"Garland Pope" <gpope@kcc.com> writes:

> I have been authoring an intranet application for IE3 and our company is
> switching to IE4 soon

This, BTW, exemplifies why it might be wise to author for an intranet
as if you were authoring for the WWW.

> was quite surprised at the difference in the appearance of the pages in
> IE4.

Is there any reason to expect that pages would look similar on different
versions of IE? After all, they don't look similar even on the _same_
version of IE (although in an intranet, you might try to force all people
to have the same browser settings, no user style sheets, identical
monitors with identical resolution settings etc etc).


Jukka Korpela

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

(Miguel Cruz) wrote:
>Greg Berigan  <gberigan@cse.unl.edu> wrote:

>>s.ki        kabu.ki      kha.ki     lo.ki     low.ki        sukiya.ki
>>tur.ki      taji.ki      ti.ki      whis.ki   disc-jock.ki  pass.ki
>>hokeypo.ki  turn-the.ki  eat.ki/wi  internet.ki/osk   french.ki/ssing

>Like I said, I feel sorry for a country whose best hope for badly needed
>infrastructure (like a 10-foot-high sea wall around the entire country) is
>the proceeds from the sale of the domain taji.ki.

Depends if anyeone wants a domain named after the Iranian language of
the Tajik people.

I was just using wildcard matches on an on-line English dictionary
though.  Still, "s.ki", "whis.ki", an "internet.ki" might be
valuable.  Some sex site may want french.ki.


Greg Berigan 

--------------------

Suzie-Q wrote:
> Greg Berigan wrote:
> > Some sex site may want french.ki.
>
> And let's not forget kin.ki.

You people are spoo.ki.


Miguel Cruz

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

<drclue@drclue.net> wrote:
> I often get ragged for not validating to some particular HTML DTD, but I
> live in the real world and not spec land.

You might benefit from reading the Web Design Group's "Using a Custom DTD"
<http://www.htmlhelp.org/tools/validator/customdtd.html>, which explains
how you can add proprietary markup (e.g., <INPUT WIDTH=...>) to a public
DTD and then validate documents that use proprietary markup.
-- 
Darin McGrew

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

I've got Netscape 1.22 (and even 1.1, I believe); I'll put them on my
webspace (http://pobox.leidenuniv.nl/~piest/netscape/) for those
interested.

Homme A. Piest

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

On Thu, 10 Sep 1998 16:06:37 GMT, martian@net1fx.com (Jason Berkan)
wrote:

>On Wed, 09 Sep 1998 23:35:09 GMT, catnip8@geocities.com (catnip)
>wrote:

>>You're right, but I'm curious as to why someone would want to launch a
>>.bat file from a browser...
>
>It's usually done because the person (or their boss) is trying to use
>the browser as the OS to shield the user from the horrors of
>attempting to understand an OS when almost everyone knows how to click
>on a link to go somewhere on the web.

<grumbles> Luddites...</grumbles>

catnip

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Nick Wagg wrote:
> 
> IMO background music is not welcome on a web site, any more than
> while I am being kept on hold on the telephone.

Off the top of my head, I can't think of a single more obnoxious mistake
than putting background music on a page.


Greg Miller

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

CIWAH should stand for Come In We Always Hassle.

Dave Lowe

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Alan J. Flavell" wrote:

>Well, I've sometimes been tempted to slip in rude[*] comments about the
>users of the Tweedles, that only non-Tweedles users will see.  Except
>that it doesn't quite work, for the reason I just gave. 

Sure it does!

<TABLE>
<TR><TD>Netscape users won't read this, because I don't close the
TABLE. Tee-hee.</TD></TR>
</BODY>

<OBJECT CLSID=#>
</OBJECT>
IE users won't see this, because their browser just crashed.

:-)


Tor Iver Wilhelmsen

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

> I don't see what's to stop a browser being able to automatically provide
> the multi-columnar rendering you describe, given a document of any
> length. 

The same thought occurred to me on reading Greg's interesting piece.
Of course an author should be able to _disable_ multi-columnar rendering.
<p class=tinc>
Perhaps an HTML element called something like <multicol> could be used
to delegate responsibility to a browser, for extended text passages
calling for flexible rendering ;^)
</p>

-- 
Nick Kew

----------------------

Nick Kew wrote:

> <p class=tinc>

aka satire?

> Perhaps an HTML element called something like <multicol> could be used
> to delegate responsibility to a browser, for extended text passages
> calling for flexible rendering ;^)

<porcine class=aviation>
Maybe <HTML LEVEL="2.0"> would do the trick?  

With a linked style sheet, natch...
</>


Alan J. Flavell

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%


On Fri, 11 Sep 1998 17:02:48 GMT, catnip8@geocities.com (catnip) diligently
typed:

Hi catnip,

I have a comment about your CatnipML(tm) as it relates to HTML:

]<grumbles> Luddites...</grumbles>

It seems to me that CatnipML differs considerably from HTML, in that cML
(if I may be so bold) uses a "tags as commands" approach, where elements
appear as verbs.  HTML, however, is the opposite.  Its elements declare
structures; therefore they appear as nouns.  So, a cml2html converter might
translate your line above to: <P CLASS="grumble">...</P> or possibly <Q
CLASS="grumble">...</Q>.

Thank you for tuning into another episode of "Hey, It's Friday!"
I've been your host,
Chris Burch  ;)
PS  Look for my forthcoming "CatnipML for Dummies" at major bookstores near
you!

-----------------------------------

>It seems to me that CatnipML differs considerably from HTML, in that cML
>(if I may be so bold) uses a "tags as commands" approach, where elements
>appear as verbs.

Actually, catnipML uses nouns and verbs. The next revision will
include the capability to use adjectives and adverbs as well. ;-)

>  HTML, however, is the opposite.  Its elements declare
>structures; therefore they appear as nouns.  So, a cml2html converter might
>translate your line above to: <P CLASS="grumble">...</P> or possibly <Q
>CLASS="grumble">...</Q>.

I'm a commanding kind of woman. What can I say? :-)

>Thank you for tuning into another episode of "Hey, It's Friday!"
>I've been your host,
>Chris  ;)
>PS  Look for my forthcoming "CatnipML for Dummies" at major bookstores near
>you!

That's a bit redundant, isn't it?

<goddess>catnip</goddess>

------------------------------

On Tue, 15 Sep 1998, catnip wrote:
<snip>

</snip>
> 
> <goddess>catnip</goddess>

<giggle><chuckle><snort><guffaw>Goddess?</guffaw></snort></chuckle></giggle>

Thanks, I needed a good chuckle this am..only 10 AM and it's been a LONG day..
:)


Jody F. Kerr

--------------------------------


]Actually, catnipML uses nouns and verbs. The next revision will
]include the capability to use adjectives and adverbs as well. ;-)

All right!  My book sales will double!  Now if someone could just write a
browser that completely mangles CatnipML ("Browser X CatnipML for
Dummies"), they would quadruple!  Plus with Catnip Style Sheets and
catnipScript, I'll make a fortune!!  Thank you, catnip, thank you!

Bermuda, here I come!
Chris Burch

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Curious George wrote:
> I'd like to hear opinions, what is the consensus on using DHTML exclusively.

Strange that you should be posting this question in the Unofficial SAVE THE LYNX
NG.


Jeff Thies

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Robert G. Eldridge wrote:
> catnip wrote:
> > Dave Lowe wrote:
> > > catnip wrote:
> > >
> > > > Hillary learns HTML:
> > > > 
> > > > Why doesn't this validate??
> > > > 
> > > > <bill_clinton> I did <em>not</em> have sexual relations with <b>that
> > > > woman,</b> Miss Lewinsky.</bill_clinton>
> > >
> > > You forgot the <lie_like_hell> attribute... ;)
> >
> > You're right. I ran it through the Starrish Inquisition Validator(tm)
> > and recieved that error message. :-P
>
> I think catnip that you need to send the Validator authors an email as
> the initial error in the markup was in the invalid element name.

I'd do that if I they'd grant me immunity from testifying. ;-)
(Everybody and their dog has been subpoenaed.)

> Names are discussed in rfc1866 at 3.2.3:
> 3.2.3. Names
<snip>
> This seems to make it fairly clear that the problem is the underscore
> in the element name.

It may be tough to convince the author of that fact, since
underscoring is his forté.

> Might I suggest <P> with a little CSS might be more appropriate, like:
> <P class=bill> ..... </P>
> Your stylesheet would then define the class .bill using appropriate
> properties.
> In this case the key words to look for in the specs might be:
> pseudo, transparent, slanted, ex, small and none.

That should work. I'd add a few more key words though: private,
nobody's business, family matter...

> This solution has the added benefit of a more widespread use, given
> the class name  ;-)

Yes, I can see how various other properties could be used: Schlake,
Gates, etc...

Perhaps the next spec, HTML 4.let's_get_real, will solve this problem.
;-)

catnip

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

NetsPlorer x.0

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

catnip (catnip8@geocities.com) wrote on MDCCCXXXVIII September MCMXCIII
in <URL: news:35fadc1d.168949245@news.nucleus.com>:
++ On Sat, 12 Sep 1998 05:10:28 GMT, aray@nmds.com (Arjun Ray) wrote:
++ 
++ >On Fri, 11 Sep 1998 08:19:33 -0700, ".." <drclue@drclue.net> wrote:
++ 
++ >| As a software engineer , I consider a spec to be like a buildings
++ >| foundation.
++ >
++ >Interesting analogy. I'd consider a spec more akin to an architect's
++ >blueprint (and the technological facts, such as strength of materials
++ >and load distribution parameters etc., that would make the blueprint
++ >*sensible*.) 
++ 
++ Architects actually have specification manuals (and addendums) that go
++ out with each project. That was the analogy I was also thinking about
++ in reference to HTML specs.


I don't see that analogy at all. Once you have an architect's blueprint,
it's pretty much decided how the building looks like. You might be able
to add a few internal walls, but you can't add an extra floor, or move
the elevators.

The HTML specifications are more what the laws of physics are for the
architect. If he ignores them, the resulting building will be unstable
and collapse at an unpredicted time.



Abigail

----------------------------

On Tue, 15 Sep 1998 04:49:23 GMT, aray@nmds.com (Arjun Ray) wrote:

>On Tue, 15 Sep 1998 04:28:41 GMT, catnip8@geocities.com (catnip)
>wrote:
>
>| They obviously wouldn't substitute sand for concrete.
>
>Why not? 
>
>     Download rose-tinted glasses ***NOW***!!1!
>
>Smoke and mirrors optional.

There must be a JavaScript script for that! ;-)

catnip

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

ingmar.greil@gmx.net (Ingmar Greil) wrote:
| Thus spoke aray@nmds.com (Arjun Ray) :
 
| >But why P instead of PARA? And yes, things like UL, OL, LI, DL,
| >DT, DD, TR, TH, TD, and so on are definitely not examples of user
| >friendly nomenclature.
| 
| might be; they do, however, support writing code by hand.

True, but what good is the ability to write by hand when the meanings
aren't known? Cart before horse, I think. The natural reaction of a
non-tech user to a <UL> isn't likely to be "Oh, that's easy to write"
as much as "WTF is that?!"

The source of HTML's initialisms is an example DTD in Annex E of ISO
8879 (see p.530 in the SGML Handbook.) Unfortunately, the example
isn't simple; however, it's functional, *provided* the users already
know what the initialisms stand for. I suspect the example was a
reworking of some DTD used by technical writers in an existing
documentation system. 

The implicit lesson is that names in a DTD, especially "short forms"
such as abbreviations or initialisms, need to be in tune with the
vocabulary of the people using them (so, e.g., it's not unreasonable
to expect a technical writer to know what an Unordered List is, and
thus find something like UL readily grasped and quite convenient.) The
utility is much less to a broader user community of people including
those who don't have document preparation as a primary activity: they
now have a learning curve to hurdle. By dint of experience, many
real-life DTDs have tended to use longer and thus more immediately
meaningful (or more easily remembered) names. 

| I wouldn't want to write UnorderdList, ListItem, TableHeader,
| TableRow, TableData, WhatHaveYou, by hand every time.

IMHO, that's overstating the case somewhat. LinuxDoc (and its "parent"
QWERTY) has names like List and Item; an ordered list is Enum (again,
a borderline case); for tables, would names like Header, Row, and Cell
really be that burdensome?

On another tack, readily meaningful names have a definite pedagogical
value: the user community might just have learned about structural
markup all the faster had they been given tags like <LIST>, <ITEM>,
<HEADING> and so on. Instead, there are probably a good many out there
to whom <UL> is no more (or no better) than some mystical incantation
meaning "indent".  
 
| >| If we want to make it the exclusive domain of the technical elite,
| >
| >The requirement seems to be computer literacy.
| 
| ACK.

Quite.


Arjun Ray

-------------------------------------

On 12 Sep 1998 18:30:26 GMT, Greg Berigan <gberigan@cse.unl.edu>
wrote:

>No, very good design.  The tags used frequently (P, A, B, I, EM, TR) and
>repeatedly (LI, TH, TD) have short names to conserve both bandwidth and
>fatigue.

Isn't it interesting that "B" and "I" are among the frequently used
tags? _Adequate_ use of them is very limited. Obviously, in this area
HTML was somewhat distorted from the very beginning. Naming the
logical elements "STRONG" and "EM" conveys--along the line you
mention--the implicit message that they are to be used rarely as
compared with "B" and "I".

> It is better to get people asking what UL is for and getting
>an answer than for them to start using UnorderedList without knowing how
>to use it or that it contains ListItems.

Yes, we have to live what we have got. There's not much point in
changing element names, even if they were originally selected badly.
 And it is noteworthy that the element name BLOCKQUOTE hasn't
prevented people from reading it as 'indent'.


Jukka Korpela

---------------------------------------

Greg Berigan <gberigan@cse.unl.edu> wrote:
| aray@nmds.com (Arjun Ray) wrote:
| >ingmar.greil@gmx.net (Ingmar Greil) wrote:
| >>aray@nmds.com (Arjun Ray) wrote:
| 
| >>> But why P instead of PARA? And yes, things like UL, OL, LI, DL,
| >>> DT, DD, TR, TH, TD, and so on are definitely not examples of user
| >>> friendly nomenclature.
| 
| >> might be; they do, however, support writing code by hand.
| 
| > True, but what good is the ability to write by hand when the meanings
| > aren't known? Cart before horse, I think. The natural reaction of a
| > non-tech user to a <UL> isn't likely to be "Oh, that's easy to write"
| > as much as "WTF is that?!"
| 
| And once the simple explanation is given,

By whom? How many people out there do you suppose actually went
looking for explanations as opposed to simply "learning by doing"? 

The basic problem starts very early, because the first impulse is
usually a question of the form "what does this do?", while the correct
explanation -- which may never be -ahem- "given" -- would answer the
question "what is this?" With some variant of the Mosaic family as the
"teacher" it doesn't take long to "learn" that <UL> apeears to achieve
an indent, at which point the neophyte need look no further for -ahem-
explanations.

This isn't just likely for non-techs. Four years ago, we hired a kid
who had been puttering with computer hardware for soem time already.
He got up to speed with basic Unix sysadmin tasks in a very short
time: by no stretch could he be called computer illiterate. Well, he
announced his first shot at a home page on the internal staff mailing
list; this is how his masterpiece started:

  <P><P><P><P>
  You have reached...
  <P><P><P>
  <UL><B><H3>My Home Page!!</UL></H3></B>

He "explained" to us that he had "worked out"(!) that P "makes a blank
line", UL "does an indent", B "turns bold on", and H3 "sets a medium
large size". Easy stuff, this HTML, piece of cake!

Go further: among all those with an introduction to HTML similar to
that kid's, how many could have turn around and written pages about
their "knowledge" for the benefit of others?! The gibberish out there
masquerading as "HTML Tutorials" has *definite* origins, if you care
to think about it. An integral part of the puzzle is the fact that
cryptic names are taken for *exactly* what they are: computerese that
you don't *need* to know about, just figure what it does already.

Now, you tell me, Greg: was this kid's "experience" unusual? Would you
be surprised? Why?

| Actually learning the tags, what they are for, and how they are used
| are good things.

That's the point.Wwhat do you suppose promotes pedagogy: curiosa like
<UL>, or readily apprehensible words like <LIST>? 

| Consider how often each of TR and TD are used in individual pages
| today and ask yourself if you really want to download all those extra
| bytes

My personal druthers? I wouldn't mind at all. The byte differential
argument has no real basis (it has also been put forward as a
justification for empty end tags.) The savings are marginal, and in
any case if economy were the goal the answer is compression. 

| just so that you are decreasing the intellectual hurdles to publish
| online.  (They're quite low enough already.)

I beg your pardon? The situation is one where even *smart* people are
getting it wrong! There are plenty of people out there to whom finding
out what UL "really means" is just a waste of time. Are you going to
accuse them of insufficient intellectual curiosity?

| Literacy is a good thing and I am in favor of things which encourage
| literacy.  I don't think we need to dumb down everything for the
| Cro-magnon intelligences.
| 
| There is a saying:  Make something so simple that even idiots can use
| it and only idiots will (want to) use it.

Then you must be quite satisfied with the state of the Web today.


Arjun Ray

----------------------------------

ingmar.greil@gmx.net (Ingmar Greil) wrote:
| Thus spoke aray@nmds.com (Arjun Ray) :
 
| as with any new language, there is a certain learning curve. i for
| once prefer writing TD and have some HTML beginner have to look it up
| -- once, than the other way round.

The point, I think, is the circumstances under which a beginner would
actually look it up *and* find an accurate answer, as opposed to just
copying what seems to "work" for whatever reason. Noone is denying the
convenience of TD when the meaning is known. The issue is whether TD
would encourage a beginner to find out all about it, as opposed to
accepting the first semi-plausible explanation that comes along. It's
much harder to concoct (or accept) bogus explanations for words with
plain meanings. 
 
| >QWERTY) has names like List and Item; an ordered list is Enum (again,
| >a borderline case); for tables, would names like Header, Row, and Cell
| >really be that burdensome?
|
| would that really help non-tech users that much?

I think it would. When it comes to *learning*, an entire translation
layer (from geeky non-words to plain English) will have been avoided.

(This is all hypothetical anyway. The HTML tagset isn't going to
change in this respect. Nor will the quality and quantity of bogus
explanations, even from apparently "reputable" sources.)


Arjun Ray

----------------------------------

> | >But why P instead of PARA? And yes, things like UL, OL, LI, DL,
> | >DT, DD, TR, TH, TD, and so on are definitely not examples of user
> | >friendly nomenclature.


Contrary to popular belief, not every HTML author can speak English.
Therefore "PARA" does not necessarily have any more meaning that "P".

Similarly, "TableHeader" is not especially more useful than "TH"
to someone who does not speak English, and is probably less useful.



.. Roger Carbol

--------------------------------

Greg Berigan (gberigan@cse.unl.edu) wrote on MDCCCXXXVIII September
MCMXCIII in <URL: news:6te88o$to$1@unlnews.unl.edu>:
++ aray@nmds.com (Arjun Ray) wrote:
++ 
++ > IMHO, that's overstating the case somewhat. LinuxDoc (and its "parent"
++ > QWERTY) has names like List and Item; an ordered list is Enum (again,
++ > a borderline case); for tables, would names like Header, Row, and Cell
++ > really be that burdensome?
++ 
++ Consider how often each of TR and TD are used in individual pages today
++ and ask yourself if you really want to download all those extra bytes
++ just so that you are decreasing the intellectual hurdles to publish
++ online.  (They're quite low enough already.)

Oh, come on. Get serious. Many pages have 100k or more on pointless inline
graphics. And you complain about 1 or 2 bytes per TR or TD more?

I suppose you never put in an extra space either, to save precious bytes.



Abigail

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

On Tue, 15 Sep 1998 23:28:37 -0400, "zero" <teehanr@hotmailNOSPAM.com>
wrote:

>Ok, I'm designing this web site, and I want it to be accessible to users
>with antiquated browsers that don't support style tags and stylesheets.

I think you have the mistaken impression that stylesheets dictate.
Browsers that don't support stylesheets simply will not show whatever
stylesheet enhancements you've added.

>  Is
>there a way to globally define the default font face for the document in the
><body> tag, or anywhere else?  Basically, I'm looking for the easy way out,
>I don't want to have to change the font face for every new header or
>whatever.

Of course, the easiest way is just to leave it alone. Most viewers use
their own font preferences. If you don't want to change the font face
'for every new header or whatever', then you're out of luck with html.
The idea of using stylesheets is to simplify the presentation of the
page, which html does not offer.

catnip

--------------------------------------

On Tue, 15 Sep 1998, zero wrote:

> Ok, I'm designing this web site, and I want it to be accessible to users
> with antiquated browsers that don't support style tags and stylesheets.

That's how it was designed.  All that you have to do is to work with the
design rather than against it.  That means, broadly speaking, putting
the mandatory structure into the HTML, and the optional presentational
enhancements into the CSS.  Or, to put it another way, you make it so
that the HTML part will successfully carry the content of your message
even when it isn't feasible to honour the stylesheet suggestions (e.g
the indexing robot).

>  Is
> there a way to globally define the default font face for the document in the
> <body> tag, 

In what way would that be a mandatory part of your message?  Do you
suppose that the indexing robot would fail to understand your message
because it isn't seeing it in the correct font?  Then you would have a
serious problem, considering that the indexing robots are the way that
quality visitors (as opposed to random surfers) are finding web sites
nowadays.

And the same goes for text mode browsers, of course (think palmtops,
cellphones...).

>  Basically, I'm looking for the easy way out,

The easy way out of what?  A problem that wasn't there until you created
it, I'd suggest.  If users aren't happy with the font face and size
that their browser uses by default, then they can either learn how to
change it themselves, or they can learn to live with it, or wait until
they get a CSS-capable browser.

The easy way out is to work with the design: content in HTML, optional
presentation in the stylesheet.  Using legacy presentational markup is
sort-of feasible, if you're more worried about the appearance on legacy
browser versions than on your content (there's only 24 hours in the day,
and when you're working on one you aren't working on the other,
presumably). But the popular browsers haven't implemented that too well,
so you'll be giving yourself extra work to avoid the presentational
markup interfering with the stylesheet. 

In the end, of course, it's your choice.  One can only point to the
issues.


Alan J. Flavell

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Bob Mariotti <bobm@cunix.com> wrote:
> Example - we need an image 770 pixels wide.  How many inches would
> that be?

