If you ask me, attempts to force people not to "leave a site" are rather effective but in a direction opposite to the intended one. Jukka Korpela %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > remember there's no IE for unix platforms, and lots of people out > > there have older computers that can't handle the resource demands of > > a current browser. > > True False. Partly ;-) IE for Solaris is quite competent at dragging a medium end Sun machine down into the ground... :) Robert Gormley %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% I wrote a short intro to CGI in C a while ago (http://members.xoom.com/_XOOM/kaeness/iesdev/cgi/19990615.html). It's a bit out of date, but works. If you want something newer and more debugged, mail me. Kae Verens (kverens@hotmail.com) %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Claus Färber wrote: > Mike Jones wrote: > > the Image tags an attribute called NATURALSIZEFLAG. > > I don't seen to be able to find any documentation on what this attribute > > is supposed to do... > > It labels the HTML as created by a broken software application. > heheheh and the author of the page as completely incontinent...I mean incompetent. Emile Axelrad %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Tom Line wrote in message <37ee4754@news.iglou.com>... >I've been observing mistakes and other odd human behavior at the >government agency where I work managing computer support. I'd like to >discuss and identify more errors to try to prevent them and direct >scholars on important issues to study. > >1. Fundamental Attribution Error. Seems to stem from need to direct >blame. This would be like accusing your doctor of causing your >disease. When you're sick, you see the doctor, therefore doctors cause >sickness. In my case, ALL user computer problems are caused by the >computer department. > >2. (Unknown Error) Mr Bean Error. An organization makes a poor choice >of how to resolve a problem and applies a series of half ass fixes; >each fix causing it's own problems. This digs into some serious money >in government where I work. Workers then become frustrated and blinded >by details and are hindered by resolving the more fundamental causes. 3. Tool Selected Before Problem Stated (or Understood) Error (subset of "Disorganized Religion Error"). You can see this one all over USENET: Linux Rules, Windows Sucks (and its corollary: Windows Rocks, Linux Bites). The tool drives the problem to be solved, rather than the problem leading to the selection of appropriate tools. ---Michael B. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% "You're about as scary as dead bumblebee, your threats have all the plausibility of Harvey Keitel in drag and you're as interesting as a gif in a text viewer." -- Doktor Pete %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Sue Sims spouted: > >"...the XML declaration in the first line _must be omitted_ or else >MSIE 4.5 Mac (the current shipping version for Macintosh) will >display a document as text..." > >The full message is here: > Oh dear. So that means we would be required to break our markup just to work around a bug in one particular browser? Specifically, in IE? The phrase '...and on that day Satan will be skating to work' springs to mind. ;) Jedi Master Yoda %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Jason O'Rourke wrote: > Roving Reporter wrote: > > REBUS wrote: > > > Tina - White Lake Web/IServPro wrote: > > > > Raj Dutt wrote: > > > > > Isofarro wrote: > > > > > > lawrence wrote: > > > > > > > Tina - White Lake Web/IServPro wrote: > > > > > > > > Sweets wrote: > > > > > > > > > I have seen pages that have hidden their source code. If you > > > > > > > > > right click and view source... or select View Source Code from > > > > > > > > > the menu, a page appears but it is blank. Does anyone know what > > > > > > > > > is needed to achieve this? > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > My browser must be broken, because that didn't work for me. > > > > > > > > > > > > Are you using a registered copy of the tag? It works > > > > > > fine for me. > > > > > > > > > > Your webserver has to have to support bmE, (blackMagik extensions). > > > > > I believe version 1.1 of bmE had a bug in the invisible tag, but > > > > > this was corrected in version 1.2 > > > > > > > > Unless you're on a Mac. Then you're pretty much screwed. > > > > > > Hell NO...we here at _1born/min_ have just release our version of > > > blackMagik call SucRPunch. It works on all browsers, all web servers, and > > > better yet, we guarantee that your code will never be seen by anyone, not > > > even the FBI, CIA, or NSA. How sure are we, well, if someone every reads > > > your code after installing our product, we'll give you a 100,000 dollars. > > > How's that! > > > > > > Here's all you do, just go to www.dumdum.com, and download the SucRPunch.exe > > > file. Afterwards, simply select Start|Run and type format C: and press > > > enter. Viola, your done. > > > > > > (Note: no instructions are included for those using macs because we figure > > > your smart enough to figure it out yourself.) > > > >I hate Macs...no command line interface. > >there is one for version X, though it felt peculiar for the few minutes I >spent helping my web developer out. Version X? Now that's one I've never heard of. I have heard of this mythical Mac command line interface, but never met it yet. I'm going to go for Linux soon, much to my relief. (Compared to Windows.) Roving Reporter ----- WebSlave wrote: > REBUS wrote: > > WebSlave wrote: > > > REBUS wrote: > > > > Hell NO...we here at _1born/min_ have just release our version of > > > > blackMagik call SucRPunch. > > > > > >And I figure you're smart enough to know that .exes won't run on a > > >Mac... > > > > Uh...yeah, sure. (Hint: I also assume that mac users know irony when they > > read it.) > > Naturally. Mine was just another useless comment on this useless thread. > Nothing personal intended. Oh, nothing personal taken, just clarifying the clarity of clarification. REBUS ------ John Forrest Tomlinson wrote: > Tina - White Lake Web/IServPro wrote: > > > > > > Opera (3.6 on Windows) doesn't support it. Actually, it does. Tina %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Darin McGrew writes: > Nick Lilavois wrote: > > even Flash is being used for more than it was first intended. > > I'll have to take your word for it. I've never noticed a problem from not > having Flash support. Just for your edification and amusement, borrow someone else's machine which has got flash support, and point it at my local council's website, . Then go and make yourself a leisurely cup of coffee. Then come back and see whether the front page has downloaded yet. What? It hasn't? Sit back and sip that coffee... What? your coffee's gone cold? Don't worry, you have plenty of time to brew another. *Now* you know why these 'designers' like using Flash on the Web. They've all got shares in Nescafe. Simon Brooke %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Daniel wrote: > Nick Lilavois a écrit dans le message : > > Put this in your HEAD (needed by MSIE4): > > Exactly what I want ! Thanks a lot Nick ! :) > Just great :) I read that comment *so* wrong! this ______ \ / \ / ||_ .' \ / | | <@ \ | _\ | =| \ _| | | Veronica Karlsson %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Arjun Ray wrote: >| Nick Lilavois wrote: >| : The problem is when a technology is invented, for example >| : Java, and there is no gizmo in the standard to support it. >Try it again, this time without the persuasive force of a familiar >name: > "The problem is when a technology is invented, for example > Foo, and there is no gizmo in the standard to support it." >How and why should a standard have a gizmo to support Foo? What makes >the absence of a gizmo to support Foo a *problem*? For that matter, >why is it that supporting Foo demands a *gizmo*? There's an overall rule that we all have to abide too, and that "designer d00des" live by, as in the "second law of thermodynamics" [2'nd td], without it no life would even be possible. It postulates that over a sufficiently long period of time, all things that happens in the world has to loose its energy and decay into a total uniform equilibrium, somteimes known as "total chaos" Eventually it may happen that "islands of a higher degree of order" (synergetic effects) develops inside such a system, but that then has to bee at the cost of an increased level of disorder in another part of the same system, since the [2'nd td] still rules, and the average level of energy inside the system as a whole, has to be maintained in a continuously decaying direction. One of these "synergetic energy inputs" we may have seen in recent history was the "outburst" from TBL and his close by enthusiasts. And naturally [2'nd td] then must bring its champion Marc Andreesen (with equally enthusiastic friends) into the ring, to