On my current display, 770 pixels could be anywhere from 7 to 19 inches,
depending on the selected resolution.  The projection display used for
meetings would use a few feet (at least) to display a 770-pixel image, and
770 pixels is a bit more than an inch on the laser printer down the hall.
(To be fair, most browsers print images scaled to about 1 point/pixel.)

Ultimately, you don't know how big 770 pixels are.  Welcome to the WWW.
-- 
Darin McGrew

-----------------------------

The size of a pixel is one pixel by one pixel.  It may be rendered as
either square or rectangular, depending on the aspect ratio of the
user's monitor.  The physical size of a pixel is determined by either 
the dot pitch of the user's monitor, or by the interaction between the
monitor's physical display area and the number of pixels (width and
height) selected by the user, depending on whom you ask.

In other words, pixels do not translate to inches, nor to meters.
Don't even try it.
-- 
Diane Wilson 

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

WYSI(sorta)WYG

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

My .signature said:
>> What's another word for "synonym"?

Jenna Farnham  <webmaster@webmoose.com> wrote:
> I can't take it any longer. <g>  Synonyms for "synonym" include
> "equivalent," "analogue," "counterpart," "parallel" "correspondent,"
> "similitude," and "correlate."

Darn.  Now I have to go and change my .signature again.
-- 
Darin McGrew, mcgrew@alumni.stanford.org, http://www.rahul.net/mcgrew/

"The man who never makes mistakes never makes much of anything."
                                                     -- Waite Phillips

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

> This suggests that their authors are either simply unaware of the
> standards, or don't care; I suspect the former is more likely.

Well, I would say there's another reason, one that I've heard often. 
Having noticed that many of the sites that conform to the standards look
rather plain, and many of the sites that are full of graphic content
don't validate, there's lots of people who have convinced themselves
that it must be impossible to produce any kind of interest in a page
without breaking validation. I've lost count of the number of people who
have told me, for example, outright that it's impossible to use
javascript in a validated HTML page, and they clearly believed this
assertion, in spite of it being untrue.


Alan J. Flavell

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Abigail wrote:

> ++ Off the top of my head, I can't think of a single more obnoxious mistake
> ++ than putting background music on a page.
> 
> Oh, I can.

Now HERE'S a thread that everyone can contribute to.


Guidelines:  These should be obnoxious mistakes actually spotted in
the wild, not merely "hypothetical" monstrosities.  URLs to live
examples encouraged!


Some of my favourites:

1)  <BLINK>.  Say no more.
2)  Marquee scrolling.  Uhg.
3)  ALT="IMAGE"




.. Roger Carbol

--------------

Phil Stripling wrote:

>   4)  "Sorry, this page uses frames, and your browser doesn't support
>       them. Come back when you have a modern browser."


This is pretty obnoxious all by itself, true; it's even MORE
obnoxious when one's browser DOES support frames, and for some reason
the site thinks that it doesn't.



.. Roger Carbol

-----------------


like i visited this site the other day, telling me that "although i
was using the finest browser (NS) i were supposed to get a more recent
version, namely 3.0 or 4.0" 

(i use 4.5)


Ingmar Greil

--------------------

5) One of my BIGGEST peeves is the thoroughtly obnoxious Java (I think it's
Java) thingy that makes the screen scroll through umpteen colors before the
page loads.  What's the point, other than to make the visitor's blood
pressure rise??  The use of that ought to make the page owner's computer
blow up.

Leslie

----------------------

6) Annoying MIDI BG sound
7) Stupid annimated anything

The Bartman

----------------------

bartman wrote:
> Ed Valle wrote:
> > After hearing all your arguments, your boss decides to go ahead and
> > implement 1 - 7. What do you do?
>
> <yesman>He pays you and you do it and don't worry what other people
> say because ......well? Your site would just blend in with the rest
> and no one would really care as long as your boss is happy.<yesman/>

As long as there's nothing to prove you created the page :)


Greg Miller

-----------------

++ After hearing all your arguments, your boss decides to go ahead and
++ implement 1 - 7. What do you do?


Tell him he's an idiot and wish him good luck?



Abigail

-----------------

>Tell him he's an idiot and wish him good luck?
 
        Is that advice or just the way you get new jobs?


Gil Harvey

------------------

++      Is that advice or just the way you get new jobs?


Oh, I might use a bit less strong language, but I would certainly
tell him the plan is foolish.

And no, my boss doesn't have the notion he knows all and disagreement
in a reason to fire someone. Quite the opposite actually.



Abigail

-----------------

>After hearing all your arguments, your boss decides to go ahead and
>implement 1 - 7. What do you do?

Tell him to kiss off, it's my goddamn web site and he can't screw with
it.  Then collect my things and go have a chat with HR...

Hey, you can realistically do only three things:

1) Try to pursuade  him the error of his ways.

2) Quit.

3) Do the work the way he wants.

Jeff

PS:  Has anyone's boss really done this?  I hear stories, but I've
never had anyone attempt anything like this with me.

Jeff Cochran

-----------------

>1) Try to pursuade  him the error of his ways.

        Hey, what a concept! 

>2) Quit.

        Yeah, quitting is smart.

>3) Do the work the way he wants.

        Only if buying groceries and paying rent is important to you.

>Jeff
>
>PS:  Has anyone's boss really done this?  I hear stories, but I've
>never had anyone attempt anything like this with me.

        Well, in the past year I've had occasion twice to tell a HTML
'author' "just do it my way" - so to answer your question, yes there
are times when the boss says  it's  my way or the highway.

Gil Harvey

--------------

> PS:  Has anyone's boss really done this?  I hear stories, but I've
> never had anyone attempt anything like this with me.

Yes. Not (yet) w.r.t. HTML, but I had to put a 'large' image on the
'frontdoor' of our site. After that I built in a backdoor, of course.

Also my job has mutated from being a 'mere' technical writer to another
job (which includes maintaining our web-site), because my boss (and
others) simply refused to make the tech writing possible. Even though I
did show him/them all kinds of recommendations available in the
literature. So I refused to do any tech writing. Period.

By now that boss and others mostly are gone. 


Auke Reitsma

----------------

>After hearing all your arguments, your boss decides to go ahead and
>implement 1 - 7. What do you do?


Forge an e-mail from the Internet Administrative Headquarters, telling
her/him to stop doing that.*

If your boss will believe that 1-7 are good things, getting her/him to
believe that such a place exists will require little to no effort.

* Many moons ago, I received a tech support call from a user who was worried
he had "broken the Internet". I told him I had received a message like the
one above, but that they would have the Net re-booted within the half hour.

8^)

Dan McGarry

-------------------

catnip wrote:
>Dan McGarry wrote:
>>If your boss will believe that 1-7 are good things, getting her/him to
>>believe that such a place exists will require little to no effort.
>
>LOL!
>
>>* Many moons ago, I received a tech support call from a user who was worried
>>he had "broken the Internet". I told him I had received a message like the
>>one above, but that they would have the Net re-booted within the half hour.
>
>You could have told him it was the annual 'web clean-up day'. :-)


A guy I know has a t-shirt that says:

"The Internet is full. Go away."

I'm thinking of marketing them. I figure they'd sell like hotcakes to
purists -- if I can ever find one....

>I like your style...


<blush>*

(* For those who have been following the CatnipML thread, <Blush> is an
empty tag which can have a color attribute, if the author so desires, i.e.
<blush color="#FF0000">. Color can also be suggested with Catnip Style
Sheets [CSS]. The latter is the recommended approach.)

8^)

Dan McGarry

----------------------

Dan McGarry wrote:

> A guy I know has a t-shirt that says:
> 
> "The Internet is full. Go away."

Does it have a footnote like:

        Occupancy by more than 4294967295 IP addresses unlawful.




.. Roger Carbol

--------------------

>I'm thinking of marketing them. I figure they'd sell like hotcakes to
>purists -- if I can ever find one....

I'd wear that. :-)

>>I like your style...
>
><blush>*
>
>(* For those who have been following the CatnipML thread, <Blush> is an
>empty tag which can have a color attribute, if the author so desires, i.e.
><blush color="#FF0000">. Color can also be suggested with Catnip Style
>Sheets [CSS]. The latter is the recommended approach.)

I'm impressed! You've obviously read my specifi*cat*ions.

catnip

------------------

>I like your style...

        Weel , at least now we know who the caller was....

Gil Harvey

-------------------

> After hearing all your arguments, your boss decides to go ahead and
> implement 1 - 7. What do you do?
> 

(*Run away, run away, ruuun awaaay!!!!*) (imagine Monty Python's the Search
for the Holy Grail)

and, yes, I'm working on a site right now that *almost* fits the description.
Just about every piece of annoyance except the midi background...I told them
our server didn't support it...lol 


Jody F. Kerr

-----------------

> After hearing all your arguments, your boss decides to go ahead and
> implement 1 - 7. What do you do?

I suppose it depends on how desparate you are, and whether anyone would
possibly, conceivably, ever find out that you had anything to do with
the page/site.

It goes without saying that you do not list this one on your resume.


Diane Wilson

-------------------

Greg Miller wrote:
>Leslie wrote:
>> 5) One of my BIGGEST peeves is the thoroughtly obnoxious Java (I think it's
>> Java) thingy that makes the screen scroll through umpteen colors before the
>> page loads.  What's the point, other than to make the visitor's blood
>> pressure rise??  The use of that ought to make the page owner's computer
>> blow up.
>
>Nah, that's a generic JavaScript effect, not Java.


Well, what ever it is is annoying as hell!!

OK, another pet peeve are those annoying pop up boxes that ask, "Hey
dude/dudette, what' yer name??"   GRRRRRRRRRR!!!

Leslie

---------------

>OK, another pet peeve are those annoying pop up boxes that ask, "Hey
>dude/dudette, what' yer name??"   GRRRRRRRRRR!!!

I never found those to be annoying until I went to a site that popped
out 6 questions in a row, that you had to respond to before the site
ever loaded.

Don't think I've ever gone back to that site.

Jason Berkan

----------------

*and* marquee, status bar scrolling, messages that suggest you change
your display settings or browser. I'm on 28.8 and a grey Java box sends
me to the *stop* or *back* button ...Actually the worst thing I ever saw
was the screen color changing thing done with an animated gif as the
background. Don't believe me? It was at
http://www.geocities.com/ResearchTriangle/Thinktank/6446 

It's NOT me but it has to be seen to be believed :-]


Gerry Coughlan

----------------

>  4)  "Sorry, this page uses frames, and your browser doesn't support
>      them. Come back when you have a modern browser."


If you try to browse the document at http://www.waff.com/site/ in a
browser with no frame support or with frames turned off, you get a
blank page.  Not even a rude "You need a better browser!" message.

Browsing the site with frames turned on exposes you to the opposite
extreme: documents so busy with images and effects that it's almost
too cluttered to read.  Then again, according to our resident troll,
that's what sets the site apart from the "bland and boring."


Chuck Taylor

-------------------

> 1)  <BLINK>.  Say no more.
> 2)  Marquee scrolling.  Uhg.

Hmmm. In the right situation, and used sparingly, both <blink> and
<marquee> can be useful. And they both degrade gracefully on browsers
that don't support them, so it doesn't cause compatibility problems. I
do agree, however, that they are often seriously misused, and in any
case the effects they produce should really be done by CSS.

> 3)  ALT="IMAGE"

Now that is annoying.

My "favourite" mistakes are:

1) Background sound.
2) ALT="something.gif (1246k)"  
3) "This site is best viewed at 800x600"

Mark Goodge

----------------

> > 1)  <BLINK>.  Say no more.
> > 2)  Marquee scrolling.  Uhg.
> 
> Hmmm. In the right situation, and used sparingly, both <blink> and
> <marquee> can be useful. 
>..

Yes, a blink is still better then an animated image of the text, and a
marquee is still better then a java news ticker, not that this would
make them really good, but...


Philipp Lenssen

---------------

> > 1)  <BLINK>.  Say no more.
> > 2)  Marquee scrolling.  Uhg.
> 
> Hmmm. In the right situation, and used sparingly, both <blink> and
> <marquee> can be useful. And they both degrade gracefully on browsers

_Very_ sparingly ...
 

> 1) Background sound.
> 2) ALT="something.gif (1246k)"  
> 3) "This site is best viewed at 800x600"

When I see that last one, I occasionally mail an advice to replace that
by:
    "This site is best viewed with your eyes closed.
    And hopefully you are deaf ..."


Auke Reitsma

----------------

I recall a site whose sole means of navigating was a collection of
Java-applet buttons.  Anyone accessing the site without a Java-capable
browser had no way of following links to any other document on the
site, or even knowing the links were there (without examining the
document source directly, that is).

The "Best Viewed With Internet Explorer 4.0" slogan was featured
prominently on the site, of course.

Sorry, no URL to provide.  The author has made some wiser decisions
since then.


Chuck Taylor

-----------------

> Now HERE'S a thread that everyone can contribute to.

Amen! :-)  No specific URLs -- these sites tend to be booted out
of my bookmark list pretty quickly! -- but here are some things
I've "experienced" while surfing...

1) Non-resizable, non-scrolling frames smaller than their content.

2) Items necessary for navigation enclosed in #1, above.

3) Background images of 700K or more in size; the corollary
    is background music of similar size.

4) Identical or close colors for text and background.

5) Making all navigation and/or use and/or content of the
    site plug-in dependent.

6) Lots more, but my blood pressure's going up... <g>


Privatepages Webmaster

----------------

>1) Non-resizable, non-scrolling frames smaller than their content.


I've found that the IntelliMouse's scroll button overrides the no-scroll
specification... I wonder if this is a bug... hard to believe it could be an
actual on-purpose feature. ;)


Erika Stokes

------------------


A penfriend of mine had a homepage made for IE3 (this was some time ago
and the page has since disappeared, so no URL). He wrote it all by hand
in some text editor, but I don't think I have ever seen a "WYSIWYG"
editor create anything as bad as that! It was THE worst page I have ever
seen (or rather - NOT seen, since big parts of it were totally
inaccessible in any of the browsers I tried...).

It contained, among lots of other things:


* A CV where everything was spaced out with &nbsp; everywhere (not a
table in sight on the entire site, even in cases where tables would have
made sense).

* URLs containing spaces.

* Pages with frames that told my NS3 that it couldn't handle frames.

* A bold statement saying "I'm a professional web designer" (I did
finally manage to get him to at least remove the word "professional"
from that). This was in the form of a link leading to a page where he
bragged about his web design skills, such as "some links I have made"
(i.e. that he was capable of creating <a
href=not-really-a-correct-url>some stuff</a> and other similarly trivial
things).

* Images of text everywhere (and not an alt attribute in sight).

* All files in one huge directory, leading to some very strange file
names. Perhaps not a big problem for a visitor to a site, but a big
problem in the long run for its owner.

* A guestbook (or something similar) where the html code for the
"submit" and "clear" buttons were written inside the "message" window.

* A HUGE gif animation right at the top (no height and width specified)
that looked like a small gif animation followed by some lines of
whitespace (i.e. most of the animation looked just the same as the
background colour).

* His name in animated letters (a little bit further down) that didn't
work because he had not specified the files as "filename.gif" but only
as "filename" ("I don't know how you can say that is wrong, it works
just fine in IE...").

* Tons of links to pages that he hadn't made yet.

* Pages with a design with a "fixed background" where one was to get
blocks of text into view by scrolling so that they came across a white
part ("reading area") of the, otherwise black (same as the text colour),
background (of course, most of those blocks of text were bigger than the
white "reading area").

* <a href=http://whatever/whatever/whatever.htm><li></a>


etc etc etc.  I don't think I remember anything _good_ about that site
(I looked at it quite thoroughly because he asked me to, but he didn't
want to hear the things I had to say about it when I did...).


Veronica Karlsson

------------------

>* His name in animated letters (a little bit further down) that didn't
>work because he had not specified the files as "filename.gif" but only
>as "filename" ("I don't know how you can say that is wrong, it works
>just fine in IE...").

I am amazed this isn't more common among Win95/98 users: by default,
Explorer hides filename extensions for registered types (probably so
they can make it look a little more like Mac's Finder). Unless the
user already knows Windows, they can mistakenly assume the extension
is not there.

But I guess that's "user friendly". :-)

Tor Iver Wilhelmsen

------------------

>* A HUGE gif animation right at the top (no height and width specified)
>that looked like a small gif animation followed by some lines of
>whitespace (i.e. most of the animation looked just the same as the
>background colour).

This reminds me of another example of this idiocy. Kesmai, the company
that produces the online WWII air-combat simulation Air Warrior, in
the first incarnation of their web page, had a 2 _megabyte_ animated
GIF image at the top of their page. I never could force myself to hang
around on that page long enough for the GIF to load completely to find
out what they actually had in it that would justify such a
Brobdignagian (sp?) banner image. They eventually got a clue and fixed
it, though...


Sean R. Malloy

------------------

Tom Bach wrote: 
> I don't think anyone mentioned the use of red or dark blue text
> on a black background, or yellow text on a white background.

This used to really irritate me.  Then I figured out that "selecting"
the text by click-dragging it resulted in it being highlighted in
nice, easy-to-read colours.


Now it only irritates me on principle.

ObObnoxMistake:  RaNDoM caPITals


Roger Carbol

-----------------

> I don't think anyone mentioned the use of red or dark blue text
> on a black background, or yellow text on a white background.
> 

Made worse by having to wait for an oversized background graphic in order 
to see the text.  Made even worse by rarely being worth the effort to 
read.

BigDoggy

-----------------

>Made worse by having to wait for an oversized background graphic in order
>to see the text.  Made even worse by rarely being worth the effort to
>read.

Even worse if you have a big black background file that takes 5 minutes to
load, and white text, but they don't specify a background colour so you're
forced to wait for it.

zero

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%


Roger Carbol wrote:
> 
> Guidelines:  These should be obnoxious mistakes actually spotted in
> the wild, not merely "hypothetical" monstrosities.  URLs to live
> examples encouraged!
> 

A penfriend of mine had a homepage made for IE3 (this was some time ago and the page has since disappeared, so no URL). He wrote it all by hand in some text editor, but I don't think I have ever seen a "WYSIWYG" editor create anything as bad as that! It was THE worst page I have ever seen (or rather - NOT seen, since big parts of it were totally inaccessible in any of the browsers I tried...).

It contained, among lots of other things:


* A CV where everything was spaced out with &nbsp; everywhere (not a table in sight on the entire site, even in cases where tables would have made sense).

* URLs containing spaces.

* Pages with frames that told my NS3 that it couldn't handle frames.

* Pages with a design with a "fixed background" where one was to get blocks of text into view by scrolling so that they came across a white part ("reading area") of the, otherwise black (same as the text colour), background (of course, most of those blocks of text were bigger than the white "reading area").

* <a href=http://whatever/whatever/whatever.htm><li></a>

* A bold statement saying "I'm a professional web designer" (I did finally manage to get him to at least remove the word "professional" from that). This was in the form of a link leading to a page where he bragged about his web design skills, such as "some links I have made" (i.e. that he was capable of creating <a href=not-really-a-correct-url>some stuff</a> and other similarly trivial things).

* A guestbook (or something similar) where the html code for the "submit" and "clear" buttons were written inside the "message" window.

* A HUGE gif animation right at the top (no height and width specified) that looked like a small gif animation followed by some lines of whitespace (i.e. most of the animation looked just the same as the background colour).

* His name in animated letters (a little bit further down) that didn't work because he had not specified the gif animations as "filename.gif" but only as "filename" ("I don't know how you can say that is wrong, it works just fine in IE...").

* Lots of links to pages that he hadn't made yet.

* Images of text everywhere (and not an alt attribute in sight).

* All files in one huge directory, leading to some very strange file names. Perhaps not a big problem for a visitor to a site, but a big problem for the site owner ("professional web designer" indeed!). 

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Hee guys,

the
alt.comp.i.dont.know.how.to.ask.her.out.otherwise.so.i.ll.send.her.my.ascii.fantasy
group is somewhere else...


Marcel Duyn

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Jody F. Kerr wrote:
> 
> > <goddess>catnip</goddess>
> <giggle><chuckle><snort><guffaw>Goddess?</guffaw></snort></chuckle></giggle>

That may be a nesting error... have you tried validating it?  I'm pretty
sure a <guffaw> isn't allowed to nest within a <giggle>, though the
other way around might be allowed if you're not using the Strict DTD.


Daniel R. Tobias

----------------------

>That may be a nesting error... have you tried validating it?  I'm pretty
>sure a <guffaw> isn't allowed to nest within a <giggle>, though the
>other way around might be allowed if you're not using the Strict DTD.

catnipML *only* uses a loose DTD. ;-)

catnip

----------------------

> That may be a nesting error... have you tried validating it?  I'm pretty
> sure a <guffaw> isn't allowed to nest within a <giggle>, though the
> other way around might be allowed if you're not using the Strict DTD.
> 

checked it in my new manual...those are all perfectly correct, and nested
properly..:)

Jody F. Kerr

----------------------

>]>I have a comment about your CatnipML(tm) as it relates to HTML:
>
>]Actually, catnipML uses nouns and verbs. The next revision will
>]include the capability to use adjectives and adverbs as well. ;-)
>
>All right!  My book sales will double!  Now if someone could just write a
>browser that completely mangles CatnipML ("Browser X CatnipML for
>Dummies"), they would quadruple! 

Try Dogscape of Canine Explorer (CE). Lynx is the *only* browser so
far that is catnip friendly. :-)

> Plus with Catnip Style Sheets and
>catnipScript, I'll make a fortune!!

You misspelled 'we'.

>Bermuda, here I come!

Not without me!! (Are you cute?)

catnip

------------------------

]Not without me!!

...and my wife...

](Are you cute?)

No cuter than, say, a Muppet baby caught stealing cookies.  q:)

Well, real life is uncomfortably invading this thread, so I guess we can
end it and agree to split the earnings.

HAND, catnip (and everyone else)

Chris Burch

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

>>>25 percent of visitors who's browsers support javascript have it turned
>>>off.  Don't rely on it.
>>
>>Proof please.....
>
>87.35% of all statistics are made up on the spot.
 
        That I can accept :-)


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

This group is not here for the purpose of newbies.

comp.lang.c isn't there for newbie C programmers asking why they can't
use i = i ++;. sci.math isn't there for people with homework questions.
Nor does rec.puzzles exist to explain Monty Hall over and over again.

If you want to teach HTML to newbies, create a group for them.



Abigail

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

William Schlake wrote:
>Recommendations are what they are: recommendations. Look, the W3C
>itself says in can't control and certainly can't enforce if or not
>it's membership (which includes Netscape and Microsoft) follow the
>recommendations. So how can a puny band of outspoken purists who have
>no pull and nearly zero influence on web authors and absolutely NO
>influence with browser producers effect change?

That may well be your opinion, but you'd do well to provide it in the
greater context of the Internet and its standards. Have you noticed that
most standards are promulgated in the form of RFCs? The acronym, of course,
means Request For Comments. IMHO, that's a very polite way of saying, "This
is how we choose to do things. If you have a better idea, let us know."


Dan McGarry

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

NetsPlorer 3.x

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Dan McGarry wrote:
| William Schlake wrote:
|
| >then explain why Netscape and Microsoft have been wildly successful
| >by not following them.
| 
| Post hoc error:
| 
| 1) NS and MS did not follow standards.
| 2) They were both very successful.
| 3) Therefore, they were successful *because* they didn't adhere to
|    the standards.

Besides that, it's all too easy to talk of "success" without context.
The problem here is that there's only one "history of the world" (a
very fruitful source of post hoc fallacies:-)), so the question of
"successful -- in relation to what?" necessarily involves an element
of hypothesis or projection. Bastiat's parable of the broken window
(from his essay on "That Which Is Seen And That Which Is Not Seen")
has an important lesson in this regard

  <URL:http://laissezfaire.org/hh7135c2.html>

where, if you're not careful, you'd have to agree that the hooligan
lobbing a rock into a window must have been a great social benefactor
for all the paying work he "generated".

The "success", as such, was Netscape's: they "proved" that there could
be a market for Tag Soup, where software, reasonably easy to use and
too stupid to do anything but respond stolidly to random mishmashes of
tags, could be popular. That Which Is Seen is that, yes, plenty of
people have found Tag Soup to be of immediate benefit. That Which Is
Not Seen is that Tag Soup not only has no future but also stands in
the way of the long term benefits that structured markup brings.

Microsoft is simply trying to take over what Netscape hath wrought:
their "success" has always been in swallowing the "success" of others.

But not being innovative -- not really a consideration, since all real
innovation on the Web came to an end about four years ago, but the
hype of "latest and greatest" is too hysterical to ignore -- doesn't
mean being stupid: Microsoft realized the value of the W3C as a source
of ideas (and feedback.) Hence their overt stance of cooperation: the
strategy is one of staying *between* the W3C and Netscape, who OTOH
are too stupid to realize that their pretensions to being innovative
are not just unfounded but precisely the smokescreen behind which
Microsaurus will eat Mozilla's eggs.

Call this "success" if you like. I call it farce. (Cue the OGT.)


Arjun Ray

   )
 .'     _.--.__                          _._
(    .-'       `-.      _.--.        _.-'   `-.
 `--'             `._.-'     `-.__.-'          `.
                                              .  )
                                              |`/
                                              |__\
+--+-------------------------------------------------------------------+--+--+
|==|          Netscape: LFB: Economics in One Lesson, The Applied      |()|[]|
+--+-------------------------------------------------------------------+--+--+
|File  Edit  View  Go  Bookmarks  Options  Directory  Window             Help|
|---------------------------------------------------------------------+-----+|
|[Back][Forward][Home] [Reload][Load Images][Open][Print][Find] [Stop]|T\\ Y||
|          +--------------------------------------------------------+ || \\|||
| Netsite: |http://laissezfaire.org/hh7135c2.html                   | ||.--.||
|          +--------------------------------------------------------+ +-----+|
|============================================================================|
|                                                                            |
|  Chapter II: The Broken Window                                             |
|                                                                            |
|  LET US BEGIN with the simplest illustration possible: let us,             |
|  emulating Bastiat, choose a broken pane of glass.                         |
|                                                                            |
|  A young hoodlum, say, heaves a brick through the window of a baker's      |
|  shop. The shopkeeper runs out furious, but the boy is gone. A crowd       |
|  gathers, and begins to stare with quiet satisfaction at the gaping        |
|  hole in the window and the shattered glass over the bread and pies.       |
|  After a while the crowd feels the need for philosophic reflection. And    |
|  several of its members are almost certain to remind each other or the     |
|  baker that, after all, the misfortune has its bright side. It will        |
|  make business for some glazier. As they begin to think of this they       |
|  elaborate upon it. How much does a new plate glass window cost? Two       |
|  hundred and fifty dollars? That will be quite a sum. After all, if        |
|  windows were never broken, what would happen to the glass business?       |
|  Then, of course, the thing is endless. The glazier will have $250 more    |
|  to spend with other merchants, and these in turn will have $250 more      |
|  to spend with still other merchants, and so ad infinitum. The smashed     |
|  window will go on providing money and employment in ever-widening         |
|  circles. The logical conclusion from all this would be, if the crowd      |
|  drew it, that the little hoodlum who threw the brick, far from being a    |
|  public menace, was a public benefactor.                                   |
|                                                                            |
|  Now let us take another look. The crowd is at least right in its first    |
|  conclusion. This little act of vandalism will in the first instance       |
|  mean more business for some glazier. The glazier will be no more          |
|  unhappy to learn of the incident than an undertaker to learn of a         |
|  death. But the shopkeeper will be out $250 that he was planning to        |
|  spend for a new suit. Because he has had to replace a window, he will     |
|  have to go without the suit (or some equivalent need or luxury.)          |
|  Instead of having a window and $250 he now has merely a window. Or, as    |
|  he was planning to buy the suit that very afternoon, instead of having    |
|  both a window and a suit he must be content with the window and no        |
|  suit. If we think of him as a part of the community, the community has    |
|  lost a new suit that might otherwise have come into being, and is just    |
|  that much poorer.                                                         |
|                                                                            |
|  The glazier's gain of business, in short, is merely the tailor's loss     |
|  of business. No new "employment" has been added. The people in the        |
|  crowd were thinking only of two parties to the transaction, the baker     |
|  and the glazier. They had forgotten the potential third party             |
|  involved, the tailor. They forgot him precisely because he will not       |
|  now enter the scene. They will see the new window in the next day or      |
|  two. They will never see the extra suit, precisely because it will        |
|  never be made. They see only what is immediately visible to the eye.      |
|                                                                            |
+----------------------------------------------------------------------------+
|[m/-O]                                     [_______________________] [[\/]!]|
+----------------------------------------------------------------------------+

-----------------------

> Can you tell me what enforcement measures are used to make sure Net admins
> and application authors comply with the other RFCs? I'm not aware of any.

You or I:   "I have this new product.  It works (well, sometimes) with
            its own protocols which follow my whim"
The World:  "Yes, very interesting.  Don't call us."

Microsoft:  "I have this new product.  It works (well, sometimes) with
            its own protocols which follow my whim"
The World:  "Gimme, Gimme!!!"

At first sight, Netscape looks like a counterexample to the above.  But
I contend they're not: for all NS has done to proprietarise HTML, they
do still render standard HTML (frequently not very well, but that's
another story).

MS's failure to implement HTTP is in an entirely different category,
and has done enormous damage:

You or I:   "I have this new product.  It works completely reliably with
            all the standard protocols, but shows up blank in MSIE"
The World:  "Yes, very interesting.  Don't call us."

(that actually paraphrases a true story)

Now isn't it ironic that the one and only great innovation to come from
MS was their decoupling of early DOS from the hardware it ran on?
A common - the next best thing to open - standard for the Clones gave an
OS that was on its merits a nonstarter the momentum to crush the likes
of Apple and Acorn (PC networking - in 1983 IIRC).


Nick Kew


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

IE3 and NN4 are, in my opinion, CSS-*dis*abled. There is a lot of perfectly
valid CSS1 that will render your pages totally unreadable in those browsers
(and in a few cases, possibly even crash Netscape) if they have style sheet
support enabled.

Braden N. McDaniel

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

On Wed, 23 Sep 1998 15:31:54 GMT, Jason Berkan <martian@net1fx.com> wrote:
>On Wed, 23 Sep 1998 10:11:54 GMT, steve.pugh@dial.pipex.com (Steve
>Pugh) wrote:
>
>>On Wed, 23 Sep 1998 09:27:55 GMT, cepl@fpm.cz wrote:
>>
>>>What do you think about use of closing tags which are not mandatory (like
>>></P>)? Use it or not use it?
>>
>>Use them.
>>
>>Some of them such as </TD> are not required but cause bad things to
>>happen (in Netscape for example) if omitted.
>>
>>Often stylesheets fall apart if closing tags are omitted.
>>
>>Putting in all the closing tags often makes it easier to edit pages,
>>find errors, etc. simply because you can see where things end as well
>>as where they start.
>
>I agree.  Some others have laughed at me because of the extra work I
>do putting in the optional closing tags.
>
>But to me, it makes the HTML source far easier to decipher when I can
>visually see where one tag ends and the next begins.  That way I don't
>have to remember that because I put an <H2> tag in the previous <P>
>tag was automagically closed.

I'll "laugh" if you're expending a *lot* of additional effort on this. 

It doesn't need to involve a large amount of work; if you've built
reasonably "well-formed" HTML, you can go back and fix those tags using
Dave Raggett's "tidy" program.
   <http://www.w3.org/people/Raggett/tidy/>
perhaps using the "-xml" option that (surprise, surprise!) puts in end
tags everywhere that they're needed.

Note that XML *requires* end tags; one of the things from SGML (and thus
from HTML) that it has omitted is the notion of "tag minimization." That
is the root cause for it being valid (under *any* criteria) to omit
certain HTML end tags.  By eliminating the syntax associated with "tag
minimization," XML got significantly simpler. 


Christopher Browne

--------------------------------

An interesting observation: after dismissing an early, bloated and bugridden
versions of HotJava back in '95, I recently downloaded the latest version.
It's become quite a respectable browser, and one feature is an HTML
parser that will add all the optional closing tags.

To my own surprise, I'm sufficiently impressed I'll probably fork up the
license fee to bundle it with my own software.

-- 
Nick Kew

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

martian@net1fx.com (Jason Berkan) writes:

>SNIP<
> However, what about other browsers, like speech browsers?  Do they use
> ALT?  And would a blank ALT for decoration graphics cause problems?

http://www.cast.org/bobby/
http://home.paralynx.com/sky/
http://www2.cdepot.net/~mist/access.htm

-- 
Phil Stripling 

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Here's a very insightful posting that appeared on
netscape.public.mozilla.general a few weeks ago (message ID:
<19980918024539.19164@rocky.jgk.org>):

Hi,

Just want to let you guys know that you're going to get lots of
questions on this list about Netscape products, forever and ever.

Did you ever find a door that has a handle that looks like you should
pull on it and you pull on it and you find that actually you're supposed
to push on it even though it looks like you're supposed to pull on it?
It may even say "PUSH" on it somewhere, but it doesn't matter, people
will pull on it because it looks like a pull handle.  If you do this,
you may feel stupid about it, but you shouldn't, it's just bad design.
People try to fix such things with signs like `this is not a urinal',
but that's just a confirmation that something is really not right.

--
Joe Keane, amateur mathematician

[end quoted material]

Braden N. McDaniel

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Peter Bendall   <Peter@EMBL-Hamburg.de> wrote:
> There now appears to be a tendency for certain classes of
>  users to disable Javascript - especially after publicity
>  for programs such as the one which can display the users
>  hard disk!

Yep.  I enable it for intraweb sites because they generally use it for
necessary business applications.  I disable it outside the firewall because
of the security issues and because it's usually used to annoy me by popping
up windows, hijacking my status bar, etc.

In general, it isn't hard to accomodate users without JavaScript support.
For example, instead of <A HREF="followLink()">, put a real URL in the HREF
attribute and put the JavaScript function in the ONCLICK attribute, like
this: <A HREF="http://www.rahul.net/mcgrew/" ONCLICK="followLink()">

Just make sure the JavaScript function returns the correct value to prevent
the browser from *also* following the link normally.
-- 
Darin McGrew

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

I have a few rules of thumb when it comes to page/site design:

1)  the user is always right.
2)  if the user is wrong, see rule 1.
3)  if a page doesn't display as intended in Macintosh and PC versions of
    Netscape 2.04, then it needs a re-work.
3a) if the page uses multimedia extensions, JavaScript or frames, then
    the page has to appear as desired in Netscape 3.0 on both Mac and PC
    platforms.

Just because there's a new tool to use doesn't mean that it *must* be
used.


Geoffrey Peters

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

BKrapf9811 wrote:
++ 
++ Is there a way to change the color of a bullet?


Crayons.


Abigail

-----------------

Zilbandy wrote:
> Abigail wrote:
> >
> >Crayons.
> 
> Sometimes, you are an incredible store house of useful information.
> Then, there are the other times....

No, no; she's right.  I just tried it.

Er, how d'you get wax off the screen?


Derek.Moody

-------------------

> Er, how d'you get wax off the screen?

Heat gun.


Jude Crouch

------------------

Tor Iver Wilhelmsen wrote:
> Zilbandy wrote:
> >
> >My apologies to Abigail. I didn't have any crayons, but a magic marker
> >did the trick. Now, if I could just get the dot to move when I scroll.
>
> Basic animation technique: wipe it off, move the content a pixel or
> two, then paint it. Repeat as necessary.

Boy, I'm so ashamed. I didn't think of this technique either. :(
<going off to set myself in the corner>

> (Or pester browser manufacturers to support <UL SRC> or CSS2.)


Zilbandy

--------------------

>Er, how d'you get wax off the screen?

I think you needed to write <BLINK> on the box before taking the crayon
out.


Peter Seebach

--------------------

>Crayons.
>
Excellent answer! I was going to suggest that a bullet will change colour
once it enters flesh, but anyway...


Peter (Cybiades)

--------------------

>++ Is there a way to change the color of a bullet?
>
>Crayons.

I tried that, but it leaves coloured wax around the entry wound.

This is actually covered in the alt.crazy.women.with.guns FAQ. Nail
polish seems to be the best option.


catnip

-------------------

Gil Harvey wrote:
: catnip wrote:
: >I tried that, but it leaves coloured wax around the entry wound.
: >
: >This is actually covered in the alt.crazy.women.with.guns FAQ. Nail
: >polish seems to be the best option.
:  
:       Oops, this is off topic the "moderator" will trash it :-)

So that's why I couldn't post:

Remove wax with armament-cleaning-oil.

?

[I guess I'm ''immoderate''.]


Buzz

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Since the subject of browsers other than the famous "Big Two" sometimes
comes up here (especially when they get labelled "Cheesy Brand-X
Browsers" by one vociferous participant), I thought I'd point you to a
page in my site where I discuss the various browsers other than Netscape
and IE that I've managed (thus far) to run into:

http://www.softdisk.com/comp/dan/webtips/brand-x.html

If anyone knows of any other browsers not on the list on that page, and
can point me to a site where they can be obtained, they can e-mail me. 
I've heard of a browser called "Alis Tango" that's more advanced in
internationalization, but I'm not sure where to obtain it, so if anyone
knows, please let me know.

Daniel R. Tobias

--------------------

Well, a review of the execrable Amaya should be in order...if you look hard
enough, you can still find Ariadne, Cello, WinWeb, Oracle Power Browser etc,
out there...Tucows lists Surfin' Annette and BeConn, two I haven't tried.
Believe it or not, I've tried most of them, Opera included...there must be a
reason people usually turn to the Browser Kings(tm) after all, the fact that
MSIE is bundled with an install of Windows 9x is a pretty good reason to
begin with...actually, the browser bundled with OS/2  versions 3 and 4,
WebExplorer 1.x, ain't bad...I don't think it's been ported to any other
platform. Someone correct me if it has.


Peter (Cybiades)

--------------------

Cybiades wrote:
> 
> Tucows lists Surfin' Annette and BeConn, two I haven't tried.
>

Surfin' Annette is a childrens browser.  It uses the IE engine, but
scans content for objectionable words and phrases before displaying the
page.  It is used alongside IE or Netscape, but is intended for
children.


Jm Taylor

--------------------

Daniel R. Tobias wrote:
> 
> I've heard of a browser called "Alis Tango" that's more advanced in
> internationalization, but I'm not sure where to obtain it, so if anyone
> knows, please let me know.

http://www.alis.com/internet_products/browser/browser.html

-- 
Sue Sims

--------------------

Daniel R. Tobias <dan@softdisk.com> wrote:

> If anyone knows of any other browsers not on the list on that page, and
> can point me to a site where they can be obtained, they can e-mail me.

There's an Apple-only browsing solution called "CyberDog" that embeds
browsing functionality into the Mac OS VUI; it infortunately relies on
OpenDoc, an Apple software technology that's been put on indeterminate
hold for now.

What is really scary about CyberDog is its incredible speed, stability
and overall performance - not to mention minimal system requirements -
compared to either of the Big Two.

More Mac users are turning to CyberDog as Netscape *still* cannot write
a disk caching system that doesn't cause directory damage to the
browsing machine, and whilst MS-IE is far more stable, is considered by
some to be part of the Evil Empire, but mostly because it (a) puts 114
extensions into the Mac's System Folder, (b) is a supreme example of
bloatware - 21 MB installer, just to surf the web? and (c) looks damn
ugly.

Another to consider: there are still a lot of old, pre-colour Macs out
there using the internet, running ancient programs such as Mosaic 1.00a
or ElNet's MacWeb.


Geoffrey Peters

--------------------

Geoffrey Peters) writes:

>SNIP< 
> Another to consider: there are still a lot of old, pre-colour Macs out
> there using the internet, running ancient programs such as Mosaic 1.00a
> or ElNet's MacWeb.

Whoa! I am on a Power Mac G3 with MacWeb -- it is not just those ancient
Macs who are still using older browsers. MacWeb is a very capable browser
with JavaScript and Java pre-disabled. :-> It is remarkable how much faster
pages load without all the fluff, and all I lose are the click-through Java
advertising applets relentlessly animating at the tops and bottoms of the 
pages.

Please, all you HTML authors, start marking up your files for MacWeb. (The
Windows version is WinWeb, naturally enough.) It is not just for ancient
pre-color computers, Mac or Windows.


Phil Stripling

--------------------

Phil Stripling wrote:
:Whoa! I am on a Power Mac G3 with MacWeb -- it is not just those ancient
:Macs who are still using older browsers. MacWeb is a very capable browser
:with JavaScript and Java pre-disabled. :-> It is remarkable how much faster
:pages load without all the fluff, and all I lose are the click-through Java
:advertising applets relentlessly animating at the tops and bottoms of the 
:pages.

My favorite feature of MacWeb was that it let you set exactly how you
wanted the various tags to be displayed, such as emphasis, code, address,
etc. You could set the font, color, style, etc. for them. One of the few
browsers that gave you that level of configurability. I really wish it
hadn't vanished into obscurity after being bought- it had nice handling of
ALT text, did a fairly decent job of quick loading, and was nicely
configurable.

I'd have a hard time using it today because of the lack of frames support
(I don't need frames themselves, but it'd be nice if it were updated to
the level that Lynx supports), tables support (not a huge deal, but for
some pages it can get messy), and if I remember correctly, progressive
jpegs didn't display.

If someone were to get it now and to fix it up a little to support some of
the current stuff in a fashion to at least allow it to navigate sites
using these things, it'd be a nice browser without a lot of overhead. I
don't suppose there's any chance of getting it released to public domain-
I'm sure people on old Macs sick of buggy and huge Netscape and IE would
like a nice "lite" graphical browser with a bit more knowledge of current
tags.

-Cari D. Burstein

--------------------

(Cari D. Burstein) writes:

>SNIP< 
> I'd have a hard time using it today because of the lack of frames support
> (I don't need frames themselves, but it'd be nice if it were updated to
> the level that Lynx supports), tables support (not a huge deal, but for
> some pages it can get messy), and if I remember correctly, progressive
> jpegs didn't display.

Some good news, some bad -- it does tables now (version 2.0 is available
somewhere), but still not frames and progressive jpegs. I got version 2
on a CD with a magazine, so I don't know where it is available on the
Internet. Even without frames, it is still useable for many of my
purposes. 

> 
> If someone were to get it now and to fix it up a little to support some of
> the current stuff in a fashion to at least allow it to navigate sites
> using these things, it'd be a nice browser without a lot of overhead. I
>SNIP<

Yepper -- The Mac's own Opera, as it were. As an alternative you might try
CyberDog. We are really getting away from ciwah and into ciwmacbrowsers or
something, so I will bid this thread adieu.


Phil Stripling

--------------------

(Geoffrey Peters) wrote:
> Another to consider: there are still a lot of old, pre-colour Macs out
> there using the internet, running ancient programs such as Mosaic 1.00a
> or ElNet's MacWeb.

Except for the fact that the developers gave up on it, MacWeb was far and
away the best browser of its time.  Absolutely blew the doors off Mosaic,
Netscape, and Explorer for rendering speed.


Roy Smith


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

ROADRASH@MINDremovethisSPRING.COM (Check the signature please!)
schreef op Fri, 02 Oct 1998 02:26:21 GMT in
<36143701.3808499@news.mindspring.com> in
comp.infosystems.www.authoring.html:
<snip>
: "Lack of restraint" may not be *the* problem, but IMO it's (in theory)
: the one that's most easily solved.  Simple observation of the "golden
: rule" [1]would go a long way towards making this newsgroup better.
: 
: [1]  "Do unto others as you would have them do unto you."
: 
: Simple, eh?

In principle, yes, but let me introduce you to:

---The Way Of The Newbie---

1)  They surf the web.
2)  They make their own HP.
3)  Something doesn't work.
4)  They search the web.
5)  They find out newsgroups exist.
6)  They read news.announce.newusers.
7)  They find ciwah.
8)  They read the FAQ.
9)  They read recent posts, and search dejanews.
10) They post their question.

Nice huh, but *oh shock* most only do steps 2,3,5,7 and 10. so:
11) They get flamed.

Should they not be flamed?, They'll never learn otherwise! (Get a
polite answer, think: ''fine'', and ask YASQ?)

How does this bear on restraint? (Don't mention/mind the trolls, they
_want_ to be flamed! (they feed on fire.))

<snip>


Buzz

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

(Timo L.) wrote:

>Sortof like the IMG-Tag, just that instead of inluding an image, it's
>supposed to include the text from the source.