fight it all, and bring everything back into a slowly decaying equilibrium, targeting full chaos once again :) Jan Roland Eriksson %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Park wrote: > i remember seeing somewhere a website where it checks and finds all the > webpages on the internet that have a link to your site. does anyone > have the URL to this site? This tool will help you view what pages the different search engines list for your URL: http://www.eons.com/position.htm The results can be pretty interesting! Jeffrey Ellison ------- Hotbot will do it- just select "links to this url" in the "look for" selector. Nick Lilavois ------- Into the 'Search' box type link:your.domain.name/page Example link:jasmine.org.uk/~simon This only finds pages which AlatVista has indexed, but there are rather a lot of those. Simon Brooke ------- just look at the referer field in your server logs. Mark ------- www.linkpopularity.com Anthony Butcher ------- Try www.linkstoyou.com rhett %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Sam Spade Tools http://www.samspade.org/t/ %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% here's a php script I put in my 404 redirect page that emails me when a link is dead: (I have php parse my html files) Mark ----- "David Wier" writes: > Where can I find more information on all the tags for PHP? No, I don't know why it's .net not .org, either. Simon Brooke %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Cool URIs never change: http://www.w3.org/Provider/Style/URI.html %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% On Thu, 11 Nov 1999 17:17:42 GMT, isoma spouted: > >By "best used with" I thought they'd meant "hard to use without" - so >often the case, IME. To my mind the most common translation seems to be 'not tested with anything except'. JMY %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Date: Wed, 3 Nov 1999 Anyone who isn't using a CSS-supporting browser by now is clearly more interested in browsing the content with a stable browser, than in fussing about the appearance. Fine, that's their right too. Alan J. Flavell %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% http://www.asktog.com/ ----- PS Obligatory Usability tip: try reading Nathan Shedross' Information Interaction Design for some thought provoking ideas about information and interaction: http://www.nathan.com/thoughts/unified/ Samu Mielonen %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Making Online Information Usable http://world.std.com/~uieweb/online.htm %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% On Sat, 06 Nov 1999 19:26:27 +0100, Rijk wrote: >On Sat, 06 Nov 1999 21:54:14 +0800, Kelvin Yiu >wrote: > >>Is there any program which can remove all font tags or properties for >>me??? > >Yes, TIDY ! > >http://www.w3.org/People/Raggett/tidy/ And to follow up on myself: There's a online version, made available by Chami, at http://www.chamisplace.com/asp/hk.asp . Tidy is also built into HTML-Kit, a nice HTML-editor, see http://www.chami.com/html-kit/ Rijk van Geijtenbeek ------- "Soetjianto" writes: > Is there a tool to do the reverse? I have created pages that use CSS and now > I want to make them compatible for browsers v.3. Please help. Q: How many Zen masters does it take to change a lightbulb? A: None. The universe revolves around the bulb, and the master leaves it alone. -- simon@jasmine.org.uk (Simon Brooke) %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > It isn't only the long lines. When I resize my browser, the copy > is still crammed against the sides of the window. It's like making > the reader hit his head against a wall every time he begins and > finishes a line. Reading text is not like sprinting to first base; one's eyes do not need open space in which to decelerate before going to the next line. Otherwise we'd write each paragraph in a spiral starting from the center so that you could continually track the text. art repr - e I eronic s I V a e C I'd n S o y t t A t l (N l o a l o e t n p amr a c i a u v o o ti e m n h e . tiw pu : )- greg@apple2.com %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% | >(I hate having to perform Jeopardectomies.) he was refering to a phenomenon that has lately become all too common but no less rude than it always has been, thanks to the spread of OE. Here is a periodic posting on the issue: Have you heard of the Good Net-Keeping Seal of Approval? You'll find evaluations (which will explain why OE is *particularly* usenet-hostile) and pointers to competent newsreaders. :ar %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% >Nick> lynx -dump document.html | ispell That should be "lynx -dump -nolist ..." %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Claus Färber wrote: >Dave Eastabrook schrieb/wrote: >> That's a case where "Use any browser" is a bit of a luxury, even >> though of course your site is visible outside your own locum. > >Nope. In every situation, using browser specific features that make your >site incompatible with HTML is the luxury, not writing portable code. OK, this is a little out of context now, but I used to totally agree with you and still have the "Use any browser" lozenge somewhere on the sites. But it dawned on me that if the crap 95% of websites out there even worked properly with one browser, it'd be a vast improvement :) Dave Eastabrook %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% skip@ejourney.com wrote: > HOW DO YOU MAKE IT SNOW IN HTML FOR A STATIONERY PAGE? You have to dance in a circle three times while chanting snow, snow, snow. Sorry, but I don't have a clue what you are asking about! John McFarlane %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% In article <383A467F.87E6A89F@orthogon.de>, "Kai L. Buerhoop" wrote: > This is a multi-part message in MIME format. Mimes, like GUIs, need to communicate without language. They use a lot of gestures and are limited in what they can communicate. The command line is complicated and has more of a learning curve. The same is true of any human language. Many a time, in a country where I did not speak the language, I have managed to communicate certain things by pointing, nodding, smiling. However, I can communicate so much more when granted the use of words, and I can communicate more swiftly. We should not expect GUIs to be as useful as command lines for the expression of complicated computer commands. The GUI is for pointing and grunting. gnohmon8715@my-deja.com %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% On 27 Nov 1999, Abigail wrote: > Bill Godfrey (bill-godfrey@usa.net) wrote on MMCCLXXIX September MCMXCIII > ++ Get yourself a copy of Lynx, it's a good approximation of what a brailler > ++ or speech synthesiser will "see". > > Speech synthesisers don't have more trouble with Lynx than with MSIE; > in fact, I know blind people who prefer MSIE instead of Lynx, as both > forms and frames are much harder to navigate in Lynx than in MSIE. There may be some misunderstanding here. Assessing a page with Lynx is a reasonable way for a sighted author to try dissociating the textual content from its spatial positioning in the typical graphical browser. In that way, it allows the author to assess whether their content is still comprehensible when that spatial positioning is taken away. Perhaps this was what Bill was getting at in relation to braillers and speech synthesisers. As such, I'd say it's a useful authoring tool for promoting accessibility. What your opinions might be on Lynx's benefits as an operational browser for various kinds of disability is actually a separate issue. By the way, I have read that MSIE has a number of accessibility features that are not usually described to normal users (nor even available via the normal accessibility-related preferences dialogs), that are only brought into play when IE discovers that it is being invoked in a relevant (e.g speech browser) environment. As such, a normal sighted author using MSIE would have no clear idea what it might do in that environment. I get uneasy feelings when I look at MS's web pages that are intended to promote disability access, and compare them with the actual design of their own pages for the mass audience to see to what extent they are applying the principles that their own disability access pages are trying to promote. Looking at the extent to which their web authoring tools do (or don't) promote good disability-compatible authoring practices is another source of discontent. It's little wonder that many misguided authors are claiming that they can't afford the extra effort that they believe they would need for designing in a disability-friendly fashion. This in spite of the fact that HTML was originally designed to maximise the compatibility of content with the widest range of browsing situations, and much of the effort these authors are putting in is in fact damaging accessibility. Alan J. Flavell %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Kurt Tappe (ktappe@voicenet.com) wrote on MMCCLXXXVII September MCMXCIII in : ++ ++ One hopes that if you're a programmer you don't choose to omit closing ++ parentheses and semicolons... :-) Personally, I cringe when I see HTML ++ code that doesn't have closing tags. It's just....unclean. I dislike languages where you have to put in tokens than can be interfered from the context. The computer should serve the user - not the other way around. Abigail %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Kurt Tappe wrote: >Would
    and
indent the same distance? I use them and they work >well. This page might give another (and more correct) view on (UL) "indenting" perhaps? -- Jan Roland Eriksson %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% The Idiot's Guide to Solving Perl CGI Problems http://www.perl.com/CPAN-local/doc/FAQs/cgi/idiots-guide.html %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Ole Yearian wrote: [...] | Realistically, any instance of "e-Commerce" [...], distance learning | (my bag) or other PC based transactional application absolutely | *MUST* offer a web-based alternative. The market demands it, if | nothing else. The "web-based alternative" shibboleth is a classic case of "solutions as requirements". Arjun Ray %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% J Kirby Inwod wrote: > I have patched together a solution that is working for me. Using a couple CGI > scripts anyone who attempts to reach a broken link on my site gets taken to > http://www.kirwood.com/404.htm Check out the 404 Research Labs at http://www.plinko.net/404/ which has scores of more functional (some downright impressive) solutions. miguel %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Tim Thomas wrote: >Great, Alan! > >But, that wasn't his question. He didn't ask whether it was right or wrong. >He asked how it could be done. Trouble is, the answer is "Well, it can't. Froce doesn't work on the WWW." A customer of mine needed help setting the correct MIME type for a tdownloadable typeface that she had uploaded to her webserver. She asked me what the final page looked like, because she wasn't in front of a net-connected computer at the time. It took quite a while for her to understand that I actually preferred Lynx on Linux to IE on Windows, and no, I wasn't going to reinstall my computer just to check the presentation of her webpage. isoma %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% You step in the stream, but the water has moved on. That page is not here. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > I agree. A decent OS should let the user set policies that determine the > privileges granted programs (including the OS itself). Unfortunately, > putting this control in the user's hands is not in Microsoft's best > interest. The fact that a person's tools now have the capability to act > against their owner's interests is scary. I think this issue goes beyond > privacy. Windows needs a better way to manage file associations, including locking an extension so another program won't change its text description. I hate how Netscape changes my HTML file which is associated with a text editor to a "Netscape Hypertext Document" associated to its browser. > One thing that would help this situation would be one of those watchdog > web sites that would list all the sneaky little activities of popular > software packages. Or maybe there already is one? The Windows Annoyances Page (http://www.annoyances.org) used to have a list. All see the Interface Hall of Shame/Fame (http://www.iarchitect.com). Dave (david@pdh.com) %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Jukka Korpela wrote: > Gary McCafferty wrote: > > I could be changing some sections every now and again and it will > > be a major headache to go through 100-200 pages and change the > > links. > > You won't need anything like that if your pages contain just a link > to the index page. Or at most links to the section index pages too, > making it a total of seven links, which should (really should!) be > rather permanent. Yes, but this is only a "solution" constrained by the inadequacies of explicit direct linking, on the grounds that this is unfortunately "all we have". Gary's basic problem, it seems, is maintenance of navigational links within pages, so advising that these should point to permanent targets is writing off the real problem as unsolvable:) > If you need to modify the set of such links (say, to add a new > section, after careful consideration, trying first to put new things > under old sections), then it's a one-time job to find a tool which > does a simple string replacement over a set of files and run it. This is basically the "preprocessor" solution. Search-n-replace has robustness problems unless the string to be found is designed to be specifically identifiable. So, if you need to take the trouble, you might as well consider a good "preprocessing" system - which means forget about the word "preprocessing" and think in terms of the HTML format as the result of *post*-processing. The fundamental problem here is that the HREF and SRC attributes in HTML take (mutable) URLs as values, and mutable values are bad for "source" formats (they should be filled in parametrically "on the way out".) This is one of the "design decisions" in HTML optimizing for the ephemeral, against maintenance. So, the robust solution - *if* maintenance is a concern - is to use a better source format and convert to HTML for Web publication. The "better source format" needn't be radically different, just a more rationalized HTML. Some interesting historical notes: Dan Connolly originally proposed a better design for HREF, but this was shot down by TimBL: The same issue came up with Eliot Kimber's critique of an early HTML DTD: (from ) which Dan Connolly responded to quite a while later, echoing TimBL's sentiments. (from the www-talk archive that used to be at EIT and has yet to be fully restored at lists.w3.org) Arjun Ray %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% "A.J. Norman" wrote: > MåTT W¡££¡åM§ wrote: > > Are there any significant reasons why companies still choose to use > > PDF documents over HTML? From what I can tell the only advantage is > > that PDF documents are displayed in a fixed manor that does not vary > > with resolution and are therefore more printer friendly. > > Got it in one. We're in the process of putting our exam papers on the > web, and this term's short notice kludge was to scan the things and > zip them up as archives of GIF files with an HTML index. The > long-term solution is to use PDF since it can handle the presentation > of illustrations and formulae and making the files is no more > difficult than printing the original Word documents (so the > secretaries can do it). This is a bit OT, I suppose, but why on earth are you scanning the exam papers instead of converting them to PDF right off? I used to work as the assistant to a professor at MIT, and for one term (I think it was 1995 or 1996, freshman-level chemistry class) I scanned exams (quizzes, homework, etc.) for our web site (which I designed, built, and maintained, btw. Even a secretary can do it...) This was after attempting to convert the Word documents to HTML, and finding that formulae and equations got converted to badly-munged GIFs. PDF was a godsend, even if I did have to convert the files to postscript format first. Anyway, for us, HTML and PDF complemented each other very nicely. When the layout of a document you're distributing on the web really, truly is crucial, PDF is the way to go. Terra Cholfin ----- > This is a bit OT, I suppose, but why on earth are you scanning the > exam papers instead of converting them to PDF right off? Because we didn't have the software, didn't have time to buy it and show the secretary how to use it, and the papers had to be on the web immediately if not sooner. As I said it was a short-term solution which I wouldn't recommend to anyone. Andrew Norman ----- Terra Cholfin wrote: > This was after attempting to convert the Word > documents to HTML, and finding that formulae and equations got converted > to badly-munged GIFs. Did you try using rtftohtml to do the conversion of the Word document, then a graphic conversion program to convert the resultant BMPs or PICs to GIFs? This gives much better result than any converter Microsoft has produced? David R Newman ----- "MåTT W¡££¡åM§" wrote: > > Are there any significant reasons why companies still choose to use PDF > documents over HTML? From what I can tell the only advantage is that PDF > documents are displayed in a fixed manor that does not vary with resolution > and are therefore more printer friendly. Having been obliged to use PDF on a few occasions, i can tell you my reasons: 1) Good enough reason: Printer friendly (as you mentioned). Certain things like application forms, etc. just plain print better from PDF. 2) Less good reason: Laziness/apathy. A whole slew of PDF files were given to me. My choices... turn them into HTML, or just throw them on the web as is. They would have been fine in HTML, and would have loaded quicker... but it was simpler not to bother. Especially since other people would be making changes to the PDF, and it would be a pain in the butt to update the HTML regularly. I have learned it is easier to teach people how to use Fetch than to teach them HTML. BTW: if anyone knows how to convert a PDF image into a format readable by graphics programs, please let me know. Eva L. Hulse ---- MåTT W¡££¡åM§ wrote: > Are there any significant reasons why companies still choose to use PDF > documents over HTML? Several reasons that I can think of: 1 - What you see on the screen is what is printed. This can be critical in documents with legal or regulatory requirements. With HTML, you have little control of printing functions such as pagination, headers/footers, etc. 2 - Can be created from almost any electronic format. If you can print to PostScript, it can be converted to PDF easily with no loss of fidelity - as opposed to using the MS Office capabilities to convert to HTML, which produces awful results if a user has any web browser other that IE. 3 - PDF offers very high control over presentation. This can be important for technical material requiring mathematical expressions. In HTML, just about the only way to do this is by embedding bit-mapped images of the math. 4 - Great portability. Acrobat plug-ins are available for almost every serious platform and web browser. 5 - Ability to zoom in on complex details. Robert Haar ----- "MåTT W¡££¡åM§" wrote: > Are there any significant reasons why companies still choose to use PDF > documents over HTML? Yes, they want to make the file size so large that people will give up waiting for the downloads, and not read them. Especially in countries where people pay per minute for telephone calls. David R Newman ----- The whole point of PDF files is so people can publish over the web. PDFs are acceptable by professional printers, so you can place a PDF document of a book on a webpage for download, and then someone can grab it, take to a printer, and print an exact copy of the original. Also its heaps easier to create. A click of a button compared to all the time spent laying out an HTML page. Eric Hall ---- Eric Hall wrote: > The whole point of PDF files is so people can publish over the web. No, Adobe designed the PDF format for various purposes of document interchange. I haven't checked the history facts, but PDF surely could predate the Web, and the general idea of having a digital data format which tries to present the appearance of a printed document is very old. > Also its heaps easier to create. It depends. You need a program to generate PDF, unless you are a masochist or guru or both. You don't need a program to generate HTML, except in the trivial sense of having a text editor. > A click of a button compared to all the time spent laying out an HTML page. If layout is what you want, PDF is absolutely superior to HTML, as are many other languages and programs, some of which are rather old. Control over layout is almost strictly complementary to universality, adaptability to varying browsing situations and user agents, and automatic processability of content for purposes like indexing for search engines. So it's a good idea to make up your mind. Note that you can take both - it just takes some extra work. (You could, for example, write the document in HTML, run it through html2ps with the settings you prefer, and convert the resulting PostScript file to PDF using Ghostscript, then perhaps edit the PDF file (probably using Acrobat) to modify the appearance. And you could make all three versions available on the Web. Guess which one will most likely be found via search engines, be viewed on a handheld device, or listened to while fishing?) Jukka Korpela ------ Jukka Korpela wrote: > If layout is what you want, PDF is absolutely superior to HTML, as are > many other languages and programs, some of which are rather old. Yes, I agree with what you say, but (this being c.i.w.a.h) I would have to say "if layout is what you want above all else: think again". > Control over layout is almost strictly complementary to universality, Is that how you spell "antithesis"? ;-) Fortunately, the WWW recognised the dichotomy between presentation and content that exists (like it or not), and provides the tools for offering both. Now, the artists are going to tell us that their content is a vital part of their presentation, and their presentation is a vital part of their content. Well, fair comment, but the reality of web presentation says otherwise, and IMNSHO the true artist (which I am not, but I think I can recognise one when I see one), understands the medium in which they are working, and exploits its strengths, and knows how to bypass its weaknesses. This was as true of the classical painters, and cathedral builders, and so on, as it would be for those who chose the WWW as their medium. Sadly, many specialists in paper publishing are trying to factor out the strengths of the WWW and make themselves reliant on its weaknesses. seasons greetings and peace to all (spammers and those who won't read the FAQs may be excluded ;-) Alan J. Flavell ----- "MåTT W¡££¡åM§" wrote: > > Are there any significant reasons why companies still choose to use PDF > documents over HTML? Obviously , having a predictably consistant rendering is probably high on the reasons list , however , same is really in conflict with the more suggestive nature of HTML and other web standards. PDF provides few fallback rendering abilities and often annoys users with it's bulk, which in the end does not IMHO justify it's existance. > From what I can tell the only advantage is that PDF > documents are displayed in a fixed manor that does not vary with resolution > and are therefore more printer friendly. Also, PDF is more difficult to > modify by the average user as they lack the appropriate software (Acrobat > Writer, WordPerfect 2000, etc.). Any more ideas on why this is so? If you are ORDERED by a concern to commit PDF , life is what one is paid for but if your free to make a choice , why would you want to , without cause reduce audiance share and at the same time increase the burdon of those truly trying to view the information? Dr. Clue (A.K.A. Ian A. Storms ------ dantso@cris.com wrote: > drclue@drclue.net wrote: > > PDF provides few fallback rendering abilities and often annoys users > > with it's bulk, which in the end does not IMHO justify it's existance. > > Oh ? How would you suggest that the IRS distribute 1040 tax > forms over the Internet ? Remember in many cases forms are required > to be absolutely uniform so they can be reliably machine read. I would suggest that given the web-age, one could design 1040 renderings for the web at large , and provide instant on-line submission thus largely avoiding the need for a printed version at all. If one is trully needfull of a non-web version downloadable yet by the web such should probably be renderable in the web-space w/o a specific plug-in. Machine reading should be adapted to the general context. I first designed OCR systems and hardware in the days of the C64, so limitations of machine reading less functionable then what I could achieved over a decade ago are no excuse at all. I trully expect my humble freeform OCR of 198? or so should be more than surpassed by any recent technologies available off-the-shelf. PDF is possibly viable outside the actual webspace , like a word document, however neither has anything really to do with web side presentation, and attempts to force same are IMHO a joke at best. Dr. Clue (A.K.A. Ian A. Storms) ------ Gil Harvey wrote: > dantso@cris.com wrote: > > Oh ? How would you suggest that the IRS distribute 1040 tax > > forms over the Internet ? Remember in many cases forms are required > > to be absolutely uniform so they can be reliably machine read. > > I suggest they go to a flat rate tax, eliminate 1040's and all > the other crap, lay off a few thousand people and ... Well you get the > idea. Actually , although it goes far afield of the context , I would be in favor of eleminating all taxes save a flat national sales tax which requires the goverment to perform in such a fashion as to allow all to spend in order for them to spend. Dr. Clue (A.K.A. Ian A. Storms) ------ .. wrote: >I would suggest that given the web-age, one could design 1040 >renderings for the web at large , and provide instant on-line submission >thus largely avoiding the need for a printed version at all. Except, of course, for the requirement that you keep copies of your completed 1040 submission for 7 (or is it 3?) years. The printed copies come in handy come audit time; or if you want to apply for a morgage. Think for a moment. Would you really want the agency responsible for checking