This is Frequently Asked Question number 15. Please consult
http://www.htmlhelp.com/faq/wdgfaq.htm#15


Jukka Korpela

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

...
The content itself is fine, well written, but the markup seems to need some
fine tuning by hand. Have you run the pages through a validator and a
linter? You might pick up some helpful suggestions. Take a look at 
http://ugweb.cs.ualberta.ca/~gerald/validate/
   A Kinder, Gentler HTML Validator
http://www.webtechs.com/html-val-svc/
   HTML Validation Service
http://www.imagiware.com/tools.html
   Link to Doctor HTML for page checks.
http://www.cast.org/bobby/
   CAST: Bobby
http://home.paralynx.com/sky/
   An Adventure in Awareness
http://www2.cdepot.net/~mist/access.htm
   Web Accessibility sites
http://www.blooberry.com/
   Helpful information on what tags do in various browsers


Phil Stripling 

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Bernd Luehr wrote:
>I want to convert Ansi-Textfiles to HTML.
>I am searching for a tool, that can do that.

The c.i.w.a.tools group would be more appropriate, but the basic
problems and logic behind such conversions looks like an HTML question
to me, so I'll discuss it shortly.

First, I'd like to clarify that you probably mean plain text files in
Ascii encoding. ("Ansi", or "ANSI", is the name of a standards
institute which has produced lots of standards for quite different
things. Sometimes "Ansi" refers to some character encoding, but the
denotations vary so much that one should really name the specific
encoding.)

A plain text file is a sequence of characters and control codes
("control characters") such as line breaks. It may have most
complicated logical structure, but there is no uniform, standardized
way to _express_ that structure.

An HTML document is a document structured using a specific markup
language.

Thus, to convert a plain text to an HTML document you need to define
_how_ the conversion is to be done, _logically speaking_ (never mind
the techniques now). For example, you might define that an empty line
in the plain text indicates a paragraph break (so <P> must be
generated into the HTML document). Some rules like that are pretty
obvious. But if you want any better structure into the HTML document
you need to define some useful mapping from the textual structure. In
particular, how are headings presented? There are many usages in this
area. Some text to HTML converters seem to try make rather "clever"
guesses, but in my (limited) experience with them, they just guess
wrong too often.

So _generally speaking_, the process cannot be automated. For text
files written according to some specific rules, _specific_ conversion
program could be developed relatively easily.  But in the lack of
uniformity, it would probably save time to do things simply:
- create a very simple partial converter which just changes empty
lines to <P>s, "entifies" the characters <>& and things like that
- then use just an editor - preferably, a programmable editor like
Emacs - to add HTML markup for headings, emphasis, lists, and things
like that; of course you need to see the intended structure and
express it in HTML, but in most cases it's something that humans learn
to do fast and programs have great difficulties with.


Jukka Korpela

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Auke Reitsma wrote:
> I'm trying to understand DTD content models. The one thing I don't
> understand is the meaning of the '+' for the head element and the second
> '+'  for the body element:

>   <!ELEMENT BODY O O (%block;|SCRIPT)+ +(INS|DEL) -- document body -->
>   <!ELEMENT HEAD O O (%head.content;) +(%head.misc;) -- document head -->

Instead of me trying to describe the syntax of DTD's with the
possibility of saying something wrong, go here...

http://www-tei.uic.edu/orgs/tei/sgml/teip3sg/index.html

...this is one of the easiest to understand descriptions I have ever
seen on SGML and their DTD's. And it does give clear answers the
questions you stated up here.


Jan Roland Eriksson

--------------------

Auke Reitsma wrote:

> I'm trying to understand DTD content models.

Try http://www.ozemail.com.au/~dkgsoft/html/sgml.html for a quick tour
of the parts usually needed for understanding an HTML DTD.

> The one thing I don't
> understand is the meaning of the '+' for the head element and the second
> '+'  for the body element:
> 
>   <!ELEMENT BODY O O (%block;|SCRIPT)+ +(INS|DEL) -- document body -->

The "+" after the parentheses means that there may be one or more
occurrences.

The + before INS/DEL says that these may be present in addition.  I.e
that they don't belong to %block, but they could occur in addition as
elements of BODY.  There's a technical fiddle involved here, because INS
and DEL are sometimes block-level and sometimes inline-level constructs.
This needs a bit of fiddling to accommodate in the DTD, as I understand
it. (if we're lucky then Arjun will pop up and give a more accurate
explanation than I can). 


If you look at e.g PRE you find the converse happening:

<!ELEMENT PRE - - (%inline;)* -(%pre.exclusion;) -- preformatted text -->

There are some kinds of things included in the %inline content model
that aren't allowed inside PRE.  They are excluded by the -(...)
construction after the (%inline)*

does that help?


Alan J. Flavell

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Cornuts wrote:

>Can a ftp link (like
>ftp://archive:oldies@archive.netscape.com/archive/index.html ) work in a
>non netscape or explorer browser like AOL? If anyone has a browser other
>than navigator or explorer can you see if this works. 

HotJava 1.1.4: converts the example above to
ftp://archive/archive/index.html and hence can't find it.

Mosaic 3.0: works fine.

Oracle PowerBrowser 1.5.04: Doesn't handle ftp from the look of it.

Opera 3.21: works fine.

UdiWWW: doesn't support ftp.

Ariadna: supports ftp but can't handle the above address.

Lynx 2.8.1: works fine

pwWebSpeak: doesn't appear to support ftp.

BTW, it works in Netscape as far back as 1.22

That's all I have on my machine at the moment.


Steve Pugh

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Phibit 100 wrote:

>I'm having a problem at www.chrislevis.com/nakedfriends.html.   I have an
>onClick event that will work in Exporer, but not in Navigator (both newest
>versions).

Then it's a problem with browser support to the client-side scripting
language, assumably JavaScript. Ergo, not an HTML problem.

But there are things which you can do in HTML to prepare for
situations where scripts for events don't work. (Actually, your page
does not work in IE either, unless a user has left a security hole
wide open, surfing around without disabling JavaScript.)

><BODY bgcolor="FFFFC0">

This alone could ruin everything, by the way. Due to the syntax error,
this might set background error to anything, depending on the
browser's error recovery mechanism - FFFFFC0 is undefined as a color
value. Moreover, even if you used technically correct "#FFFFC0", you
might manage to set background color identical to or very similar to
the text color. See
http://www.hut.fi/u/jkorpela/HTML3.2/5.11.html

>             <td><a onClick="document.bigx.src='/pagepics/joeblack.jpg'";><img
>src="/pagepics/joeclothes.jpg" height="215" width="180" name="joeclothes"></a>

In addition to the lack of ALT attribute for the IMG element, this
suffers from the following problem: when client side scripting is
disabled or not supported by the browser, there is no way in which the
user could access the image /pagepics/joeblack.jpg (except through
peeking at the source code!). Not ever by simply following a link.

A natural solution would be to include HREF="pagepics/joeblack.jpg"
into the A tag (perhaps with BORDER="0" in the IMG element - in a case
like this where you would rather explicitly say that the image is a
link, it could make sense to suggest the suppression of a border).

However, I'm afraid browsers would easily get confused when JavaScript
is enabled in such cases. It would probably be safer to include a
SCRIPT element which writes out the code for the situation where
scripting is enabled and, after that, a NOSCRIPT element which
contains a simple <A HREF=...><IMG ...></A>. What the HREF points to
is up to you to decide. It could be just the image. Or it could be
something that more closely corresponds to what happens when scripting
is enabled and works - it could point to an HTML document similar to
the current one but with the image replaced by another image. This
would get somewhat complicated of course and you would probably like
to use server-side scripting to generate such documents on the fly.

BTW, your code now implies - when it works - that after clicking the
page still tells the user to click on the image to get the image
changed - to something it now _is_, and clicking on it wouldn't reveal
anything new.


Jukka Korpela

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

CB wrote in message ...
>
>How come so many web sites now have a main page with tiny print and a
>ton of text and links crammed onto it? Why is this good? It makes me
>spend time searching for what I want to click on because there's so
>many choices. One would think there should be more of a hierarchy,
>where there's not so many links on the main page.


What one group of people consider "good design" does not neccesarily
constitute the same thought in a second group of people. Simpler put:
What the Americans consider good design may not be the same as what
the Germans will consider good design. Naturally, the reverse is true.

I also dislike the "sardine can" approach to home pages, but apparantly,
this is what the "mob" wants (who does all this market research anyway?).

One of our clients complained there was too much white space on the
pages, so we wrote back saying "We, removed all the white-space, and
replaced it with breathing-space. We agree, it does look better". He got the
message (luckily).

-iSs-

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Jason Berkan wrote:
>
> I didn't know that servers would automatically send out the
> Last-Modified header.  But you did use "most" HTTP servers.  If my
> server does not send that header, all I can do is advise my friendly
> webmaster that I would like the server to send that out.

D.J.Deloire has a wonderful site for checking stuff like this.
Servers http headers especially at...

http://www.delorie.com/web/headers.html

Check out all the rest of his tools too, they _are_ really good.
Here...

http://www.delorie.com/web/

Point your webmaster to it too, he might learn a thing or two.

-- 
Jan Roland Eriksson

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Christopher M. Glenn, Ph.D. wrote:

> In IE4, the sytax below automatically word wraps (there is a vertical
> but not a horizontal scroll bar).  In V4 of Netscape, there is a
> horizontal scroll bar, and an entered sentence does not word wrap.  As
> I use IE, perhaps it is my (mostly default) settings in Netscape, but
> I'd like there to be a vertical scroll bar if needed, and automatic
> word wrap.  [And no word limit.]
>
> Is there in-common syntax for both IE4 and Netscape 4?
>
> Thank-you.
>
> <textarea name="Q10" rows=7 cols=50 size=50></textarea>

Try: <textarea name="Q10" rows=7 cols=50 size=50 wrap="hard"></textarea>



--
Jerry Park

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Buzz
-- 
Validator: Invalid: Element <G> is not defined.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Liz Knuth wrote:
> catnip wrote: 
> > Nobody wrote:
> > > Where do you start School. or just teach your self. and put out adds
> > > that you make web pages for xxx amount of money on the web.Does anyone
> > > make a decent living doing it this way?
> >
> > I don't...really. That's why I do phone sex on the side. It brings in
> > the bucks and I can work on web sites while I'm 'talking'.
>
> I think you took that "xxx" too narrowly.  And you didn't tell Nobody 
> that you webmistress an xxx page that promotes your other, er, 
> service.  _They_ bring in the bucks.  Your secret is safe with us.

Ssshhhh!! Now everyone will get into developing xxx sites. Geez. :-(

catnip

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Let me see if I've got this straight.

When writing HTML, the following characters can simply be typed as they are
(without the "&#n;" character escapes):

************
(copied from http://www.hut.fi/u/jkorpela/HTML3.2/3.1.html)
! " # $ % & ' ( ) * + , - . /0 1 2 3 4 5 6 7 8 9 : ; < = > ?@ A B C D E F G
H I J K L M N OP Q R S T U V W X Y Z [ \ ] ^ _ ` a b c d e f g h i j k l m n
op q r s t u v w x y z { | } ~
************

The following characters can be viewed by all browsers, but they need the
character escapes (i.e. instead of typing ALT-0174 for "®", I would need
"&#174;"):

ˇ ˘ Ł ¤ Ą ¦ § ¨ © Ş « ¬ ­ ® Ż° ± ˛ ł ´ µ ¶ · ¸ ą ş » Ľ ˝ ľ ż Ŕ Á Â Ă Ä Ĺ Ć Ç
Č É Ę Ë Ě Í Î ĎĐ Ń Ň Ó Ô Ő Ö × Ř Ů Ú Ű Ü Ý Ţ ß ŕ á â ă ä ĺ ć ç č é ę ë ě í î
ďđ ń ň ó ô ő ö ÷ ř ů ú ű ü ý ţ ˙
************

All other characters, for the sake of universal accessability, should be
avoided.

Is this correct?


Gern Blandston

-----------------------------------

>When writing HTML, the following characters can simply be typed as they are
>(without the "&#n;" character escapes):
>
>************
>(copied from http://www.hut.fi/u/jkorpela/HTML3.2/3.1.html)
>! " # $ % & ' ( ) * + , - . /0 1 2 3 4 5 6 7 8 9 : ; < = > ?@ A B C D E F G
>H I J K L M N OP Q R S T U V W X Y Z [ \ ] ^ _ ` a b c d e f g h i j k l m n
>op q r s t u v w x y z { | } ~
>************

Almost.  You need to escape the & (&amp;), < (&lt;), > (&gt;), and "
(&quot;) where they would be mistaken for HTML markup.

>The following characters can be viewed by all browsers, but they need the
>character escapes (i.e. instead of typing ALT-0174 for "®", I would need
>"&#174;"):
>
>ˇ ˘ Ł ¤ Ą ¦ § ¨ © Ş « ¬ ­ ® Ż° ± ˛ ł ´ µ ¶ · ¸ ą ş » Ľ ˝ ľ ż Ŕ Á Â Ă Ä Ĺ Ć Ç
>Č É Ę Ë Ě Í Î ĎĐ Ń Ň Ó Ô Ő Ö × Ř Ů Ú Ű Ü Ý Ţ ß ŕ á â ă ä ĺ ć ç č é ę ë ě í î
>ďđ ń ň ó ô ő ö ÷ ř ů ú ű ü ý ţ ˙
>************

Wrong.  You can type ALT-0174 as long as your document is encoded as
ISO-8859-1.  See
<http://www.htmlhelp.com/tools/validator/charset.html> for more
information on specifying character encoding.

>All other characters, for the sake of universal accessability, should be
>avoided.

For the sake of universal accessibility, some of the characters you've
cited should also be avoided.  See
<http://ppewww.ph.gla.ac.uk/%7Eflavell/iso8859/iso8859-mac.html>.

If you're writing in a language for which ISO-8859-1 is a sufficient
encoding (e.g., English, Spanish, Portuguese, Italian), then in
general I would avoid characters outside ISO-8859-1 due to poor
browser support.  However, saying that "All other characters ...
should be avoided" would be overly cautious--and make your task
impossible--if you were writing in a language such as Chinese,
Japanese, Korean, Russian, Greek, Vietnamese, Thai, etc.


Liam Quinn

---------------------

Liam Quinn (liam@htmlhelp.com) wrote on MDCCCLXIV September MCMXCIII in
<URL:news:361dbb7b.2699363@news.sprint.ca>:
++ 
++ Almost.  You need to escape the & (&amp;), < (&lt;), > (&gt;), and "
++ (&quot;) where they would be mistaken for HTML markup.

Yes, but I've never managed to find an example where you would need &gt;.
(Save for broken browsers).



Abigail

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Stephen o'o' Boyles wrote:
> What, if any, difference is there in putting
> "
>    (versus)
> &quot;
> into a html file, as they both appear to give the same result.
>   Is there any reason to choose one over the other?


   The standard quote sign " is adequate under normal
circumstances.  The literal entity &quot; is normally
used where the sign itself is used to enclose a string, 
such as an attribute value:

<IMG SRC="pix/barney.gif" ALT="My &quot;pet&quot; dinosaur">


Warren Steel

---------------------------

On Thu, 08 Oct 1998 09:23:54 GMT, Stephen o'o' Boyles spoke unto the 
people of comp.infosystems.www.authoring.html in these words: 

> What, if any, difference is there in putting
> "
>    (versus) 
> &quot;

> into a html file, as they both appear to give the same result.
>   Is there any reason to choose one over the other?

&quot; was mistakenly left out of HTML 3.2.  So if you have any docs 
that should conform to 3.2, use &#34; instead.

But I suspect that wasn't your question.  :-)

There are 4 special characters in HTML: < > & " .

< often means, "A tag starts here."  If a tag _doesn't_ start there, 
you just want to display that character, you should use the entity &lt; 
or its numerical equivalent &#60; .

> means, "A tag ends here."  I have not seen a browser choke on this if 
outside a tag, and not preceded by an un-entified < .  It could be 
big trouble _inside_ a tag, say, in your META description or ALT text 
or inside comments.  If you want to play it absolutely safe, substitute 
&gt; or &#62; when > does not end a tag.

" in HTML, as in some other languages, marks the beginning and end of 
a string, e.g., <a href="http://www.yahoo.com/">Yahoo!</a>.  In this 
example, A is the element, href is the attribute, and the URL is the 
value.  Values with anything besides letters a-z and A-Z (no 
diacritics/accents), numbers, hyphen, or period/full stop must be in 
quotes.  You can use double quotes (", not smart quotes or curly quotes) 
or single quotes (').  You can also use pairs of single quotes inside 
double quotes and vice versa, e.g., 
<IMG SRC="pix/fred.gif" height=175 width=57 ALT="Fred says, 'Hello!'">
<META NAME="keywords" CONTENT='New York, "Big Apple", Knicks, sports'>
But you can't use double quotes nested inside double quotes, because 
the browser will think your string ends at the second quote.  You can't 
have 
<META NAME="description" CONTENT="New York, the "Big Apple", my hometown.">
but you _can_ use entities when you need a double quote inside a 
string.
<META NAME="description" 
CONTENT="All about the Beatles' &#34;Hey Jude&#34;.">
I religiously entify all quotes that are not the start or end of a 
string, but I don't know anyone else who does.  So just be careful in 
situations where you have quotes within quotes, or where a browser might 
choke on your tags.

And finally, the ampersand (&) means, "An entity starts here."  If you 
work for AT&T, you really should use the entity, or a parser will 
tell you that there is no such entity as "&T".  The entity for 
ampersand is &amp; or its numerical equivalent &#38; .  Special cases: 
Some bozos use the ampersand in their domain names or file names.  Bad.
Causes some browsers to croak, and will also have validators raise a 
stink.  In this case, use the _hex_ equivalent, %26.  For example, if 
the original URL was http://www.r&b.com/ , in your a href make it 
http://www.r%26b.com/ .  The other common problem is that many CGI 
programs use the ampersand to separate fields, e.g., 
http://www.countersrus.com/count.cgi?id=nerdguy&account=geekhome&style=35
The best way to fix this is to try substituting %26 (the hex solution 
takes care of the http server problems with &).  If that doesn't work, 
_some_ scripts will work just as well if you use a pipe (broken 
vertical bar, |) instead of & .  And if neither of those do the trick, 
try &amp; or &#38; in the URL.
 

Elizabeth T. Knuth 

---------------------------

On 10 Oct 1998, Liz Knuth wrote:

(amongst an otherwise unexceptional reply)

>  The other common problem is that many CGI 
> programs use the ampersand to separate fields, e.g., 
> http://www.countersrus.com/count.cgi?id=nerdguy&account=geekhome&style=35

That seems to be a URL pretending to be a GET form response using the
application/x-www-form-urlencoded representation.  Very useful, indeed. 

> The best way to fix this is to try substituting %26 

As a simulated form response, this would be absolutely wrong, I'm
afraid. Writing %26 means that you're trying to _avoid_ the ampersand
acting as a delimiter, and have it be part of the data instead.  Refer
to http://www.w3.org/MarkUp/html-spec/html-spec_8.html#SEC8.2.1 for the
authoritative specification.  Applying your procedure to the above URL
would mean that you want to set the argument named "id" to the value
"nerdguy&account=geekhome&style=3".  I'm not sure how many combinations
of browsers and servers would get this pathological case correct, but if
it seems to "work" as you seem to be wanting it. then somebody would be
misbehaving.

Or else you are defining some kind of private syntax, that could indeed
appear to work, but contradicts the forms syntax in the specification.
And then how are you going to be able to supply an ampersand character
as part of the data?

The correct _URL_ for this situation is exactly the one you wrote
above:

http://www.countersrus.com/count.cgi?id=nerdguy&account=geekhome&style=35

and the rule for expressing that URL as the value of an HREF in HTML
is to turn the & signs into &amp; (or the &#number; equivalent).

Some ancient browsers used to mishandle that, but by now it's been my
experience that the browser versions in current use get it right. 

>  If that doesn't work, 
> _some_ scripts will work just as well if you use a pipe (broken 
> vertical bar, |) instead of & . 

Somewhere in an old specification or tutorial (which I can't immediately
find just now, sorry), it recommended implementing form evaluation
scripts so that they would accept an alternative delimiter, so as not to
provoke these old browser bugs; I recall "semicolon" being recommended,
but I'm sure some other character would be fine too (let's hope your
vertical bar doesn't get turned into a unix pipe in some carelessly
written CGI script).  

When you say "pipe" (the function that is typically represented by the
ASCII vertical bar character), then say "broken vertical bar" and
exemplify it with what appears to be an ASCII vertical bar (the unbroken
one), you leave us in some confusion.  But the iso-8859-1 broken bar
(decimal 166) could not validly be included in a URL without encoding
it, %A6.  I assumed that you meant the ASCII vertical bar, rather than
the iso-8859-1 broken bar (code point 166). Here's a handy reference:
ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-1.TXT
albeit with the code points in hexadecimal (7C, A6 respectively).

> And if neither of those do the trick, 
> try &amp; or &#38; in the URL.

That's the correct way of using the ampersand here.  Your alternative of
designing the evaluation script to honour an alternative delimiter and
thus avoid the issue is also a perfectly fine approach.  But your first
answer, if understood to be the URL of a GET-type FORM submission, was
emphatically wrong, and, coming from such a reliable commentator, I
earnestly hope to see you posting a clarification.


Alan J. Flavell

----------------------------

On Sat, 10 Oct 1998 12:04:47 +0200, "Alan J. Flavell"
<flavell@mail.cern.ch> wrote:

>Somewhere in an old specification or tutorial (which I can't immediately
>find just now, sorry), it recommended implementing form evaluation
>scripts so that they would accept an alternative delimiter, so as not to
>provoke these old browser bugs; I recall "semicolon" being recommended

That's right; it's in  the HTML 2.0 specification. In the plain text
version of it, RFC 1866, it's a note in section 8.2.1. The sound
advice there hasn't been taken by CGI implementors, I'm afraid.

>When you say "pipe" (the function that is typically represented by the
>ASCII vertical bar character), then say "broken vertical bar" and
>exemplify it with what appears to be an ASCII vertical bar (the unbroken
>one), you leave us in some confusion.

I have no access to the real ASCII standard, but ISO 646-1973 (the
corresponding international standard) uses the name "vertical line"
for the character meant here. It's also the Unicode name; see
http://www.hut.fi/u/jkorpela/latin1/3.html#7C
I recommend using Unicode names for characters. As you point out,
people may easily get confused with characters if miscellaneous names
- often based on the _appearance_ of a character in some fonts - are
used. The "internationalization" of HTML and the Web - i.e. better
abilities to use wider character repertoires - will increase the risks
of such confusions.


Jukka Korpela

----------------------------------

(Claus André Färber) writes:
> Nick Kew <nick@webthing.com> schrieb:
>> A more satisfactory solution would be to create a new Content-Type,
>> identical to URLEncoding except for the separator.  This would remove
>> the HTML-CGI conflict altogether, except as a matter of history.
>> Of course, this will require a change in the browsers, ...
> 
> There is no conflict. The only problem is that there are broken browsers  
> that can't handle href="foo.cgi?v1=a&amp;v2=b" correctly.

There is a problem: escaping &amp; is confusing.  For new authors, the
different types of escaping (HTML- and URL-) is bad enough, and the
need to _combine_ them in this particular situation could get a bit
mind-boggling, methinks.  For veterans, there's force of habit: even
reputable sources aren't consistent on this (I used WebTechs validation
service for a couple of years, and both their online service and download
software allowed unescaped ampersands in QUERY_STRING - which is my excuse
for getting into a bad habit).

> Any change will neither improve the situation with these old (obsolete)  
> browser (versions), nor with current browsers, where there is nothing to  
> improve.

But it's authors, not browsers, who need helping out.

-- 
Nick Kew

--------------------------------

catnip wrote:

> I've been following this thread...kinda. Can someone please summarize
> where we are right now? Thanks.

 1. In attributes, & should be escaped as &amp;, and in some cases needs
to be if what follows it could be misinterpreted as completion of an
entity.
 2. If you want to have an ampersand in the URL that is not a field
separator, you have to use %26.
 3. The browser is supposed to do the translation from &amp; to & when it
sends the query to the server.
 4. CGI authors may elect to handle broken browsers by treating receipt of
&amp; in a URL as simply & (fault tolerance) or not and instead encourage
people to upgrade to non-broken browsers (fault intolerance).

Greg Berigan

---------------------------------

Greg Berigan wrote:

>  1. In attributes, & should be escaped as &amp;, 

Excuse me.  In the language of RFCs: & MUST be escaped as &amp;  in this
context.  A failure to do so is an HTML error, which MAY be fixed up by
client agents which choose to do so. 

> and in some cases needs
> to be if what follows it could be misinterpreted as completion of an
> entity.

In a practical sense you may well be right, but it would only be fair to
make clear that in terms of the HTML spec (RFC1866 and subsequent
recommendations) that would be relying on error fixup, rather than
following the spec.

>  2. If you want to have an ampersand in the URL that is not a field
> separator, you have to use %26.

Agreed.  I tried it, with some obvious browser versions: it works to
spec, surprising as that may seem.

>  3. The browser is supposed to do the translation from &amp; to & when it
> sends the query to the server.

The spec requires it.  Apparently Amaya is broken in this respect.

>  4. CGI authors may elect to handle broken browsers by treating receipt of
> &amp; in a URL as simply & (fault tolerance) or not and instead encourage
> people to upgrade to non-broken browsers (fault intolerance).

That's OK as far as it goes, but CGI authors may elect to handle some
other delimiter (semicolon recommended) as a workaround, so that GETs
can be coded in an HREF without provoking the bug.  If there are any
real browsers that do still provoke the bug, that is.  (Sorry, Amaya is
an interesting project, but a real browser it is not, and certainly not
so long as it flies in the face of this clearly mandatory requirement of
the spec).


Alan J. Flavell

--------------------------

> Alan J. Flavell wrote:
> On Tue, 13 Oct 1998, Greg Berigan wrote:

>> 1. In attributes, & should be escaped as &amp;,

> Excuse me.  In the language of RFCs: & MUST be escaped as &amp;
> in this context.  A failure to do so is an HTML error, which MAY
> be fixed up by client agents which choose to do so.

Is this a requirement of HTML?  I don't see anywhere in the HTML spec
which prohibits, ALT="Tom & Rachel fell in the river", which is of
course legal by SGML's rules (though not XML's).

Ampersand MUST be written as "&amp;" or "&#38;" when followed by any
SGML name-start character (letters, mainly) or by "#" and a digit.


>> 3. The browser is supposed to do the translation from &amp; to &
>>    when it sends the query to the server.

I'd rephrase that as "when it parses the HTML", to emphasise that HREF
attributes are treated exactly the same as any other.


Toby Speight

-----------------------------

Toby Speight wrote:
>> Excuse me.  In the language of RFCs: & MUST be escaped as &amp;
>> in this context - -
- -
>Is this a requirement of HTML?  I don't see anywhere in the HTML spec
>which prohibits, ALT="Tom & Rachel fell in the river", which is of
>course legal by SGML's rules (though not XML's).

In that example, the ampersand need not be escaped.

>Ampersand MUST be written as "&amp;" or "&#38;" when followed by any
>SGML name-start character (letters, mainly) or by "#" and a digit.

Yes. It seems that people are discussing different things here: the
general rule (which is what you wrote) and the specific, and very
common, situation where the ampersand occurs in a URL containing a
query string with several name=value pairs, such as
http://www.countersrus.com/count.cgi?id=nerdguy&account=geekhome&style=35
mentioned as an example in this thread. I suppose Alan meant the
latter (note the words "in this context"). And of course he is quite
right in his statement, since in this context the ampersand is always
followed by a letter. (Well, my daily nitpicking: it seems that the
NAME attribute value in a form field is declared as CDATA, so it
theoretically need not begin with a letter. Unless I am exercising my
daily routine of missing something, of course.)

Jukka Korpela

------------------------------

On Wed, 14 Oct 1998, catnip wrote:

> On 14 Oct 1998 11:22:33 +0100, Toby Speight
> <Toby.Speight@digitivity.com> wrote:
> 
> >Alan> Excuse me.  In the language of RFCs: & MUST be escaped as &amp;
> >Alan> in this context.  A failure to do so is an HTML error, which MAY
> >Alan> be fixed up by client agents which choose to do so.
> >
> >Is this a requirement of HTML?  

"The context" here is, a URL that ends with
                    ?attr1=val1&attr2=val2&attr3=val3 etc.

> >I don't see anywhere in the HTML spec
> >which prohibits, ALT="Tom & Rachel fell in the river", 

Sorry, that's outside of the scope that I was intending to cover.

> >Ampersand MUST be written as "&amp;" or "&#38;" when followed by any
> >SGML name-start character (letters, mainly)  [...]

Exactly so, and that's just what's happening in "this context".

> This is exactly why I asked for a summary of this thread. It seems to
> me that an issue like this ought not be so contentious 

No ma'am, it isn't "contentious", it's just that I didn't explain
clearly what I meant by "this context", and Toby has interpreted it
more broadly than I had intended.

> and that it would have a techically correct bottom line. 

It has, as you see (Toby has already stated the relevant principle, 
above). 


Alan J. Flavell

------------------------------

On 11 Oct 1998 23:47:00 +0200, claus+usenet@faerber.muc.de
(=?ISO-8859-1?Q?Claus_Andr=E9_F=E4rber?=) wrote:

>There is no conflict. The only problem is that there are broken browsers  
>that can't handle href="foo.cgi?v1=a&amp;v2=b" correctly.

No, the real problem in practice is that _authors_ fail to write their
code that way. The statement in HTML 2.0 we're discussing about refers
to "trouble of escaping `&' characters". After all, this _is_ a
difficult topic to ordinary authors. In addition to getting the big
picture of such URLs right, they have to deal with two different kinds
of "escapes", namely using %xx for some characters in URLs due to URL
syntax and using &amp; (and perhaps other similar notations) in URLs
when they occur as attribute values in HTML. Each of these in itself
confusing, and when people confuse them with each other - as we have
seen people do quite often - it gets really problematic.

In principle, the problems can be handled by learning to use both
kinds of "escapes" and remembering to use them properly. And it seems
we have to do that in practice, too

Jukka Korpela

--------------------------------

EKNUTH@tiny.computing.csbsju.edu (Liz Knuth) wrote:
>Stephen o'o' Boyles wrote: 

>> What, if any, difference is there in putting " versus &quot;
>> into a html file, as they both appear to give the same result.
>>   Is there any reason to choose one over the other?

> There are 4 special characters in HTML: < > & " .

Since it is apparently allowed to use ' as a substitute for " in HTML, (I
stick with " and &quot; where needed) I'd classify ' as a special
character as well, requiring entification if present inside such a
delimited attribute value.

> < often means, "A tag starts here."  If a tag _doesn't_ start there, 
> you just want to display that character, you should use the entity &lt; 
> or its numerical equivalent &#60; .
> 
> > means, "A tag ends here."  I have not seen a browser choke on this if 
> outside a tag, and not preceded by an un-entified < .  It could be 
> big trouble _inside_ a tag, say, in your META description or ALT text 
> or inside comments.  If you want to play it absolutely safe, substitute 
> &gt; or &#62; when > does not end a tag.

Indeed, you can use < inside tags and > outside tags freely, with the
exception of certain parsers' problems with <!-- comments> -->.

> You can use double quotes (", not smart quotes or curly quotes) 
> or single quotes (').  You can also use pairs of single quotes inside 
> double quotes and vice versa,

Certainly not required to be paired, or you couldn't use contractions.

> But you can't use double quotes nested inside double quotes, because 
> the browser will think your string ends at the second quote.

Are you sure on this?  I need to write a simple parser right now and if
ALT='This isn"t right' sets the ALT text to "this isn" then there's a
problem.  Is this correct according to the specifications or are there
browsers that get this wrong?

> Special cases: 
> Some bozos use the ampersand in their domain names or file names.

I don't think ampersand is even allowed in domain names.  I've never heard
or seen of such a thing, even as part of the machine-name portion of the
fully qualified domain name (FQDN).

Having to escape such ampersands for such domains would serve as a major
deterrant to anyone choosing to register one.


Greg Berigan

---------------------------------

On Sun, 11 Oct 1998 22:59:20 -0500, Greg Berigan spoke unto the people 
of comp.infosystems.www.authoring.html in these words: 

> EKNUTH@tiny.computing.csbsju.edu (Liz Knuth) wrote:

> Since it is apparently allowed to use ' as a substitute for " in 
> HTML, (I stick with " and &quot; where needed) I'd classify ' as a 
> special character as well, requiring entification if present inside 
> such a delimited attribute value.

I suppose, if you are using single quotes to begin and end the string, 
and you have the same character _in_ the string.

By that standard, we can say the hyphen is a special character too, 
although using an entity won't fix the situation where it's a 
problem.  Have the wrong number, and your comments break.

> > You can use double quotes (", not smart quotes or curly quotes) 
> > or single quotes (').  You can also use pairs of single quotes 
> > inside double quotes and vice versa,

> Certainly not required to be paired, or you couldn't use 
> contractions.

Not required to be pairs.  Can be pairs, though.

> > But you can't use double quotes nested inside double quotes, 
> > because the browser will think your string ends at the second 
> > quote.

> Are you sure on this?  

Yes.  Read it again.

> I need to write a simple parser right now and if 
> ALT='This isn"t right' 

That's not what I said.  If you want your ALT text to show up as 
This isn"t right
that's your business.  It will look ignorant, but it will show up.

What won't work is double quotes inside double quotes, e.g., 
ALT="Greg said, "This isn't right.""
Where does the string end?  At the second quotation mark.  This is 
why the quotation mark ought to be an entity if it doesn't begin or 
end a string.  Though you can certainly get by with only using an 
entity for quotation marks in situations like this.

> > Special cases: 
> > Some bozos use the ampersand in their domain names or file names.

> I don't think ampersand is even allowed in domain names.  I've never 
> heard or seen of such a thing, even as part of the machine-name 
> portion of the fully qualified domain name (FQDN).

Dunno for sure.  I can think of two links on the library pages 
where I had to put the ampersand in hex.  Now, it may be that it was 
a username or part of the path, but I thought that at least one was 
in a domain name.

> Having to escape such ampersands for such domains would serve as a 
> major deterrant to anyone choosing to register one.

Lots of us are using the tilde in URLs...  Ask Jukka how safe that is.


Elizabeth T. Knuth

-------------------------------------

EKNUTH@tiny.computing.csbsju.edu (Liz Knuth) schreef op 12 Oct 1998
06:35:49 GMT in <6vs805$nai@ash.computing.csbsju.edu> in
comp.infosystems.www.authoring.html:

: On Sun, 11 Oct 1998 22:59:20 -0500, Greg Berigan spoke unto the people 
: of comp.infosystems.www.authoring.html in these words: 
: 
: > EKNUTH@tiny.computing.csbsju.edu (Liz Knuth) wrote:
<snip>
: > > Special cases: 
: > > Some bozos use the ampersand in their domain names or file names.
: 
: > I don't think ampersand is even allowed in domain names.  I've never 
: > heard or seen of such a thing, even as part of the machine-name 
: > portion of the fully qualified domain name (FQDN).
: 
: Dunno for sure.  I can think of two links on the library pages 
: where I had to put the ampersand in hex.  Now, it may be that it was 
: a username or part of the path, but I thought that at least one was 
: in a domain name.

Just looked at rfc1101:

> 3.1. Network name syntax

>    The current syntax for network names, as defined by [RFC 952] is an
>    alphanumeric string of up to 24 characters, which begins with an
>    alpha, and may include "." and "-" except as first and last
>    characters.

Is that what you wanted to know?

<snip>

hth,


Buzz

-----------------------------------

In comp.infosystems.www.authoring.html,
EKNUTH@tiny.computing.csbsju.edu (Liz Knuth) wrote:
>gberigan@cse.unl.edu (Greg Berigan) wrote:
>>EKNUTH@tiny.computing.csbsju.edu (Liz Knuth) wrote:

>>> But you can't use double quotes nested inside double quotes, 
>>> because the browser will think your string ends at the second 
>>> quote.

>> Are you sure on this?  

> Yes.  Read it again.

Yes, you are quite right.  Mea culpa.

>>> Special cases: 
>>> Some bozos use the ampersand in their domain names or file names.

>> I don't think ampersand is even allowed in domain names.  I've never 
>> heard or seen of such a thing, even as part of the machine-name 
>> portion of the fully qualified domain name (FQDN).

> Dunno for sure.  I can think of two links on the library pages 
> where I had to put the ampersand in hex.  Now, it may be that it was 
> a username or part of the path, but I thought that at least one was 
> in a domain name.

In "FAQ" <http://www.internic.net/faq/new-reg.html> it is written:

| 3. What are the valid characters for a domain name and how long can it be?
| 
|       The only valid characters for a domain name are letters,
|       numbers and a hyphen. Special characters like the underscore
|       (_) or an exclamation mark (!) are NOT permitted. Domain names
|       cannot contain spaces or begin or end with a hyphen. 

>> Having to escape such ampersands for such domains would serve as a 
>> major deterrant to anyone choosing to register one.

> Lots of us are using the tilde in URLs...  Ask Jukka how safe that is.

I believe ~ has been officially moved from the unsafe to the safe
category.  Unfortunately I don't have a reference for this, so feel free
to treat it as anecdotal. :-)

Let's take this to comp.infosystems.www.authoring.misc.


Greg Berigan

--------------------------------

On Tue, 13 Oct 1998 14:34:19 -0500, Greg Berigan spoke unto the people 
of comp.infosystems.www.authoring.html in these words: 

> In comp.infosystems.www.authoring.html,
> EKNUTH@tiny.computing.csbsju.edu (Liz Knuth) wrote:

> >gberigan@cse.unl.edu (Greg Berigan) wrote:

> >> I don't think ampersand is even allowed in domain names.  I've 
> >> never heard or seen of such a thing, even as part of the 
> >> machine-name portion of the fully qualified domain name (FQDN).

> > Dunno for sure.  I can think of two links on the library pages 
> > where I had to put the ampersand in hex.  Now, it may be that it was 
> > a username or part of the path, but I thought that at least one was 
> > in a domain name.

> In "FAQ" <http://www.internic.net/faq/new-reg.html> it is written:

> | 3. What are the valid characters for a domain name and how long can 
> | it be?
> | 
> |       The only valid characters for a domain name are letters,
> |       numbers and a hyphen. 

I checked.  The two URLs that required my putting the ampersand in 
hex were: one path, one file name.  No domain names.  i.e., 

http://www.foobar.com/baz&bat/foo.html

and

http://www.foobar.com/baz&bat.html

So the deciding factor for how to get around ampersands in URLs is 
whether or not the ampersand is being used as a field separator in a 
query to a server side program.  If yes, then &#38; or &amp; .  If no,
%26 .


Elizabeth T. Knuth

------------------------------------

On Sun, 11 Oct 1998 eknuth@tiny.computing.csbsju.edu wrote:

> In article
> <Pine.HPP.3.95a.981010111032.12962C-100000@hpplus01.cern.ch>, "Alan
> J. Flavell" <flavell@mail.cern.ch> wrote:

> > (amongst an otherwise unexceptional reply)
> 
> Hmm.

Yes, I'm sorry, the word is "unexceptionable".  That's what I meant,
anyway.

> > > http://www.countersrus.com/count.cgi?id=nerdguy&account=geekhome&style=35
> 
> > > The best way to fix this is to try substituting %26
> 
> > Applying your procedure to the above URL would mean that you want to
> > set the argument named "id" to the value
> > "nerdguy&account=geekhome&style=3".

> OK, thanks for the correction.  The heart of the problem is that the
> ampersand has special meaning in HTML as well as in many, many CGI
> programs... and the meanings are not the same.

Let's take this calmly.  The correct URL is exactly as it was stated
above.  Now, you want to represent that URL as the value of an HREF
attribute in HTML.  Therefore, you must use the applicable HTML rule for
this situation, and "entify" the thing.  The HTML syntax rules are
unambiguous on this point (as someone mentioned, apparently Amaya gets
this wrong, but the browsers that are in actual use by readers nowadays
all handle it fine).

>  Consequently, one must
> determine whether the ampersand will screw up the HTML or the http
> server.

"Hmm", as you say...

> > and the rule for expressing that URL as the value of an HREF in HTML
> > is to turn the & signs into &amp; (or the &#number; equivalent).
> 
> So you're saying it's an HTML problem, and not on the server side of the
> process?

It isn't a server problem.  It isn't an HTML "problem" either, since the
syntax has been unambiguously stated since RFC1866.  It might, however,
be a browser problem, if the browser fails to honour that syntax.

> > Some ancient browsers used to mishandle that, but by now it's been my
> > experience that the browser versions in current use get it right.
> 
> Some _scripts_ and/or servers cannot handle it correctly. 

Neither the scripts nor the servers will ever see the entified
ampersand, if the browser is behaving correctly.  I can only recommend
breaking the problem up into its logical steps and analysing the actions
of each one according to the rules applicable to that step. 

> Using the
> same browser, same version, same platform, I can get some scripts to
> work correctly (that is, they serve up the requested data) when using
> an entity for the ampersand in the HTML markup, but other scripts or
> servers absolutely refuse to handle it unless I use a plain ampersand.

URLs of your demonstrations, please?

Btw, here's a quick demonstration of the bit about using %26:

  http://ppewww.ph.gla.ac.uk/~flavell/tests/urlget.html

but you've already conceded that point, thanks.

good luck

Alan J. Flavell

----------------------------------

Alan> Alan J. Flavell <URL:mailto:flavell@mail.cern.ch>

0> In
0> <URL:news:Pine.HPP.3.95a.981010111032.12962C-100000@hpplus01.cern.ch>,
0> Alan wrote:

Alan> Somewhere in an old specification or tutorial, it recommended
Alan> implementing form evaluation scripts so that they would accept
Alan> an alternative delimiter, so as not to provoke these old browser
Alan> bugs; I recall "semicolon" being recommended, but I'm sure some
Alan> other character would be fine too.

I'm significantly less sure - if the CGI program is ever invoked with
user input (from a HTML FORM), there's nothing to prevent that input
containing your "separator" character, which is likely to affect the
meaning of the input.  Ampersand and semicolon are protected, since
they have a syntactic significance in HTTP URLs, and clients are
therefore obliged to use the encoded form for those characters.


Toby Speight

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

On Thu, 08 Oct 1998 12:57:40 GMT, guelkev@de.polygram.com wrote:

>I have a problem with transferring German Umlauts through an URL. I want to
>call a script with some parameters:
>
><A HREF="/cgi-bin/script.pl?name=Gülke">

URLs currently do not support non-US-ASCII characters like u with an
umlaut.

>In the perlscript the variable of param(name) is "G�lke"
>How can I get the correct Umlaut in my script?

Since it's your script, you can use some convention and translate to
the correct character in the script.  For example, you could use

<A HREF="/cgi-bin/script.pl?name=G%5C374lke">

so that your script receives name=G\374lke.  374 is u with an umlaut
in octal, so you can convert that to the actual character easily
within your script.


Liam Quinn

-------------------

On Thu, 08 Oct 1998 12:57:40 GMT, guelkev@de.polygram.com wrote:

>I have a problem with transferring German Umlauts through an URL. I want to
>call a script with some parameters:
>
><A HREF="/cgi-bin/script.pl?name=Gülke">

Unless I've missed something, you can't do that. Not because of a
restriction but because of the fact that in _URLs_ (which is what you
need to have as the value of an HREF attribute) only US-ASCII
characters are allowed (and even they have several restrictions).

This is explicitly stated in the Forms section of the HTML 4.0 spec
(and things have always been this way):
     Note. The "get" method restricts form data set values to ASCII
     characters.


Jukka Korpela

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

lynx -head -dump http://www.net1fx.com/~martian/

HTTP/1.1 200 OK
Server: Apache/1.2.5 FrontPage/3.0.4
Last-Modified: Thu, 17 Sep 1998 19:50:48 GMT
ETag: "72813-772-36016818"
Content-Length: 1906
Accept-Ranges: bytes
Content-Type: text/html
Date: FRI, 09 OCT 1998 19:00:59 GMT
Connection: close

--------------

lynx -head -dump http://www.ludd.luth.se/~vk/

HTTP/1.1 200 OK
Date: Sun, 11 Oct 1998 17:02:13 GMT
Server: Apache/1.3b5
Connection: close
Content-Type: text/html


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Craig Berry writes:
>
>At whom is the "best viewed with"  thing targeted?

I have a theory.

It's used to rationalize "design" decisions to the CEO of a client,
when things aren't working with the browser the CEO's niece is using.


Tero Paananen

--------------

> I have a theory...

:) Well, you've hit on the gist of my theory, too.