your 1040 to have the *ONLY* copy of your 1040? (Ignoring the fact that you still have to submit all that ancillary paper; such as 1099INTs, W2s, checks for underpayment, etc.) People are twichy about submitting credit card numbers over a secure http connection - would you realy want to send your complete financial life (ssn, address, income, bank account numbers, etc.) over the Internet? Given the gaping security holes in the big two wowsers[2], would you really trust them to keep your secrets? Think about some of the problems that commerce sites have been having this season. Do you really think you can construct a server robust enough to handle the millions of transactions required to institute an on-line 1040 filing system? >PDF is possibly viable outside the actual webspace, like a word >document, however neither has anything really to do with web side >presentation, and attempts to force same are IMHO a joke at best. This is either an application of the Golden Hammer[3]; or a fundamental misunderstanding of one of the purposes of the web - to allow the presentation of disprate document types in a common framework. [1] [2] Thanks Arjun. I just love that term. [3] The Golden Hammer Antipattern summarizes the old saw "If all you have is a hammer, every problem looks like a nail." In this case, the hammer is HTML. Timothy R Prodin ------ Timothy R Prodin wrote: > .. wrote: > >I would suggest that given the web-age, one could design 1040 > >renderings for the web at large , and provide instant on-line submission > >thus largely avoiding the need for a printed version at all. > > Except, of course, for the requirement that you keep copies of your > completed 1040 submission for 7 (or is it 3?) years. > The printed copies come in handy come audit time; or if you want to > apply for a morgage. > Think for a moment. Would you really want the agency responsible for > checking your 1040 to have the *ONLY* copy of your 1040? There was nothing in the thought that said you should not be able to print the web-rendered form > (Ignoring the fact that you still have to submit all that ancillary > paper; such as 1099INTs, W2s, checks for underpayment, etc.) Again , while having a printed version is wise , I see little reason why the computer generated versions should not be piped in within the same electronic framework > People are twichy about submitting credit card numbers over a secure > http connection - would you realy want to send your complete financial > life (ssn, address, income, bank account numbers, etc.) over the > Internet? Gosh , someone who believes it's not already there, would you like to buy a bridge? > Given the gaping security holes in the big two wowsers[2], would you > really trust them to keep your secrets? If you have written it down , it will soon be on-line , so what's the point? Everytime you even use the ATM card , your life becomes an open book, be it a hotel room, or bar tab. > Think about some of the problems that commerce sites have been > having this season. Do you really think you can construct > a server robust enough to handle the millions of transactions > required to institute an on-line 1040 filing system? Gosh my ATM and credit cards have far more traffic, so I guess the answer is "are you kidding?". If we can through the combined medium do that, a ramped response to 1040's should not even be a real sweat > >PDF is possibly viable outside the actual webspace, like a word > >document, however neither has anything really to do with web side > >presentation, and attempts to force same are IMHO a joke at best. > > This is either an application of the Golden Hammer[3]; or a > fundamental misunderstanding of one of the purposes of the web - > to allow the presentation of disprate document types in a > common framework. I think one mis-appropriates the golden hammer by reverse application to somehow validate the minority PDF format into itself being the afore mentioned hammer. When a new technology is trully required to address an issue , new technologies are free to fill the gap in a compliant manner, but to deny the standard's abilities to afford the same functionality is some kind of drug this hippy aint tried yet. Somewhere in this thread PDF was cited as being the better method of producing machine readable output, yet this is mental gymnastics , as the information required can be conveyed in long existing web-formats without requiring a side trip to paper and an ocr engine. In short , what are you thinking? > [3] The Golden Hammer Antipattern summarizes the old saw "If all you > have is a hammer, every problem looks like a nail." In this case, the > hammer is HTML. Dr. Clue (A.K.A. Ian A. Storms) ----- .. wrote: > Timothy R Prodin wrote: > > .. wrote: > > > > Think for a moment. Would you really want the agency responsible for > > checking your 1040 to have the *ONLY* copy of your 1040? > > There was nothing in the thought that said you should not be able > to print the web-rendered form Of course, then you are back to needing a method to guarantee exact layout on the page. There are, after all, legal requirements. > > (Ignoring the fact that you still have to submit all that ancillary > > paper; such as 1099INTs, W2s, checks for underpayment, etc.) > > Again , while having a printed version is wise , I see little reason > why the computer generated versions should not be piped in within > the same electronic framework Because, the information is supplied to the Goverment twice - once by your financial institution or employer, and once by you. The Fifth amendment to the constitution prohibits the government from performing computerized cross checks of the two. Insisting that we supply the same information in the same medium will erode a sacred protection against government intrusion in our personal lives. Besides, if I am typing in my own 1099INTs and W2s; how does the government test the authenticity of the numbers that I supply? The paper documents have some prima facie evidence that they are authentic. > > [W]ould you realy want to send your complete financial > > life (ssn, address, income, bank account numbers, etc.) over the > > Internet? > > Gosh , someone who believes it's not already there, would you like to > buy a bridge? There is a big difference between credit card numbers (where you have a good deal of liability protection, by law); and all the information contained in complete 1040. The risk is proportional to the exposure. The higher exposure in your complete fiscal year documents drives the risk much higher. > > Given the gaping security holes in the big two wowsers[2], would you > > really trust them to keep your secrets? > > If you have written it down , it will soon be on-line , so what's > the point? This is specifically not true. > Everytime you even use the ATM card , your life becomes > an open book, be it a hotel room, or bar tab. And this isn't either. Those transactions are point-to-point secure between the retailer and the credit card company. HTML, and its implementations in the wowsers, don't have this guarantee. > > Think about some of the problems that commerce sites have been > > having this season. Do you really think you can construct > > a server robust enough to handle the millions of transactions > > required to institute an on-line 1040 filing system? > > Gosh my ATM and credit cards have far more traffic, so I guess > the answer is "are you kidding?". You have over 100 million transactions in 3 1/2 months on one site? Remember, everybody who works needs to file a 1040 form. You don't get the information you need until 1 Jan; and it is due (or your application for extention) 15 April. > > This is either an application of the Golden Hammer[3]; or a > > fundamental misunderstanding of one of the purposes of the web - > > to allow the presentation of disprate document types in a > > common framework. > > I think one mis-appropriates the golden hammer by reverse application > to somehow validate the minority PDF format into itself being the > afore mentioned hammer. Specifically what I meant in alluding to the Golden Hammer was your application of "All I have is HTML, let me use it; despite all its shortcomings on this particular problem (Government documents that have a large legal burden) and in spite of the fact that better solutions exist." > but to deny the standard's abilities to afford the same functionality > is some kind of drug this hippy aint tried yet. The basic problem here is how do you assert absolute presentational control over your information and distribute it in a platform indpenedent manner? > Somewhere in this thread PDF was cited as being the better method of > producing machine readable output, As a side effect of having a requirement to be stored in a paper format; the presenational control that PDF allows eases this machine readability. > In short, what are you thinking? That there is an overburden of legal requirements in this particular