Craig Berry

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Lazarus wrote:
> Darin McGrew wrote:
> > Lazarus wrote:
> > > Michael Goff wrote:
> > > >How do you make it so when some clicks your hyperlink it opens in another
> > > >window automatically?
> > >
> > > Use the <A HREF="wherever.whatever" TARGET="_new"></A>
> >
> > If this works, it does so by accident.  The correct value is
> > TARGET="_blank".
>
> It doesn't matter what the target is it could be "wiggyhoggy" as long as it
> is not a prior defined target name, the browser will open a new window.
>

Check the spec again.  target="_blank" is special, *always* suggesting a
new window.  target="_new" is invalid and expected to be ignored.


Erik Seaberg

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Liz Knuth wrote:
> catnip wrote:
> > Nobody wrote:
> > >
> > > Where do you start School. or just teach your self. and put out adds
> > > that you make web pages for xxx amount of money on the web.Does anyone
> > > make a decent living doing it this way?
> >
> > I don't...really. That's why I do phone sex on the side. It brings in
> > the bucks and I can work on web sites while I'm 'talking'.
>
> I think you took that "xxx" too narrowly.  And you didn't tell Nobody 
> that you webmistress an xxx page that promotes your other, er, 
> service.  _They_ bring in the bucks.  Your secret is safe with us.

Ssshhhh!! Now everyone will get into developing xxx sites. Geez. :-(

catnip

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Veronica Karlsson <dream@on.spammer> wrote:
: Sure, if you treat an "intranet" the same as you would a www site, no
: problem.

: I've recently had some interesting discussions with a "designer". He
: said "We don't need to concider [some www problem] because we're making
: an intranet and the company has a contract with Microsoft and will be
: using IE4 for the next two years anyway". He never seemed to think of
: the fact that in two years nobody (meaning people like him) will want to
: _touch_ something as old and useless as IE4... "What? Problems with new
: browser versions? Well, the site will probably be redesigned long before
: that anyway, so that won't be my problem...".  *grrrr*

Then there's also the peculiar habits that intranets have of turning into 
extranets.  Company A decides to standardize on IE4 as its internal 
browser (in order to simplify installation and maintenance) but chooses 
to write its internal documents in standard HTML rather than an 
IE4-specific flavor.  Company B decides to standardize on IE4 and to 
"take advantage of" all its proprietary HTML "features."  Company C, 
who's a vendor to both A and B, decides to standardize on something by 
Netscape, and then decides it wants to electronically exchange data with 
A and B.  Which of A and B are going to be able to pull it off without 
much hassle?


Eric Bohlman

---------------------

Then, the boss gets one of those cool PDA/CellPhone/WebBrowser and wonders
why he can't view the site.....(its happened twice on projects I've worked
on).


REBUS

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Will wrote:
>Anyway now for something on the subject......IMHO the  people who
>hate frames in this group are is not representative of the general users.


    That may be so, but that doesn't mean that general users
want, like, or understand frames.  Many less informed users
(according to their own reports to me, however anecdotal)
are confused and disoriented by frames.  They wonder why
their arrow keys fail to scroll the area they were looking
at--since they don't know about frames, they don't know how
to establish "focus" on one specific frame.  When frames
appear without borders, it's even worse, since they can't
find the frame they need.  They're also confused about what
to do with their back button.  Not to mention the bookmark
problem.  Frames break so many models that people rely on
(URLs, the history stack, the window as viewport) that 
it's no wonder they annoy so many people.  I suspect 
that most of their attraction is to authors, not Web users.

Warren Steel

--------------------------------

On Tue, 13 Oct 1998 vinman@my-dejanews.com wrote:

> What is the bookmark to a CGI that gets an input stream from a
> POST operation?????

POST operations are defined to be apt for operations that are not
idempotent.  In plain languages that means operations that change the
state of the universe.  Since you can't reverse the state of the
universe (you can't un-charge that new auto to your credit card merely
by pulling down a bookmark), your problem is insoluble. 

If, in fact, the operation is idempotent, then you could equally well
implement it as a GET (implementation details permitting), and that,
you could bookmark.

Alan J. Flavell

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

William Schlake writes:
>
> in excess of 95% is free for content.
                           ^^^

Hmm, I think you mispelled "of." (sorry, couldn't resist :-)


Decklin Foster

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

From the HTML4 DTD regarding element P ...
http://www.w3.org/TR/REC-html40/struct/text.html sect. 9.3.1

              <!ELEMENT P - O (%inline;)* -- paragraph -->
              ^-------^ | | | ^---------^ ||           |||
                  |     | | |      |      ||           |||
SGML declaration -+     | | |      |      ||           |||
Element name -----------+ | |      |      ||           |||
Start tag required=(-) ---+ |      |      ||           |||
End tag   optional=(O) -----+      |      ||           |||
Content model for element ---------+      ||           |||
SGML comment start -----------------------++           |||
SGML comment end --------------------------------------++|
SGML declaration close ----------------------------------+

(reading a DTD is not that difficult after all)


Jan Roland Eriksson

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

jm3553@my-dejanews.com wrote:
>BTW are you really as arrogant as your posts suggest or is your English just
>coming across that way.

No, my English is so far from perfect that I fail to express my
arrogance adequately. Asking stupid questions is forgiveable; we all
do that. Giving wrong answers is worse when they are given on a forum
where one could easily avoid the worst mistakes. Refusal to admit that
wrong answers are wrong, after people having clearly pointed out that
they are wrong,  is really bad. Trying to cast some smoke over such
stupidity by useless babbling and personal attacks is something I lack
words for.


Jukka Korpela

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

"My pedantry quota is exceeded, so I'll just add this:"

Jukka Korpela

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

On 15 Oct 1998 15:48:21 GMT, tprodin@ford.com (Timothy R Prodin) wrote:

> William  G. Schlake wrote:
> >So while the "experts" at W3C saw a need for an ending tag on things
> >like headings and lists, the boneheads goofed with the most commonly
> >used tag of all making the ending tag for paragraphs optional. 
> 
> But, from the content model of lists,  for example, that they must
> contain at least one list item.  The content model of <LI> shows that
> it can contain either block or inline elements - So, we have no
> unambigous method of determining when the containing is closed.  So,
> it's closing tag is not optional.  This isn't the case with <P>.

From the HTML4 DTD again, section 10.2

<!ELEMENT UL - - (LI)+  -- unordered list -->
<!ELEMENT OL - - (LI)+  -- ordered list -->

Both element UL and OL has start and end tags _required_ and both have a
content model of _minimum_one_ or more list item elements LI

<!ELEMENT LI - O (%flow;)*  -- list item -->

Element LI has the start tag required _but_ the end tag optional, and a
content model of %flow (which is eq. to %block | %inline)

This is fully OK for an SGML parser of course since any LI element will
automatically end if a new LI is found _or_ if a closing tag </UL> or
</OL> is found in the data stream.

And according to specs, LI can only exist...
inside <UL>..</UL> or...
inside <OL>..</OL>

There's nothing ambiguous here.

This is not "rocket science" as the saying goes, and any one who has
spent some hours writing text processing software do know how to parse a
text data stream into specified components. An I'm fully convinced that
programmers by the "big two" do know how to do that too.

Thing is, they have never been pushed in that direction since none of
the browsers is designed to do an SGML parse of text/html into a
structured document tree. Instead they are designed to act on "tag soup"
i.e. regard tags as a sort of dynamic rendering config switches.
(and that goes for Lynx too, in case some one is wondering)

All this started when Marc Andreessen and Jim Clark "hijacked" the NCSA
Mosaic team together with Lou Montulli (of Lynx fame) back in April 94,
and set out to make the first Netscape browser. They made a choice not
to go along with the already available SGML technique, but instead to
follow a "tag-soup" route that they invented them self.

As Arjun once said, "the really bad thing about this is that 'tag-soup'
was never standardized" had it been, we would all have known what to
stick to and valuable working hours could have been saved for a lot of
web authors all over the world. The words "economical efficiency" are
very real in my world at least, but I loose a lot of that on browsers
not showing a standardized behavior.

A bit later, (a single person probably) with one of the "big two"
manages to convince his marketing department that CSS might be a good
thing to adopt. Then we find that implementors runs into a hell of a lot
of problems trying to adopt the CSS spec into an already existing
"tag-souper" something that would have been a "breeze" to implement for
a rendering programmer, had he been handed a properly parsed, classified
and/or ID'ed document tree.

Instead he finds himself in a clear mess, trying his best not to break
what's already there, but still inform his rendering engine what it
ought to do with all these style rules.

So no wonder that they need some help every now and then by a closing
tag, even if the specs does not call for it. That's reality of today.

-- 
Jan Roland Eriksson

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

I can't resist posting an exchange from another newsgroup:

    [ debating with trolls ]
>>>reminds me of the Monty Python "Argument Clinic" skit.  Make a valid and
>>>factual point, and the other fellow just goes "No, it's not."
>>>After a while it becomes obvious that no intellectual ground will be
>>>gained.  The  ... troll will simply ignore the mounting evidence that
>>>proves them wrong and repeat the same old tired (and debunked)
>>>arguments.
>>>It is a waste of time to debate them....

> >No, it's not. 
> Don't give me that, you snotty-faced heap of parrot droppings! 

   Anyone who wishes to see this exchange in its original
context is invited to look in news.admin.net-abuse.email
for the thread entitled "Debating a troll is like the Monty 
Python "Argument Clinic" skit."  Thanks to James C., Will V. 
and Dave P.

-- 
  Warren Steel

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Subject: Re: Comment out SSI
   From: "Alan J. Flavell" <flavell@mail.cern.ch>

On 12 Oct 1998, cs - Elton Kong wrote:

> /*
>   <!--#exec cgi="abc.cgi" -->
>  */
> 
> Does this make sense?

(This asnwer assumes Apache.  SSI isn't well-standardised: read the 
documentation for the server that you use).

If you want to comment-out an SSI directive, simply remove the "#" - it
then becomes nothing more than an HTML comment, and will be sent out
as-is, to be ignored by the browser. 

There is (unfortunately?) no SSI comment, say <!--#comment ... -->,
that would cause the server to remove the string entirely before sending
the document to the browser.

Btw: don't use exec cgi.  Use include virtual.  Slemko explains it from
time to time, better than I could, and repeats the advice several times
a day, it seems.  It's in the Apache documentation for mod_include.


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Craig A. Keefner writes:
> Any tools/viewers available for desktop systems to simulate viewing
> webpages thru ce or palm-type device?

  Certainly; have a look at

    http://www.pilotfaq.com/sfwg_01.htm

  and go on from there. Note also the emminent group:

    comp.sys.palmtops.pilot

  where more help is probably available.


-- 
 Tina Holmboe

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Michael Fry <michael.fry@careerbuilder.com> writes:

> Anybody have thoughts on the use of pull-down lists for Web navigation?

Sure. We need an element for navigation, somewhat like the
recently-depreciated MENU.

<nav>
<item href="news/today.html">News</item>
<item href="bio.html">About Us</item>
<item href="misc/links.html">More Sites</item>
</nav>

This could be displayed as a drop-down menu in the browser's native
widget set, or like: 

[ News | About Us | More Sites ]

with each bit of text being a presented exactly like a normal
link. Which method of presentation you wanted could be specified in a
stylesheet, but users could really pick anything that suits
them. (WebTV users might prefer not having to use a mouse, for
example.)

When rendered as a drop-down menu, there are several advantages: the
user doesn't need to turn client-side scripting on if they'd rather
not, and the author doesn't have to provide a fallback CGI. That part
also means less load for the server administrator.

Note that the name itself is just regular text, so you can use an IMG
element instead (with ALT, of course). If your widget set supports
pictures in drop-downs (Macintosh, GTK, etc.), you can use it like
that. Or, you can just set it up as a sidebar "totem pole" of images
like we see on so many websites now... Stylesheets should provide for
a lot of reusability of the same NAV element, especially if you can
use OBJECT instead of IMG, as there should be some way to specify in
the stylesheet when you want the image version and when you want the
text with that. (if not, that really should be implemented.)

> * Do pull-downs run counter to the model of Web navigation that users
> have come to rely on (i.e. "To get around a Web site, I should click on
> 'buttons' and highlighted text.")? 

As implemented. If HTML had come up with a solution to this first, it
could be rendered just like a UL or anything else.


Decklin Foster

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Darren Bennett wrote:
> 
> ok...I need some quick reference on what browsers support CSS and at what
> level they support them (ie.  CSS1 or CSS2).....

Eric Meyer maintains a mastergrid of CSS support here:
http://www.webreview.com/guides/style/mastergrid.html


Sue Sims

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

In article <6vbmd1$eo9$1@nnrp1.dejanews.com>,
 <d_estiny@my-dejanews.com> wrote:
>I am doing a project for my Business Communications course at Northern
>Illinois University.  I am new to Usenet and I was wondering what some of the
>current user's biggest pet peeves in hearing from newbies. Please respond.
>Thank you.

People doing homework projects.  Thanks for asking.
--
Kate Wrightson

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

On Thu, 15 Oct 1998 23:21:03 +0100, "Derek.Moody"
<derek.moody@clara.net> wrote:

>In article <3625c19e.46327644@news.nucleus.com>, catnip
><URL:mailto:catnip8@geocities.com> wrote:
>> On Wed, 14 Oct 1998 22:28:50 +0000, Chris Jamison <chris@mu.org>
>> wrote:
>> 
>> >frames are only hated by those who lack the skill to properly use them,
>> 
>> Tell us...what are the skills that viewers need to 'properly' use
>> frames?
>
>Weeelll, he might have a point there.  Here's what I do with frames.  On
>those rare framed sites that look as though they _might_ have some
>worthwhile content I launch each frame in a new window.  Most of the time
>that is enough to break out of the frameset, then if anything is worth a
>return I just bookmark that file, ie. the frame rather than the page.

My question was sarcastic. Since this is an html authoring group, I
assume he was saying that frames are hated by authors who don't have
the skills to properly use them.

My point is never to forget who you are authoring for and how they are
affected by your choices, especially when it comes to implementing
frames.

BTW, *many* users have no idea what to do when it comes to
bookmarking a frame. I believe that it's either the author's duty to
explain the process on their framed site[1] or to simply offer a
non-framed alternative.

catnip
[1]with deference to the varying UAs.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

On Thu, 15 Oct 1998 22:05:14 GMT, catnip8@geocities.com (catnip) wrote:

> On Thu, 15 Oct 1998 20:13:15 GMT, rex@css.nu (Jan Roland Eriksson)
> wrote:
[...]
> >There's nothing ambiguous here.

> Oh, yeah...[twirls blonde hair]...nothing ambiguous there...not at
> all...[confused look]...especially that whole %flow thing which
> probably ties into the data stream thing...reminding me of watching
> the water slowly trickle down a mountainside on a hot summer day...but
> I digress...

But the thing is, catnip, it is not that difficult to understand in the
first place. All it takes is to be "kicked off" in the right direction
by some one... (and then grasp the meaning of the word "structure")

Please do me a favour and go here...

http://www-tei.uic.edu/orgs/tei/sgml/teip3sg/index.html

...it's one of the most easy to understand SGML sites you can find.
Do as I did, print it out and use it as night time reading in bed.

Your second 'Q' about %flow... Well,

first of all (%flow;)* with that 'star' at the end tells you that %flow
can appear zero or more times, that all it means.

%flow  in it's turn is of course defined in other terms, namely as...

%block _or_ %inline, which in its turn ie defined as...

for %block...
"P | %heading; | %list; | %preformatted; | DL | DIV | NOSCRIPT |
BLOCKQUOTE | FORM | HR | TABLE | FIELDSET | ADDRESS"

for %inline...
"#PCDATA | %fontstyle; | %phrase; | %special; | %formctrl

and as you can see there are more "lower level" atoms to be found in
these definitions.

Think of %something as being a "macro name" that contains a definition
of some lower level details.

Now read up on that SGML site I pointed you to.
Tomorrow you will know more than I do today...


-- 
Jan Roland Eriksson

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Steve Pugh wrote:

> HTML Validators return the following error:
> 
> document.write('</H1>');
>                  ^
> Error: end tag for element OBJECT which is not open;

Well, it's certainly an HTML syntax error, and the HTML4.0 spec
suggests a very simple remedy: stick a backslash ahead of the
forward slash.

> Does anyone know exacyly why this happens?

Yup, it's the SGML content model that the DTD specifies for SCRIPT.  I'm
not sure why it referred to OBJECT in your diagnostic, but it's
definitely a syntax error.


Alan J. Flavell

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

> 2) How do I determine the <!doctype> of my page?

See <http://www.htmlhelp.org/tools/validator/doctype.html>.  If you're
using proprietary markup like <MARQUEE>, you might want to create a custom
DTD by adapting one of the public DTDs.  There's a good explanation of how
to do this at <http://www.htmlhelp.org/tools/validator/customdtd.html>.

Darin McGrew

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

HTML terminology is as follows:

tag: the individual markup bits, enclosed in angle brackets
element: a sequence of text enclosed in a tag and end-tag
attribute: a name-value pair included in the start tag of an element
entity: a sort of macro that represents special text symbolically or
numerically.  Marked by an initial '&', and terminated by a ';'


Stanley Friesen

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

spoo <spoo@oops.ksc.nasa.gov> schrieb:
> Has anyone found how to do the equivalent of
> <IMG SRC="..." ALIGN="CENTER">
> in HTML 4.0 STRICT, with CSS ????

IMG#xyz     { left-margin: auto; right-margin: auto }

fup2

--
Claus Andre Faerber

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

(William Schlake) writes:

> I'll tell you. You want things YOUR way. If you can't have your way,
> you whine, stamp your feet and become obnoxious as you have
> demonstrated repeatedly in several newsgroups time and time again.
> 
> When I drive on the highways I see many drivers who speed, tailgate
> and drive carelessly. That doesn't cause me to stop driving, nor does
> it make me take down license numbers and write a nasty note to the
> Secretary of State's Office demanding they revoke the guy's license.
> 
> I go to a fancy restaurant I don't let the fact that I may have to
> wait to be seated and it's more crowded than I would like spoil the
> reason I went in the first place...to have an exceptionally good meal.
> 
> I pick up a magazine because I see several interesting articles I'd
> like to read so I don't get bend out of shape because in reading the
> articles I also need to flip past pages that have advertisements on
> them.
> 
> The web is a big place. Over 300,000,000 web pages out there and
> growing. If a certain site has information you want then take a deep
> breath and accept it's THEIR site to do with what they want, not some
> site for your to nitpick or whine about because they have their
> content arranged in a way you don't like, use features you don't like,
> your browser doesn't support or they used markup you don't approve of.
> 
> Accept every web site as it is. Be adult. Go visit another site if you
> don't like the one you're at. Not too much to ask of an adult, but way
> too much to ask of a little boy apparently.

Wow.

I think this posting sums up what's wrong with the WWW today. I am
someone who believes that software and electronic data should work for
people, not the other way around. When people tell me that paper is
just "nicer" than reading things on a computer, it seems so obvious to
me that the reason for this is that we've allowed data formats (such
as HTML) to do too much. We have become slaves to our tools. (How
would you like changing the prescription in your eyeglasses every time
you read a printed page from a different person?)

I know that Schlake is probably a bot, a collective of cynics, or just
mentally unstable, but you're kidding yourself if you don't think he
represents (at least in part) the attitudes of *many* people out
there. I think everyone needs to read the text I quoted very, very
carefully, and think about just where this is taking us. Following
standards is not enough. Everyone has to start empowering users, and
explaining to them that they *really do matter*. Most people who were
born long before PCs assume that *they* caused a mistake like "Error,
your browser doesn't support frobnitz." We've been engineering
computers for long enough - it's time we did something about this
simple interface assumption (We've practically dropped the ball that
TBL handed us). Computers are just tools.

[ P.S. inspect this message carefully before flaming :-) ]


Decklin Foster


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

As I said to the Glasgow taxi driver on a recent trip to the airport,
"looks as if we've had summer for this year then", to which he quipped
"it was a Thursday, I think".

Alan J. Flavell

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Graham Murray wrote:
>William  G. Schlake writes:
>> ... As far as
>> web authors using non standard markup that's their right too no matter
>> how much Alan gets his shorts in a bunch.

And, we don't even know if Alan wears underwear. Well, *I* don't know
anyway...

catnip
(Speaking of underwear, where's Chris Gray?)

---------------

> And, we don't even know if Alan wears underwear. 

Just for the record, and totally off topic, in British English
"shorts" is an outer garment (short trousers).  I suspect youall
are referring to what we call (under)pants.  Shorts would be an
appropriate garb in summer, at least for leisure wear.

As I said to the Glasgow taxi driver on a recent trip to the airport,
"looks as if we've had summer for this year then", to which he quipped
"it was a Thursday, I think". 

And we don't discuss what's worn under the kilt...

Alan J. Flavell

---------------

William G. Schlake wrote:
>But Alan, you didn't answer. The HTML world is holding it's breath and
>wants to know what the ultimate authority wears:
>
>[ ] baggy boxers
>[ ] regular briefs
>[ ] bikini briefs
>[ ] kilt with undies
>[ ] kilt without undies
>[ ] designer silk
>[ ] nothing 
>[ ] _______________________________
>
 You forgot:

[ ] jock strap
[ ] g-string
[ ] long johns
[ ] women's underwear
[ ] those underwear with Velcro that you can just rip off[1]
[ ] edible undies[1]

ummm...

catnip

[1] I *read* about those somewhere, okay? Maybe it was in the Kenneth
Starr report. [innocent look]

-------------------

> [1] I *read* about those somewhere, okay? Maybe it was in the Kenneth
> Starr report. [innocent look]

Uh... yeah... right.

Damn, once in a while I peek into a thread that Mr. :o) has taken
over,  just to see what the subject has morphed into.  What do I
find??? Underwear!  In ciwah!!!!  Geez folks... have you no
modesty?!?!?

Not that any one asked, but I prefer boxers with HTML tags on them... 


Joey M. Jackson

------------------

> Not that any one asked, but I prefer boxers with HTML tags on them...
>
  Okay, but are they *W3C compliant* tags on those boxers? Or do you wear
different boxers when using IE and when using NS?

Sorry to contribute to an already way off topic (but funny none the less)
thread.

Greg (whose boxers are w3c compliant)

--------------------

The Fischers and Their Cats wrote:
>   Okay, but are they *W3C compliant* tags on those boxers?

Certainly!  They actually are part of a set... the "HTML
Reccomendation Series."  There's one for each day of the business
week.... v1, v2, v3, v3.2, and v4.  (Of course, since there wasn't
actually a v1, those look a lot like "the emperors new clothing.")
There are also the v4 variations for weekend wear.

The boxers that use a style sheet are especially useful.  I can
coordinate thier color to match the rest of what I've got on, but if
someone else dislikes my colors, they can see what they prefer.
Either way, I still look good.

> Or do you wear
> different boxers when using IE and when using NS?

Nah, I've got briefs and thongs for those.

> Sorry to contribute to an already way off topic (but funny none the less)
> thread.

What the heck, let's see how far off-topic we can take this thread.
We may as well have fun while there's no moderator to harrass us.  :-)


Joey M. Jackson

------------------------

>Not that any one asked, but I prefer boxers with HTML tags on them...

So the <P> would be on the front, then?


8^)
Dan McGarry

-----------------------

> So the <P> would be on the front, then?

No, that would be the <BIG> tag.  Using a proper XML vocabulary rather 
than HTML would offer more options.


Eric Bohlman

----------------------

> No, that would be the <BIG> tag.

Bragger.  ;-)  That's more information than I needed to have.  :-)

> Using a proper XML vocabulary rather 
> than HTML would offer more options.

Well, sure... but if you're going to be using XML, you may as well get
the XML boxers since they're quite different from the HTML boxers


Joey M. Jackson

-----------------------

> Bragger.  ;-)  That's more information than I needed to have.  :-)

That's okay.  Just use lynx and you'll never notice the difference.

BTW, are these HTML shorts fully accessible?  I mean do they have
raised letters or braille bumps?

Jude Crouch

----------------------

>BTW, are these HTML shorts fully accessible?  I mean do they have
>raised letters or braille bumps?

Oh...is that what those things are called? Braille bumps? [jots that
down]

catnip

----------------------

> So the <P> would be on the front, then?

Heh, heh, heh... but I haven't decided whether to include the </P>
yet.

Or...

<H1></H1> across the front and 
<DIV ALIGN=CENTER DIRECTON=
VERTICAL></DIV>  in the back???


Joey M. Jackson

---------------------

>Heh, heh, heh... but I haven't decided whether to include the </P>
>yet.

That would be on the urinal.

>Or...
>
><H1></H1> across the front 

LOL!!

>and 
><DIV ALIGN=CENTER DIRECTON=
>VERTICAL></DIV>  in the back???

Just use ascii art: (_|_)

catnip

-------------------

> Just use ascii art: (_|_)

Ah yes... I forgot about "butticons"


Joey M. Jackson

------------------

>Just for the record, and totally off topic, in British English
>"shorts" is an outer garment (short trousers).  I suspect youall
>are referring to what we call (under)pants.  Shorts would be an
>appropriate garb in summer, at least for leisure wear.

I'm a member of the great and mighty commonwealth too, sugar. :-) We
use both versions of 'shorts' over here. "Shorts tied in a bunch/knot"
refers to underwear.

>As I said to the Glasgow taxi driver on a recent trip to the airport,
>"looks as if we've had summer for this year then", to which he quipped
>"it was a Thursday, I think". 
>
>And we don't discuss what's worn under the kilt...

We could... :-)

>Well, it's no more off-topic than much of the stuff on this group
>recently, bleagh.

Well, Alan, you effectively dodged the whole shorts issue anyway. :-)

catnip

------------------

>As I said to the Glasgow taxi driver on a recent trip to the airport,
>"looks as if we've had summer for this year then", to which he quipped
>"it was a Thursday, I think".


Dunno, I was busy sunning myself in Zanzibar. And still I managed to find a
computer and fix it!

>And we don't discuss what's worn under the kilt...


Nothing is worn under the kilt. Its all in very, very good condition.


Rhialto

----------------

>And, we don't even know if Alan wears underwear. Well, *I* don't know
>anyway...

So he hasn't called your 1-900 number, then?


8^)
Dan McGarry

----------------

>So he hasn't called your 1-900 number, then?

I did suspect it might be him on the line one nite...

Remember, the lines are always open. Just call 1-900-LUV-HTML for some
stimulating html discussion. ;-)

catnip

-----------------

> I did suspect it might be him on the line one nite...

Really?  What made you think it was him?


Joey M. Jackson

-----------------

>Really?  What made you think it was him?

He wouldn't tell me if he had any underwear on. Not only that, he
almost choked himself when he was loosening his Pouty Purist(tm) tie.

catnip
(Luv ya Alan! *smooches* Call again soon.)

-----------------

> He wouldn't tell me if he had any underwear on. Not only that, he
> almost choked himself when he was loosening his Pouty Purist(tm) tie.

Hmmm, I dunno... nothing there would rule out the possiblity that it
was your buddy, Mr. Troll.   It's pretty obvious from his posts that
he's infatuated  with you.

Did he actually *say* it was a Pouty Purist tie?  We, of course,  know
that Mr. Troll is a closet  Purist, so he *could* have been wearing a
Pouty Purist tie.)  Were there any other clues?

Joey M. Jackson

-----------------

(catnip)
wrote:

> Remember, the lines are always open. Just call 1-900-LUV-HTML for some
> stimulating html discussion. ;-)

While we're *way* off topic and there's no moderator around....

The following is modified from a bit of junk mail that I got a long
time ago.  It needs more work, but I need some sleep.  :-)



Caller:  Hello, what do you look like?

LUV-HTML:  I am wearing a sexy outfit made by W3C, over a pair of silk
HTML boxer shorts. I work out every day, My <BODY> is toned and
perfect. My <FRAMESET ROWS="36, 24, 36">measurements are
36-24-36</FRAMESET>. What do you look like?

Caller:  I'm 6'3" and about 250 pounds.   I wear glasses and I have on
a pair of grey sweat pants.  I'm also wearing a FrontPage T-shirt with
a few spots of barbecue sauce on it from dinner...it smells funny.

LUV-HTML:  I want you.  Would you like to <LINK> me?

Caller:  Uh... OK

LUV-HTML:  We're in my bedroom.  There's soft music playing on the
stereo <BGSOUND SRC="bolero.wav" LOOP=INFINITE>
and candles on my dresser and night table.  I'm looking up into your
eyes, smiling. My hand works its way down your <BODY> and begins to
fondle your huge, swelling <P>.

Caller:  I'm gulping, I'm beginning to sweat.

LUV-HTML:  I'm pulling up your shirt and kissing your chest.

Caller:  Now I'm unbuttoning your blouse. My hands are trembling.

LUV-HTML:  <A HREF="softmoan.wav">Mmmmm</A>

Caller:  I'm taking hold of your blouse and sliding it off slowly.

LUV-HTML:  I'm throwing my <HEAD> back in pleasure.  The cool silk
slides off my warm skin.  I'm rubbing your <P> faster, pulling and
rubbing.

Caller:  My hand suddenly jerks spastically and accidentally rips a
<BR> in your blouse.  I'm sorry.

LUV-HTML:  That's OK, it wasn't really too expensive.

Caller:  I'll pay for it.

LUV-HTML:  Don't worry about it.  I'm wearing a lacy black bra.  My
soft <OBJECT>'s are rising and falling, as I breath harder and harder.

Caller:  I'm fumbling with the clasp on your bra.  I think it's stuck.
Do you have any scissors?

LUV-HTML:  I take your hand and kiss it softly.  I'm reaching back
undoing the clasp. The bra slides off my body. The air caresses my
<OBJECT>'s.

Caller:  How did you do that? I'm picking up the bra and inspecting
the clasp.

LUV-HTML:  I'm arching my back. Oh baby. I just want to feel your
tongue all over me.

Caller:  I'm dropping the bra. Now I'm licking your, you know,
<OBJECT>'s. They're neat!

LUV-HTML:  I'm running my fingers through your hair. Now I'm nibbling
your ear.

Caller:  I suddenly <A HREF="aaaachoo.wav">sneeze</A>. Your <OBJECT>'s
are covered with spit and phlegm.

LUV-HTML:  What?

Caller:  I'm so sorry. Really.

LUV-HTML:  I'm wiping your phlegm off my <OBJECT>'s with the remains
of my blouse.

Caller:  I'm taking the sopping wet blouse from you. I drop it with a
<A HREF="kerplop.wav">plop</A>.

LUV-HTML:  OK. I'm pulling your sweat pants down and rubbing your
<P ALIGN=CENTER>.

Caller:  I'm screaming like a little girl! Your hands are cold! Yeeee!

LUV-HTML:  I'm pulling up my miniskirt. Take off my boxers.

Caller:  I'm pulling off your boxers. My tongue is going all over,
in and out nibbling on you...umm... wait a minute.

LUV-HTML:  What's the matter?

Caller:  I've got a <SCRIPT> caught in my throat. I'm choking.

LUV-HTML:  Are you OK?

Caller:  I'm having a coughing fit. I'm turning all "#FF0000".

LUV-HTML:  Can I help?

Caller:  I'm running to the kitchen, choking wildly. I'm fumbling
through the cabinets, looking for a cup. Where do you keep your
cups?

LUV-HTML:  In the cabinet to the right of the sink.

Caller:  I'm drinking a cup of water. There, that's better.

LUV-HTML:  Come back to me, lover.

Caller:  I'm washing the cup now.

LUV-HTML:  I'm on the bed aching for you.

Caller:  I'm drying the cup.. Now I'm putting it back in the cabinet.
And now I'm walking back to the bedroom. Wait, it's dark, I'm lost.
Where's the bedroom?

LUV-HTML:  Last door on the left at the end of the hall.

Caller:  I found it.

LUV-HTML:  I'm tuggin' off your pants. I'm moaning. I want you so
badly.

Caller:  Me too.

LUV-HTML:  Your pants are off. I kiss you passionately-our naked
<BODY>'s pressing each other.

Caller:  Your face is pushing my glasses into my face. It hurts.

LUV-HTML Why don't you take off your glasses?

Caller:  OK, but I can't see very well without them. I place the
glasses on the night table.

LUV-HTML:  I'm bending over the bed. Give it to me, baby!

Caller:  I have to pee. I'm fumbling my way blindly across the room
and toward the bathroom.

LUV-HTML:  Hurry back, lover.

Caller:  I find the bathroom and it's dark. I'm feeling around for
the toilet. I lift the lid.

LUV-HTML:  I'm waiting eagerly for your return.

Caller:  I'm done going. I'm feeling around for the flush handle,
but I can't find it. Uh-oh!

LUV-HTML:  What's the matter now?

Caller:  I've realized that I've peed into your laundry hamper. Sorry
again. I'm walking back to the bedroom now, blindly feeling my way.

LUV-HTML:  Mmm, yes. Come on.

Caller:  OK, now I'm going to put my...you know ...<P>...in
your...you know...woman's <P>.

LUV-HTML:  Yes! Do it, baby! Do it!

Caller:  I'm touching your smooth butt. It feels so nice. I kiss
your neck. Umm, I'm having a little trouble here.

LUV-HTML:  I'm moving my ass back and forth,
<A HREF="loudmoan.wav">moaning</A>. I can't stand
it another second! Slide in!  <LINK> me now!

Caller:  <FONT SIZE="-1">I'm flaccid</FONT>.

LUV-HTML:  What?

Caller:  I'm limp. I can't sustain an erection.

LUV-HTML:  I'm standing up and turning around; an incredulous look on
my face.

Caller:  I'm shrugging with a sad look on my face, my <P
CLASS=FLACCID>.  I'm going to get my glasses and see what's wrong.

LUV-HTML:  No, never mind. I'm getting dressed. I'm putting on my
underwear. Now I'm putting on my wet, nasty blouse.

Caller:  No wait! Now I'm squinting, trying to find the night table.
I'm feeling along the dresser, knocking over cans of hair spray,
<IMG>'s and your candles.

LUV-HTML:  I'm buttoning my blouse. Now I'm putting on my shoes.

Caller:  I've found my glasses. I'm putting them on. My God! One of
your candles fell on the curtain. The curtain is on fire! I'm pointing
at it, a shocked look on my face.

LUV-HTML:  Go to hell. I'm hanging up, you loser!

Caller:  Now the carpet is on fire! Oh noooo!

LUV-HTML:  <A HREF="hang-up.wav">click</A>




Joey M. Jackson

-----------------------

>LUV-HTML:  I want you.  Would you like to <LINK> me?
>
>Caller:  Uh... OK
<snip>

Joey! <hands_on_hips> I didn't know you were taping our phone calls!

catnip

-----------------------

> Joey! <hands_on_hips> I didn't know you were taping our phone calls!

Pshaw...  as if I can afford to call a Canuck 1-900" number!!

Joey M. Jackson

-----------------------

> BTW, are these HTML shorts fully accessible?  I mean do they have
> raised letters or braille bumps?

Well, they do have a "fly" so I guess they're "accessible."  I won't
make any comment on "raised" or "bumps" in this context.  ;-)


Joey M. Jackson

----------------------

> Oh...is that what those things are called? Braille bumps? [jots that
> down]

Well, actually, no.  They are actually called "raised dots" and the
process for printing a braille page is called "embossing".

BTW, there is a whole section about Braille and SGML at
        <http://www.duxburysystems.com/js-sgml.html>,
as well as other interesting links all around duxburysystems.com .
Braille is not translated word-for-word, and uses a system of
contractions to include more content on each page.  Most of the
specs for Braille in its various forms reads much like HTML, using
block elements to indicate the various functions of the sections of
a particular document.  (Headings, lists, paragraphs).


Okay, whew, we got back to HTML.


Jude Crouch

----------------------

> Okay, whew, we got back to HTML.

Aw shucks, and this was just starting to get interesting. ;-)


Joey M. Jackson

-----------------------

>Well, sure... but if you're going to be using XML, you may as well get
>the XML boxers since they're quite different from the HTML boxers

I ordered some XML boxers through the mail, and all I got was a bolt of
cloth and a beginner's sewing book.

<sigh>
Dan McGarry

---------------------

>(Luv ya Alan! *smooches* Call again soon.)

Ah, I see his secret's safe with you....

8^)
Dan McGarry


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Andrew McCormick uttered:
>
>Netscape and IE's inability to move beyond the hack of a tag-soup "parser"
>is probably the biggest problem.  The fact that neither really care about
>helping people get work done is another.

IE has actually gotten far better, whereas Netscape...

<P>This </sub>goes </sup>up </sup>and </sub>down in bad browsers</p>


Tor Iver Wilhelmsen

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

William G. Schlake wrote in message <363e5ab1.47452258@news.enteract.com>...
>On Thu, 15 Oct 1998 00:05:23 GMT, catnip8@geocities.com (catnip)
>wrote:
>
>>Because the viewers *do* have final control. I can control the
>>resolution, colours, link appearance, images etc. no matter what the
>>owner intended to force on me in the first place.
>
>In reality you have no control at all.

And here lies the crux of your, and many other's miss-understanding. The
web, as a medium, is perfectly suited, in fact designed, to give users a
moderate amount of control over the content they wish or need to see. How
much control, well that's up to the skills of the author. The more skilled
author will provide more views and tangents into the material than a lesser
skilled one.

As I have said before, perhaps the most unfortunately fall-out of the
"Johnny-come-late-to-the-web-parade" is that they missed the Web 101 class
on "hyper-media." Instead, they see "web-pages" as nothing more than
electronic magazines or still-frame TVs (and yes, its about as exciting as
it sounds). I am often asked what my favorite web-site is; my response is
always the state DOT traffic flow website which shows where the heaviest
traffic is, and how to avoid it. I can view this site on WebTV, my Sparc, my
PC, and my Palm Pilot III through a cell phone in the car. I think this is
perfect example of a great site because there is no other medium that could
(or would) give me this type of information on demand, and customized to my
needs. This is what the web is about.

Now, if you're just in it to make money, ignore that last paragraph, learn
all the cute tricks and whiz-bang effects you can, cause there is a
gold-mine of idiots out there with lots of money. And, really, best wishes.
But don't try to tell others that you really understand what the medium is
about, or know what is good for it. If its good for you: congratulations,
but don't pretend for a minute to really understand it.



REBUS

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Adrian writes:

> The web is a place made buy the people, for the people. It is a
> space for information, business, and entertainment.

This is exactly my reason why the web should work for people instead
of people working for it. This space, as you call it, ceases to be
such a great thing when we spend more time making it work. When was
the last time you had to make a document printed on paper "work"?

> Just because you don't like what someone does with a website doesn't
> mean that they should change it just for you,

This is the Shlakeism that upsets me the most. Web designers seem to
think that simply accommodating people is a "change," and a burden for
them. The fundamental belief within this is that people have to do
work and kludge around things just to let people see stuff - whether
this refers to Lynx users looking at text or Netscape users looking at
<layer>s.

It doesn't have to work like this. Writing accessible pages means that
no one has to go to some insane "fixing it" length just to read
something. But the greater evil in acting like Schlake is teaching
naive users that they should expect having to upgrade software and
muck around with their system to get something that isn't much of an
improvement.

Luckily, a bit of a backlash against upgrade cycles may be
starting. Read <URL:http://www.useit.com/alertbox/980322.html>.

> or that you should just live with it and screw off. If you don't
> like something about a site, don't whine and bitch and complain, but
> think rather about how it sits.

No one is going to do this willingly. The web is not a personal forum
for <insert name of page developer>. This is the kind of childish
acting that makes me believe the web sucks. If the web is useless to
everyone except those on an expensive computer, it's not a global
communications medium. It's a playground for these people and no one
else.

> We are not becoming slaves to our tools. We are just expecting too
> much of them. You want your computer to do everything right now, and
> do it just how you want it to.

To hand-wave anyone else's desire for this is selfish. The most
important people are the ones who can't use the web now, because they
cant hack around with a web browser to fix *your* problems. Computers
are our tools. There is nothing hard-coded in them that prevents all
of the WWW from being readable by anyone. It's only our prejudices.

> But you are pissed because computer errors don't fix themselves and
> you have to actually fix them, becoming a slave to your computer?
> How lazy!

This expectation that errors are a fact of life on the WWW is
sick. Just because *you* have the desire to fix them doesn't mean
anything about people who just want information, not a meaningless
toy. The possibility of the web someday actually being better than
paper, and becoming accessible enough to start embedding in better
appliances than computers is more important than your indignation
about user X who doesn't have browser Y, or just sets it up so he can
read things more easily for him.

> Netscape is only a few million lines of code. Go ahead. Make it fix
> itself. Write an AI to fix all the bugs itself. The design a new
> computer with enough power to run it. No biggie. Its what you expect
> of others, so why don't you expect it of yourself?

See, your basic assumption is that people can actually go and fix
things. Ask yourself why there is any fixing to do instead. Does it
make the web more equitable to claim everything revolves around you?
You simply cannot claim something as hard as rewriting mozilla -
actually *doing* something - is the same as asking an author *not* to
do something.

> As for the point of view that something is wholly controlled by the
> person who owns it, well, that isn't pathetic. Who else would
> control it? The viewers?

Well, duh. *Information* can be presented in some way adapted to the
actual display medium. My problem with the web is it's not even close
to that state.

> Why? Do you control what's on TV?

This is why I don't watch TV. There's no potential to make it more
convenient, because it's based on the system of sitting there and
watching. Sitting is unnecessary if I just want information and I
could get my <insert name of appliance> to speak it to me when my
hands are full, or get a useful summary in my email-box.

> But in the end, who decides what stays and what goes? Gues who. The
> owner.

Throughout this idea I have had the suspicion you are talking about
content. I am not. I am talking about how you *make* this content
deliverable, with no hassle for people who want to use it and not
waste time with their software configuration when they have more
important things in their life. A little work from us on that front
now could save the annoyance of millions of people later.

> Whatever you can do with your site is what the web can do.

But can the web just simply be used? Is there no problem in
implementing it anywhere, in whatever way is convenient to the reader?
Or do we have to continually struggle with software? That's why I
posted.


Decklin Foster

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
William  G. Schlake <comments@hobsonsquare.com> wrote:
>You sure you're not a closet purist? 

Well, actually I think HTML sucks rocks.  I'm very unenthused with any
language that nobody seems to be able to write an interpreter for that
doesn't blow up all the time.  The two most popular browsers are both
pieces of crap, I can't browse anywhere without either of them
trapping within a few minutes.  The only browser that is stable does
it by throwing out all the flash and just displaying text.

Jay Denebeim

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

On Fri, 16 Oct 1998 18:25:20 -0400, bill@scconsult.com (Bill Cole)
wrote:

>In article <708fiv$jti$1@Mercury.mcs.net>, jcrouch@MCS.COM (Jude Crouch) wrote:
>
>>Oh, and in answer to your question... SEX.  Keeps `em from going to the 
>>bathroom every time.
>
>Not enough. If they really have to go bad enough, no amount of sex works. 
>
>I think HTML 5 needs a <USERRESTRAINT> tag.

I'd vote for <USERRESTRAINT CLASS=leather>

catnip

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Content should be targeted. Accessibility to that content should be
universal.


Dan McGarry

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

On Fri, 16 Oct 1998, Joi Ellis wrote:

> SSI pages are certainly cacheable. 

I'd just like to expand a little on my previous answer.  The cache
proxy that's being used for this demonstration is squid 1.1.22.

My own pages use SSI extensively, with the xbithack full option being
used.  For example, here's the headers for my welcome page:

> lynx -dump -head http://ppewww.ph.gla.ac.uk/~flavell/
HTTP/1.1 200 OK
Date: Sat, 17 Oct 1998 14:03:59 GMT
Server: Apache/1.3.1 (Unix)
Last-Modified: Sun, 17 May 1998 16:02:00 GMT
Connection: close
Content-Type: text/html;charset=iso-8859-1

which includes the last-modified date/time of the actual file
(irrespective of its dynamic inserts).  This is entirely appropriate
for boilerplate inserts (pages that contain real dynamic content 
need to be handled differently, of course).

So, here's what the cache proxy records about my welcome page, which I
had referenced just half a minute before:

170.http://ppewww.ph.gla.ac.uk/~flavell/
     Verified 35s ago, Used 35s ago, Modified 5M ago, 

See, it was modified 5 months ago.  When a reader (in this case, me) 
accessed it, squid would have done an if-modified-since request from the
"distant"  server, and the server reported that the page had not been
updated, so there was no need to fetch it again.  Working As Designed. 

Now, let me reference it again, and see what the new cache server
report says for this object:

170.http://ppewww.ph.gla.ac.uk/~flavell/
     Verified 9m ago, Used 9s ago, Modified 5M ago,

This time, the cache proxy's algorithm decided that, after a mere 9
minutes, there was no need to even re-verify the page.  Fine.

> I use SSI extensively for
> boilerplate paragraphs (common headings and footings, etc) and those
> pages certainly do cache on the client side.