problem, and in other problems similar to it, that makes PDF a useful medium type in the web. More generally, you have demonstrated a fundamental misunderstanding of what the web is really about. There is certainly no reason that PDF should be excluded as a valid type. The entire point of the web was, and is, to use HTML as a textual glue to hang together disprate media types. Timothy R Prodin ----- Timothy R Prodin wrote: > Of course, then you are back to needing a method to guarantee > exact layout on the page. There are, after all, legal requirements. While there might be legal requirements for submission , evidence has at least a different standard , which itself will over time grow to address the techonological age in which it resides, as it has for all time. > Because, the information is supplied to the Goverment twice - > once by your financial institution or employer, and once by you. > The Fifth amendment to the constitution prohibits the government > from performing computerized cross checks of the two. While I would be very happy to have amendments interpreted in favor of the people , this has not been the case for many decades. I would most happily supply some funding for the rounds required in a revolt , but the best revolt uses the tools given by ones foe. > Insisting that we supply the same information in the same > medium will erode a sacred protection against government intrusion > in our personal lives. This makes little sense to me so perhaps you will give me more information on this view, as I believe that their end will have a computer and no matter the format used to deliver the information it is only a matter of time before same is in the "system". > Besides, if I am typing in my own 1099INTs and W2s; how > does the government test the authenticity of the numbers that > I supply? The paper documents have some prima facie evidence > that they are authentic. A secure submission from a known source will over time carry some weight, however a trully hand written version woould indeed have more weight. All these issues are wonderfull points of discussion , but they do tend to travel far affield > The risk is proportional to the exposure. The higher exposure in > your complete fiscal year documents drives the risk much higher. Gosh, if you want to feed the money requirements , I could as is almost tell you when you first jerked off, which hand you used and how many strokes it took. Sure thats a major exageration, but financially it is reasonably accurate dispite goverment efforts. > > If you have written it down , it will soon be on-line , so what's > > the point? > > This is specifically not true. How is it so not true? It may be an extream statement , but the motivation behind it and it's intended rendering is accurate. > > Everytime you even use the ATM card , your life becomes > > an open book, be it a hotel room, or bar tab. > > And this isn't either. Those transactions are point-to-point > secure between the retailer and the credit card company. > HTML, and its implementations in the wowsers, > don't have this guarantee. Wow, you really believe your CC transactions can't be had? > You have over 100 million transactions in 3 1/2 months on > one site? The statement misses the response. I myself am lucky if my ISP can support my daily load , however the point is that sites like at a min "yahoo" , can support that traffic within a couple of days as opposed to three and a half months. > Remember, everybody who works needs to file a 1040 form. > You don't get the information you need until 1 Jan; and it is > due (or your application for extention) 15 April. It's all a bunch of horse shit anyway , as taxes should be collected at the point of sale and there should not be income tax at all. > Specifically what I meant in alluding to the Golden Hammer > was your application of "All I have is HTML, let me use it; > despite all its shortcomings on this particular problem (Government > documents that have a large legal burden) and in spite of the > fact that better solutions exist." While the past can be clunge to with form factors , the reality will be secure data , end of that line! > The basic problem here is how do you assert absolute > presentational control over your information and distribute > it in a platform indpenedent manner? You don't have to. The problem is itself a straw man. > > Somewhere in this thread PDF was cited as being the better method of > > producing machine readable output, It was machine born in the first place , so PDF is a simple insult to progress in that regard. > That there is an overburden of legal requirements in this particular > problem, and in other problems similar to it, that makes > PDF a useful medium type in the web. PDF is an excuse to stall progress in this context and hardly more > More generally, you have demonstrated a fundamental > misunderstanding of what the web is really about. There > is certainly no reason that PDF should be excluded as a valid > type. The entire point of the web was, and is, to use HTML as > a textual glue to hang together disprate media types. I guess you missed the point , as the web is supposed to be an effective low budget method of communications, and not as much a hotbead for selling pin striping. I doubt you'll find agreement for this point with many who have been here as long if not longer than I have. Yes , hanging external media is a real issue , but intoducing un-needed garbage was never the point. Dr. Clue (A.K.A. Ian A. Storms ----- > Oh ? How would you suggest that the IRS distribute 1040 tax >forms over the Internet ? Remember in many cases forms are required >to be absolutely uniform so they can be reliably machine read. I'd use SVG for that. If I were using HTML I'd use a form and allow submission via HTTP. Tim Bannister ----- > I'd use SVG for that. Considering the practicality and verbosity of SVG distribution, this may be a particularly appropriate method of facillitating income confiscation.... ;-) jd ----- >I'd use SVG for that. Can many browsers use that format? John Forrest Tomlinson %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Les Jones wrote: > > What are some ways of estimating another web site's traffic? If you > didn't have access to the log files, couldn't install a counter, etc., > what would you do? > Another way is Alexa: http://widener.alexa.com/sitedata?srv=-1&client=3&cli=10&url=domain.com:80/ In the above URL replace domain.com with the site of your choice. On the Alexa results page you see a traffic count in the left column. Move your mouse over the * and see the rank of the site. Karl Heinz Resch ---- REBUS wrote: > Stephane Theroux wrote: > > Karl Heinz Resch wrote: > > > Another way is Alexa: > > > > Alexa stats aren't totally reliable. For instance it states that > > my site comprises of 15 pages, when the true figure is over 900 > > pages. > > More than not reliable, totally useless. These stats are from users who > visit the site "through" the Alexa service, which includes only newbies and > the perennially clueless. (Oh, I know, that's harsh...but its true. Real > surfers only use Google.com.) no, sorry but you have no idea what you're talking about. Alexa stats indicate alexa users who visited a site with the alexa service enabled in their browser. It has nothing to do with the referring document. On the average, alexa numbers are fairly reliable for sites that have heavy traffic and have been around for awhile, but can misreport stats on sites that are new, or have a higher than average alexa user to non-alexa user ratio. There is also some question as to what alexa considers a 'site' if your site is split among different domains for instance or if your 'site' is a subdirectory of a larger domain, then your stats will be misreported. Mark --- REBUS wrote: > Mark wrote: > > no, sorry but you have no idea what you're talking about. Alexa stats > > indicate alexa users who visited a site with the alexa service enabled > > in their browser. It has nothing to do with the referring document. > > And how would the "service" be enabled if a user didn't have it installed > (i.e., wasn't using it). What percentage of browsers have "alexa" enabled. > If it isn't close to 80-90% than its stats are meaningless, or worse, > misleading. Well, it's on by default in Netscape, I'd say that alone is a nice chunk of the browser market, and while it's not on by default in IE, those kind of numbers are good enough for me. Yes, the numbers are misleading, but that's life. > > On the average, alexa numbers are fairly reliable for sites that have > > heavy traffic and have been around for awhile, but can misreport stats > > on sites that are new, or have a higher than average alexa user to > > non-alexa user ratio. > > Exactly, you just described potentially ever site. How does one know if > their or the subject's site has a "high alexa to non-alexa ratio. Make a guess based on what you know of the site I guess > There is also some question as to what alexa > > considers a 'site' if your site is split among different domains for > > instance or if your 'site' is a subdirectory of a larger domain, then > > your stats will be misreported. > > More waffleing. > > Let me restate. Alexa ratings are usless. If you're looking for a foolproof way to determine exactly what kind of traffic a site gets, you can email the admin and ask to go over the logs, but Alexa is the most useful way I've found to predict roughly what kind of traffic a site gets. Mark ---- > More than not reliable, totally useless. Reported my site as belonging to Pioneer Electronics, gave it 5 stars for freshness (I last updated about 6 months ago) and stated I had over 2,500 pages, when in fact there are about 90. apart from that, pretty good, I would say Dan Evans --- Les Jones wrote: > What are some ways of estimating another web site's traffic? If you > didn't have access to the log files, couldn't install a counter, etc., > what would you do? E-mail them saying you want to advertise with them, then take the traffic they say they are getting and multiply it by 0.1 - this formula is surprisingly accurate... "Online Services" ---- > > It's turned out to be an interesting problem, and I'm sure I've > > overlooked many solutions. What are some other ways of estimating > > another web site's traffic? Recently a few tech shows (possibly C|Net's) were running stories of a service that tracks the performance of all your competitors, for a cost. I can't for the life of me remember anything about the product name though. The story hyped the service so accurate that it could totally revolutionize the way companies do business. Maybe somebody else remembers more. Tero Paananen ---- Tero Paananen wrote: > The story hyped the service so accurate that it could totally > revolutionize the way companies do business. You mean forceing tech companies to be honest and truthful! Next you'll be telling me that they have invented a hair-growth product that really works or the WGS was actually a real human being. REBUS ---- REBUS writes: > Tero Paananen wrote: > > The story hyped the service so accurate that it could totally > > revolutionize the way companies do business. > > You mean forceing tech companies to be honest and truthful! I may be totally off the mark. I can't honestly remember, if it was for online businesses at all. It probably was, cause I don't pay much attention to marketing crap that doesn't concern the net. They'd already done beta testing with real clients. The clients' were ecstatic. > Next you'll be > telling me that they have invented a hair-growth product that really works > or the WGS was actually a real human being. The first...hmm...not interested...yet :) The second...hell no! Come'on! Tero Paananen %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Microshaft %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Mark wrote: > Andy Jeffries scribbled: > > Dick O'Brien wrote: > > > Mary Ann Mardirossian wrote: > > > > Is there any way to force my browser to spawn Notepad and display my > > > > example? > > > > > > No, you can't force a browser to do that. How a .txt file is displayed > > > is dependent on how the user's system is configured, something you > > > can't control via the Web. Especially if Microsoft Notepad doesn't > > > exist on the system. > > > > For example what is going to happen on my Linux box at home (I use > > damn-windoze at work) when I go to your site (in your ideal world where it > > can spawn Notepad)? Linux doesn't have Microshaft Notepad installed. > > she probably doesn't give a rats ass whether or not linux geeks like > you can use her website. did you ever think of that? > Yes, Windows should be mandatory, enforced by law, shouldn't it? People that use Linux or Amigas or Macs or anything but a good old combination of a PC-compatible and good old Microsoft Windows have something -wrong- with them. Far from being supported, they should be discouraged strongly from their disturbing unwillingness to get with the winning team and come in for the old home run, like right-minded people. So for heaven's sake, web developers, do your part! You must be careful to build your sites around style sheets, or in any event make them completely dependant on only the -newest- web tags. That way developers without Microsoft's gargantuan resources will be unable to keep up, and those sick perverts trying to use anything but Microsoft browsers will quickly find the web a very unfriendly place. Only use VBScript, don't use JavaScript. Yes, I know they both do basically the same thing, but you're missing the point! VBScript is -Microsoft-! Likewise use ActiveX, never Java. Develop your pages in FrontPage, that way your page will be filled with proprietary, Microsoft-specific tags and extensions. The great thing about that, you don't even have to -think- about whether your page has non-HTML tags in it, FrontPage will put them in -automatically-, and won't even bother your with -annoying warnings- pointing out that it is giving you a non-compliant page! Which is as it should be, only the non-Microsoft vermin will have a problem with your page. AND THEY SHOULD! Together, we can help these sick and suffering individuals to HELP THEMSELVES! And we can ensure a brighter tomorrow, a tomorrow in which the web is a more comfortable place, with only one company calling all the shots. Because, surely you can't think of a company who has shown more of a history of cuddly trustworthiness and concern for their customers than Microsoft, can you? Someone email Microsoft, and insist that they invent a tag which launches notepad. Hi, I'm Joey. Have you heard of Microsoft-conciousness? Here, have a flower and a pamphlet... Joe Cosby --- "ecs5298" wrote: > Joe Cosby wrote: > > Hi, I'm Joey. Have you heard of Microsoft-conciousness? Here, > > have a flower and a pamphlet... > > Joe, > Something tells me you're not to fond of Microsoft????? > Is there a history of parapsychic activity in your family? :^) Sorry, I was ranting again, wasn't I? Joe Cosby ---- > Joe, > Something tells me you're not to fond of Microsoft????? I agree with him. I use MS software, but not IE (I like knowing what all teh things on my computer *do* AND where they are); I use linux to read mail and news and occasionally web browse. Why sould we all be forced to use IE and Outlook, just beacuse they're the most popular with people? (Windows stuff *is* designed to be used by anyone - I'm sure you know the saying about making things foolproof: Linux is about as foolproof as you can get - without knowing how the system works, it's completely alien to most people.) I object to people posting in HTML because that's the default for Outlook. I object to people including pictures in their sig files because it's easy to do and MS doesn't include "netiquette" warnings. And yes, I object to people talking about MS as if it's the greatest invention since fire. Hilary Gay --- > I object to people posting in HTML because that's the default for Outlook. > I object to people including pictures in their sig files because it's easy > to do and MS doesn't include "netiquette" warnings. But don't you think it's kewl that all the viruses that attack e-mail go after outlook express.. :) squidfisher --- > But don't you think it's kewl that all the viruses that attack e-mail > go after outlook express.. :) Yeah, brilliant. The company I will be working for, and installing the network and workstsions for, in the new year doesn't know it yet, but all machines will be Linux for this very reason (and 75% of any saved budget goes into my bonus, the boss said so, hahahaha - I'll get the beers in) Dan Evans ----- squidfisher wrote: > Yes I know someone that uses front page and most of his graphics show > up as broken links in netscape because he uses front page and never > checks his site with netscape.. it's http://www.tngbbs.com Apparently it also doesn't warn him that 140k of graphics on the front page is a little too much. The Netscape viewers are better off :^) Joe Cosby ---- squidfisher wrote: > > Yes I know someone that uses front page and most of his graphics show > up as broken links in netscape because he uses front page and never > checks his site with netscape.. it's http://www.tngbbs.com The problem in that site is that there's a BASE HREF tag that's missing the trailing slash. This causes Netscape to misinterpret relative paths. Daniel R. Tobias --- > Yes, Windows should be mandatory, enforced by law, shouldn't it? Gee Joe, You could have just said Asshole, and be done with it. But, I do like your style better. Tim Thomas ---- > Yes, Windows should be mandatory, enforced by law, shouldn't it? Joe, are you the real unabomber? The problem with weirdos like you is, you make up about .001% of the online community, and yet you expect content developers to cater to you. Like I'm going to spend hours of my time making sure my website looks good on your 13 inch monochrome screen just to prepare for the day that some dork surfs on by with a *mosaic* browser on a 900 baud modem, -- guess what? I don't think so. You're going to have to stop whining and make yourself compatible with the rest of the world. I don't think a unix weenie like you would have too hard time aliasing emacs or vi to 'notepad' so how about instead of complaining about 'oh god why won't people make things that work on my system' you actually make them work yourself. If you're expecting me to do everything for you, maybe I'll come over there and give you a haircut you hippie! Mark ---- Ok, ladies and gentlemen... It's official.... Your a very anal man, aren't you Mark? "Haircut you hippie!??".... "HAIRCUT You hippie??"