> gyles19@nospam.visi.com, http://www.visi.com/~gyles19/

OK, here's your headers:

HTTP/1.1 200 OK
Date: Sat, 17 Oct 1998 14:19:34 GMT
Server: Apache/1.3.1 (Unix)
Connection: close
Content-Type: text/html

See, there's no last-modified date/time in there.

Here's what the cache says, directly after I've accessed the document:

745.http://www.visi.com/~gyles19/
     Verified 12s ago, Used 12s ago, Modified 12s ago, 

It thinks the document was _modified_, so it was sent a fresh copy.
And because it was modified so recently, if I try to access it again,
the same thing will happen:

744.http://www.visi.com/~gyles19/
     Verified 11s ago, Used 11s ago, Modified 11s ago, 

_This_ is the problem with SSI.  It's not a feature of you actively
inserting any kind of no-cache headers, as you were talking about
earlier; it's the absence of a meaningful last-modified date/time.

And indeed that's how it _should_ be, for SSI pages that are genuinely
dynamic.  But when it's being done merely to insert boilerplate (and
non-critical as to whether updates to the boilerplate will get
immediately reflected in the served documents), then the converse seems
to apply, and the use of xbithack full with the appropriate x-bit
settings is strongly recommended. 

  http://www.apache.org/docs/mod/mod_include.html#xbithack

Now, I reconfigure my browser to turn the cache server off, and set
the browser's options to "compare with the document on the network every
time".  And sure enough, every time I go to your page, the browser
reports that it's fetching the document from the distant server.
And the browser's "page info" display tells me that the last-modified
date/time of the document is "unknown".  It all hangs together.

With my own pages, in this situation the browser would still check
(I've told it to, after all), but the server would return the status
304 not-modified response, instead of sending an apparently fresh
version of the document.

(Please excuse me using your page as fodder for this demonstration.  I
don't for a moment want to give the impression that I'm targetting an
attack specifically on your pages, but they served as handy examples to
demonstrate this point.  And the same thing will happen with any of my
own pages where I've failed to set the x-bits correctly, i.e have set
the user x-bit to enable SSI but haven't set the group x-bit to.) 

-- 

                  'we use the term "word" here to mean "sequences 
                   of non-white space characters"' - HTML4.0 spec.


Alan J. Flavell

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

There is a NG dedicated to site checks. Check out alt.html.critique.


Greg Gibson

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

> > Is there a program avaliable to encrypt html pages? ...which is freeware
> > or shareware? How do these work and will they work with most browsers
> > (old and newer versions of NS and IE?)?
> > 
> 
> Yes, they are called "wysiwyg html editors". No, the code they produce
> may not work for most browsers.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Timothy R Prodin wrote:
>Actually, I am a lot more interested in gauging the interest
>as opposed to gaague interest....

This is what is wrong with c.i.w.a.h.

You can't force civility on the web; you can only suggest it.  It's up to the
individual UA (Usenet A.....) to render a post as civil as , never mind.

ScottM6146

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Michael Fry wrote:
>* Would the addition of pull-downs undermine the current navigation
>scheme or confuse users trying to make sense of the site's overall
>architecture?

I wonder how many visitors to a site actually need to acquire such a
mental image in order to find what they need. My objective is always to
try to allow any visitor to find their *particular* item of interest as
quickly as possible. I think that the only case where the visitor
*needs* to get a site overview is if an initial attempt to find that
item of interest fails, yet there remains in the visitors mind a belief
that it's on the site _somewhere_ (in which case, the site has failed,
IMO).

Colin F Reynolds

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

If you are not already familiar with Usenet, the place to start is
news.announce.newusers.  If you are interested in hearing about every new
newsgroup which is proposed, read news.announce.newgroups.  If you want to
read information about existing groups, read news.groups.reviews.  If you
have questions about where to talk about various topics or what groups are
available, read news.groups.questions. 

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

"Dan McGarry" wrote:
><colin2177@my-dejanews.com> wrote:
>[snip]
>
>>But the fact is, that's exactly what happens. So our intentions really don't
>>enter into it. If we understand and embrace the "web way" we get a whole lot
>>of benefits (universal accessibility) for little or no cost to us. We don't
>>have to change what our gui audience gets, and we get this crazy flexibility
>>as well. *Man*, we live in exciting times.
>
>Very Taoist. Think I'll publish a little chapbook entitled _Tao Te Web_.
>
>(Don't get any ideas, catnip; I thought of it first!)

Hmmm...you know me better than I thought you did. ;-)

catnip


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Democracy is three wolves and a sheep voting on what's for dinner.  You don't
want democracy, you want *fairness*.  These are not always compatible.

Peter Seebach

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

catnip <catnip8@geocities.com> wrote:

>That's just a fantasy consensus. You've heard of lurkers?

Sure, I've heard of lurkers:

(To the tune of "My Bonny lies over the ocean")
by Jo Walton

The Lurkers support me in email
They all think I'm great don't you know.
You posters just don't understand me
But soon you will reap what you sow.

Lurkers, lurkers, lurkers support me, you'll see, you'll see
off in e-mail the lurkers support me, you'll see.

The lurkers support me in email
"So why don't they post?" you all cry
They're scared of your hostile intentions
they just can't be as brave as I.

Lurkers etc.

One day I'll round up all my lurkers
we'll have a newsgroup of our own
without all this flak from you morons
my lurkers will post round my throne.

Lurkers etc.
-- 
Jay Denebeim

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

William  G. Schlake wrote:
>When is the last time you sat through a TV show
>that scrolled text for 30 minutes?

Never.  On the other hand, I typically spend well under two hours a week
watching television, and at least five or ten hours a week reading Usenet.

You certainly seem to find Usenet interesting enough.

>I doubt anyone would for more than
>five minutes. People have been conditioned to expect more than text
>when viewing a web page.

Whenever people talk about "people have been conditioned", I have these
horrible images of a training session for newbies where the instructor
gives them all breadcrumbs every time an image loads.


Peter Seebach

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

> > > ...the person who came up with the idea to moderate ciwah did so 
> > > out of frustration in attempting to deal with Schlake's trolling 
> > > after he saw that enteract.com would not handle the situation.)
> >
> > It should be obvious that this is not a good reason to moderate a
> > group.
> 
> It is not obvious to me.

I have to plead guilty to attempting to argue by assertion. Here are 
some reasons:

1. Frustration is not always a good motivator.
2. A way of dealing with trolling 
      <http://www.intersurf.com/~aevinc/aev2trol.htm>
3. Another comment on trolling
      <http://www.lava.net/~dewilson/asd/trolls.html>
4. You might also be interested in this essay
      <http://www.paranoia.com/~ricardo/faq.html>
5. In this article <http://www.swcp.com/~dmckeon/mod-faq.html>,
two particular reasons for moderation are discussed:
     i) To decrease the volume of off-topic posts, and
    ii) To increase the significance of of content of on-topic points.
6.  This reference gives the clearest guide known to me on how to
deal and live with rogues
      <http://www.cyberramp.net/~kjpodsia/rebuttal.html>

Ben
cuthbert@no.spam.leeds.ac.uk

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Q: How many UseNet junkies does it take
to change a light bulb?

A: 1,331:

1 to change the light bulb and to post that the light bulb has been changed

14 to share similar experiences of changing light bulbs and
how the light bulb could have been changed differently.

7 to caution about the dangers of changing light bulbs.

27 to point out spelling/grammar errors in posts about
changing light bulbs.

53 to flame the spell checkers

156 to write to IP administrators complaining about
the light bulb discussion and its inappropriateness to this
group.

41 to correct spelling in the spelling/grammar flames.

109 to post that this group is not about light bulbs and to
please take this discussion to alt.lite.bulb

203 to demand that cross posting to alt.grammar,
alt.spelling and alt.punctuation about changing light bulbs
be stopped.

111 to defend the posting to this group saying that we are
all use light bulbs and therefore the posts **are**
relevant to this mail list.

306 to debate which method of changing light bulbs is
superior, where to buy the best light bulbs, what brand of
light bulbs work best for this technique, and what brands
are faulty.

27 to post URLs where one can see examples of different
light bulbs

14 to post that the URLs were posted incorrectly, and to
post corrected URLs.

3 to post about links they found from the URLs that are
relevant to this group which makes light bulbs relevant to
this group.

33 to quote all posts to date, including all headers and
footers, and then add "Me Too."

12 to post to the group that they are leaving because
they cannot handle the light bulb controversey.

19 to quote the "Me Too's" to say, "Me Three."

4 to suggest that posters request the light bulb FAQ.

1 to propose new alt.change.lite.bulb newsgroup.

47 to say this is just what alt.physic.cold_fusion was
meant for, leave it here.

143 votes for alt.lite.bulb.

=====================
thanx Dragula

------------------------------------

-iSs- wrote:
> 
> Q: How many UseNet junkies does it take
> to change a light bulb?
...
> 111 to defend the posting to this group saying that we are
> all use light bulbs and therefore the posts **are**
> relevant to this mail list.

You forgot that at least one would point out that this is *not* a mail
list ;-) That, of course, could trigger a rudeness rebuttal, spinning
off ad infinitum into a sub-thread with a life of its own...
-- 
Sue Sims

---------------------------------

You forgot:

134 posts congratulating the posting on their astute sense of truth and
humor.

and

256 posts saying that these congratulating posts are off topic and should be
banned.

and

1098 posts saying that the root cause of all this is pointy-headed purists.


REBUS

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

IDIOT, n. 
            A member of a large and powerful tribe whose
            influence in human affairs has always been dominant
            and controlling. The Idiot's activity is not confined to
            any special field of thought or action, but "pervades
            and regulates the whole." He has the last word in
            everything; his decision is unappealable. He sets the
            fashions and opinion of taste, dictates the limitations
            of speech and circumscribes conduct with a
            dead-line. 
                                          
            Ambrose Bierce

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

[ STAGO = '<'  = Start TAG Open ]
[ ETAGO = '</' = End TAG Open   ]
[  TAGC = '>'  = TAG Close      ]

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Abigail wrote:
> catnip wrote:
> > Jan Roland Eriksson wrote:
> > > Richard Noteboom wrote:
> > > > Jan Roland Eriksson wrote:
> > > > [ 8< ]
> > > > 
> > > > resulting in 
> > > > <BODY>
> > > >   <UL>
> > > >   <LI><UL>
> > > >           <LI>Text</LI>
> > > >   </UL></LI>
> > > >   </UL>
> > > > </BODY>
> > >
> > >Which happens to be fully valid HTML :-)
> > 
> > What does this look like when rendered? Double indented "text"?
>
> Maybe on your browser. But with one of my stylesheets, I
> get chocolate flavoured text.

I shall implement that on every page of mine promptly. ;-)

catnip

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

              "Adding ALT to images doesn't make the images disappear."
                                                        - Reinier Post

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Kurt J. Lanza wrote:
>Sergio Marchesini <nospampimaggio@freenet.hut.fi> wrote:
>>[ 8< mercy snip 8< ]
>>
>Since you obviously believe in mind-readers (that's how we know what
>OS you are running on), why bother to post a question? Just read the
>mind of someone who knows the answer ;-)

I've found it a safe assumption, that when a person neglects to inform the
rest of the world of the OS, they are using Windows (and typically the
latest version.) The corollary is that, when the make a an unqualified
statement about the state of the Internet/Web/World, they are from the USA.

Therefore, I'd recommend TextPad.


REBUS

------------------------

>        The corollary is that, when the make a an unqualified
> statement about the state of the Internet/Web/World, they are from the USA.

Even with:
 (1)    a Finnish email address (albeit one known to be used by many
        international users).
 (2)    An Italian "organization" line (which says to me he's posting through
        a Telecom Italia newsserver) and italian name.
 (3)    Second-language English: although his spelling and grammar are rather
        better than many americans, he's clearly not at ease with the language.

To answer the original question: use Perl.  Hai bisogno solo di una linea
di Perl - se fai la ricerca su Dejanews, troverai facilmente la soluzione
su questo Newsgroup.

-- 
Nick Kew

------------------------

Corollary: a proposition that follows with little or no proof from a
previously proven statement.

(Doesn't mean that the corollary applies to the poster. Instead, since the
poster does use Windows, the previous proof, the second assumption is also
true.)

REBUS

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Ed Valle's sig:


--
"You remain, Sir, our humble and obedient servant.  Have a banana."
 
                                                               -- Alan
Flavell


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

(S.E. Morris) writes:
>I got the impression that Netscape have big plans for XML

Be afraid, very afraid.


Tero Paananen

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

"Alan J. Flavell"
<flavell@mail.cern.ch> wrote:

>Until you can suggest this kind of layout successfully with CSS on most
>of the browsers that people are using, you might just want to consider
>using the HTML3.2 <DL COMPACT> for this task, with your code numbers as
>the <DT> and the hymn details as the <DD>. This appears to be supported
>by NS4 and MSIE4, to name just two, and on browsers that don't support
>the COMPACT attribute, the fallback is harmless enough. 

(I didn't know they supported the COMPACT attribute. I had thought
nobody was going to support or use it so I never tested on the 4
versions...)

If I understand you correctly, you're suggesting the use of DL instead
of TABLE to overcome browser bugs which cause problems with links to
locations inside tables. Although the DL element is somewhat vague and
often abused (although not as much abused as TABLE!), I'd say it still
means definition list, and by using it you need to accept the
consequences of some software really treating it according to that.
Sometimes there's no strict border between description lists and
definition lists, but here the DT elements would be just code numbers,
not something you're defining or even describing.

For instance, a search engine might conceivably assume that DT
contains a term being defined and return the document when someone is
searching for _definitions_ for the keywords he has given. (In the
general case, although probably not in this one, this might lead to
strange results.) A user stylesheet might legitimately aim at
presenting DT elements very prominently. A speech synthesizer might
conceivably read the element so that the structure is explicit (e.g.:
"Definition of - -".) And so on. I'm not saying there's substantial
support to DL at present in those senses; I'm saying that perhaps we
should sometimes give more weight to make room for future advances in
user agents than to dealing with buggy browsers in a manner which
sacrifices structure.

Wouldn't a simple paragraph structure, with code number in STRONG
perhaps, be more suitable? Granted, there would be the problem that
without a suitable style sheet in effect the presentation would be
less pleasant - but it would still work, with its links. And if a
browser gets wild with links to locations within a (large) table, it
(or some other browser) might as well get confused links to location
withing a (large) definition list.

Well, perhaps there is something that might help in removing the link
problems (at least partially) without changing the overall structure.
What I mean is that the pages contain markup like
<TD ...><A NAME="M1711"></A>1711.</TD>
This means that the content of the A element is an empty string. This
is valid but it is known to cause problems to some browsers, so
usually one should put something inside the A element - and usually
there is some natural choice, in this case
<TD ...><A NAME="M1711">1711.</A></TD>

Jukka Korpela

-----------------------

Jukka Korpela wrote:

> (I didn't know they supported the COMPACT attribute. I had thought
> nobody was going to support or use it so I never tested on the 4
> versions...)

To be honest with you, neither did I, until I tried it out for the
purpose of this answer.   The Tweedles move in mysterious ways...

> If I understand you correctly, you're suggesting the use of DL instead
> of TABLE to overcome browser bugs which cause problems with links to
> locations inside tables. 

Hey, that's a bit unfair.  I'm suggesting the use of DL for associating
terms (code numbers) and their corresponding descriptions (the hymn
details).  The fact that TABLE had been used in the examples was
incidental, IMO.  However, the trolls will be highly amused to see two
suspected purists fighting over this issue ;-)

> Although the DL element is somewhat vague and
> often abused (although not as much abused as TABLE!), I'd say it still
> means definition list,

Well, there goes my "purist" star...

> Sometimes there's no strict border between description lists and
> definition lists, but here the DT elements would be just code numbers,
> not something you're defining or even describing.

It does in a sense "define" the code numbers, but I don't propose to get
into an argument over this.  You've expressed your position clearly, and
you've every right to express it, just like anyone else. 

[... well presented argument elided...]

> What I mean is that the pages contain markup like
> <TD ...><A NAME="M1711"></A>1711.</TD>

Strange, I didn't spot that.  Yes, I'd definitely recommend modifying
that,

> <TD ...><A NAME="M1711">1711.</A></TD>

Indeed. 


Alan J. Flavell

-------------------------

"Alan J. Flavell"
<flavell@mail.cern.ch> wrote:

>On Sun, 1 Nov 1998, Jukka Korpela wrote:
>
>> (I didn't know they supported the COMPACT attribute. I had thought
>> nobody was going to support or use it so I never tested on the 4
>> versions...)
>
>To be honest with you, neither did I, until I tried it out for the
>purpose of this answer.   The Tweedles move in mysterious ways...

Now that I tried it too, using my test page
http://www.hut.fi/u/jkorpela/HTML3.2/lists.html
I would like to warn against using the COMPACT attribute for a DL
element under any circumstances - even for an undeniable definition
list - since the "support" in IE 4 and Netscape 4 seems to be worse
than no support. At least an author who intends to use it should first
check how it works for his case on those browsers.

My simple test related to DL COMPACT on that page is
<dl compact>
  <dt> alpha
  <dd> the first letter of the Greek alphabet
  <dt> beta
  <dd> the second letter of the Greek alphabet
  <dt> gamma
  <dd> the third letter of the Greek alphabet.
</dl>

This is roughly how it looks like on IE 4:
alphathe first letter of the Greek alphabet 
beta  the second letter of the Greek alphabet 
gamma
  the third letter of the Greek alphabet. 

And on Netscape 4:
alpha 
    the first letter of the Greek alphabet 
beta the second letter of the Greek alphabet 
gamma 
    the third letter of the Greek alphabet.

In special cases (e.g. when the DT term elements are _very_ short, say
3-4 characters all), those browsers might get things right. But this
in turn might depend on font size and things like that.


Jukka Korpela

--------------------------

> Now that I tried it too, using my test page
> http://www.hut.fi/u/jkorpela/HTML3.2/lists.html
> I would like to warn against using the COMPACT attribute for a DL
> element under any circumstances - even for an undeniable definition
> list - since the "support" in IE 4 and Netscape 4 seems to be worse
> than no support.

Oh dear.  Looks as if my suggestion ("you might just consider") would
best be withdrawn.  Sorry about that.


Alan J. Flavell

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Alan J. Flavell wrote:
>
>And the other thing, though this has nothing to do with your problem,
>is this perennial awkwardness that when the anchor destination is near
>the end of the document, the browser positions the page so that the foot
>of the document is at the foot of the browser window, rather than the
>anchor destination being at the top of the window.
>
>I must admit that when I'm composing a page like this, I use <PRE>
>to add a fistfull of blank lines at the end, to avoid this problem.
>(That should get my purist badge revoked, but it's what I do, anyhow).


Although I understand the problem, the solution you put forward has
itself an (I think fairly serious) problem:  Although it looks fine
(perhaps even much better) on screen, printing the document could easily
result in a blank page being produced.  This seems extremely
undesirable.  It also makes other, perhaps more important, ways of
navigating around the page meaningless (pressing END in my Win95
browsers for example).

I cannot really see how the browser-makers could do anything else than
what they do now without upsetting (future) implemetation of CSS and the
object model.  This extra space would not exist in any padding or margin
specification, so the browser would just have to invent it.  Again, this
could cause problems in a variety of media (eg how should the browser
decide on its colour? - if I put a DIV flush with the bottom of the BODY
which was, say, light grey and the body was white, the extra space would
probably result in an unwanted white band at the bottom).

I have read some of your pages where you used your quite shocking un-
purist hack ( ;-) ) and was confused at the time - now I understand.
(But disapprove!).

-- 
George Lund

--------------------------

George Lund wrote:

> I cannot really see how the browser-makers could do anything else than
> what they do now without upsetting (future) implemetation of CSS and the
> object model.  This extra space would not exist in any padding or margin
> specification, so the browser would just have to invent it. 

Well, then perhaps they should indeed "invent" it.

Alternatively, set some kind of marker (cursor, highlight, whatever)
to draw attention to the anchor destination, because otherwise it's
too easy to lose one's orientation in a document like this.

I know that _some_ people say the solution is to never write a
web document that will be longer than a typical page, and then one 
doesn't need internal anchors to find one's way around it: that's
probably good advice for many superficial purposes, but for any kind
of in-depth or serious work I would find it impossibly limiting.

> I have read some of your pages where you used your quite shocking un-
> purist hack ( ;-) ) and was confused at the time - now I understand.
> (But disapprove!).

An entirely fair comment.  I was always aware it was a compromise, and
that some might not like it.  Thanks for the discussion. 


Alan J. Flavell

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

morgan wrote:
> I'm doing a page that requires numerical footnote indicators. The
> small elevated numbers? Its been a few years (ok, maybe more than a
> few) so I don't remember the term for them.


   Superscript.  There is a markup element for it, SUP.  
For footnote calls in a research paper, however, it would
be advisable to use other means of distinguishing the
numerical call from the surrounding text, instead of or
in addition to the SUP element, for example,
    ... in rural areas."(4)    or
    ... in rural areas."<SUP>(4)</SUP>

The reason for this is that not every browser can display
SUP text in superscript.  Useful information on SUP is at:
 http://www.hut.fi/%7ejkorpela/HTML3.2/5.53.html
 http://www.htmlhelp.org/reference/wilbur/font/sup.html

-- 
  Warren Steel 

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

On 3 Nov 1998,  David wrote:

> I'll get back to the point that started me on the problem of "width".  My
> pages contain many widths for cells, in %.  I don't want to use pixels
> since I don't think that it's a good idea to hardwire things any more than
> necessary. 

I agree whole-heartedly.  Nobody compels you to specify cell widths,
at all, though.

> it doesn't seem
> to fit the people-part of HTML 4.0 Transitional,

Nor 3.2 either.  The reason seems to be that NS and MS implemented
percentage cell widths in contradictory ways.

>  What options do I have?

1. Don't try to specify column width at all.  Leave it to each browser
to do the best it can with the content that you're sending.

2. It's been observed that if you specify percentage widths on _all_
columns, taking care to make them add up to 100%, it will produce
almost the same results on NS and MS.  This has been pointed out by
various people, and is in the TABLE hints web page produced by one
W.Schlake (he of the Dr. Jekyll website and Mr. Hyde usenet posting
style).  You might just want to consider that, if option 1 doesn't
appeal to you.  There's some screenshots here:
http://ppewww.ph.gla.ac.uk/~flavell/www/tablewidth.html
to demonstrate the problem and the workaround.


Alan J. Flavell

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%