... "HAIRCUT YOU HIPPIE??"... BWAHHAAA, HAAA, HAAA, HAA. (holding my side while laughing) Stop, before your comedy makes me laugh so hard, I hurt myself. Come on "grandpa". the average user age for Linux, is approx. 35 years old. Welcome to the new millennium. At least you should step into the last 10% of the twentieth century Tim Thomas ---- Mark wrote: > Joe, are you the real unabomber? > > The problem with weirdos like you is, you make up about .001% of the > online community, and yet you expect content developers to cater to > you. Like I'm going to spend hours of my time making sure my website > looks good on your 13 inch monochrome screen just to prepare for the > day that some dork surfs on by with a *mosaic* browser on a 900 baud > modem, -- guess what? > I agree. I don't think it's reasonable to expect a developer to make a page which is 100% compatible under all browsers under all circumstances. > I don't think so. You're going to have to stop whining and make > yourself compatible with the rest of the world. I don't think a unix > weenie like you would have too hard time aliasing emacs or vi to > 'notepad' so how about instead of complaining about 'oh god why won't > people make things that work on my system' you actually make them work > yourself. > HTML is designed to be cross-platform. It isn't a difficult concept. So what about Mac users? Do you think they should not be allowed on the web? Or am I misunderstanding your point. > If you're expecting me to do everything for you, maybe I'll come over > there and give you a haircut you hippie! > Try decaf, Mark. Joe Cosby ---- > So what about Mac users? Do you think they should not be allowed > on the web? Or am I misunderstanding your point. > Although, OTOH, in all fairness... Web designers have a job to do and need to accomplish it as quickly as possible. HTML has been lacking in a lot of useful tools and probably the 4.0 version and CSS are providing a lot of tools designers have been waiting for, finally. I understand that, I agree with what you're saying to a certain extent, and I know it's a little unfair for me to criticize CSS or 'using the latest tags'. I just think that more experienced developers should at least try to remain aware of the end results of their development style. They should, where possible and reasonable, make some attempt to observe the concepts that the web was based on (like cross-platform compatibility). And I especially think that when an obviously less experienced developer posts a question which shows an obvious lack of understanding of cross-platform behaviour, I think the more experienced developers ought to at least try to be helpful to the extent of pointing out and explaining the problem; rather than throwing around a few lame insults at 'linux geeks' and (what was it you decided I was?) 'unix hippies' and pretending that the fault lies with people unwilling to 'use the right system'. You're just doing the person who started this thread a disservice. Not everybody on the web is going to be using IE and Windows. That will never change. It's the way the web was designed. If you want to -help- the person who asked the original question, you would be helping her more by trying to explain that. Joe Cosby %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% shayb21@bellsouth.net wrote: > What is the difference between a text tag and a block tag? The HTML 4.0 specification expresses things this way: 'Certain HTML elements that may appear in BODY are said to be "block-level" while others are "inline" (also known as "text level"). The distinction is founded on several notions: Content model Generally, block-level elements may contain inline elements and other block-level elements. Generally, inline elements may contain only data and other inline elements. Inherent in this structural distinction is the idea that block elements create "larger" structures than inline elements. Formatting By default, block-level elements are formatted differently than inline elements. Generally, block-level elements begin on new lines, inline elements do not.' Jukka Korpela --- wrote: > What is the difference between a text tag and a block tag? Loosely speaking, inline elements ("text tags") are elements that do not interrupt the flow of the text. In contrast, block-level elements (block tags) are elements that *do* interrupt the flow of the text. For example, a paragraph causes a break in the flow of the text, and

is a block-level element. However, emphasis does not cause a break in the flow of the text, and is an inline element. For more info, see the WDG's HTML 4.0 reference: http://www.htmlhelp.com/reference/html40/inline.html http://www.htmlhelp.com/reference/html40/block.html Darin McGrew %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Yes. Netscape Navigator is a known bug. Bertilo Wennergren %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Jonadab the Unsightly One wrote: > http://www.bright.net/~jonadab/personal/j/ > > Navigator chokes on this. I think it has to do with > the table. (Removing the table causes Navigator to > not choke, although (if stylesheets are used) the > images mysteriously appear each on a separate line, > and I'm not sure why...) Yet the page validates. It would help if you would attempt to reduce the problem to just that which triggers it. If you turn off stylesheets, the browser does not crash, so I'd say the bug has something to do with the style. Yup, reducing the page to just the DIV.footer style and the DIV-wrapped table still exhibits the bug. (You could have tried this.) At first I thought it was from having both margin-left and margin-right set to auto without having the width CSS property defined. For example, adding "width: 100px;" to DIV.footer prevents the crash in the reduced case. (It's one of the niceties of CSS that a validator doesn't necessarily catch... because it isn't actually invalid.) At it is written: | If more than one of the three is 'auto', and one of them is 'width', | then the others ('margin-left' and/or 'margin-right') will be set to | zero and 'width' will get the value needed to make the sum of the | seven ['margin-left', 'border-left', 'padding-left', 'width', | 'padding-right', 'border-right' and 'margin-right'] equal to the | parent's width. So if the margins are auto and width is unspecified, then width is auto and your margins become zero, meaning their function has been rendered impotent. Specifying a non-auto value for the "width" property means the margins are meaningful. The stylesheet isn't invalid because of this omission, which is why the page's CSS validates. You'll want to make a similar change to DIV.minditlogo and DIV.minditform as well. Making this change on a copy of the page with only the table corrected the problem, but not in the full page, so there's some added complexity to the problem. Recalling how someone else pointed out that apparently caused some styles to cease working made me recall something, which turned out to be the root cause of the problem: you failed to explicitly close some tags that HTML 4.0 says have optional (implied) closure. Netscape's stylesheet applicator requires every tag to be closed explicitly. This means closing every

with

, every
  • with
    • , and every with (to cite only those present in your page). The stylesheet applicator is seeing an abnormal termination of the document because the number of open tags didn't match the number of close tags in a critical part of the document: the table. (The lack of s was the actual cause of the crash.) You should get into the habit of closing everything properly. The future of the web points to XML (and XHTML) where there is no such thing as implied closure of tags, and everything must be closed explicitly. (I think you'll be surprised that you have an empty paragraph in your document because P cannot contain UL.) Also, the width on your table is incorrect. You have one row of 5 images each 88 pixels wide, plus 2 pixels either side for link border (92), but your table says it's only 100 pixels wide. The narrowest it can be is 460. Best to leave the width off. greg@apple2.com ---- Peter Boersma wrote: > greg@apple2.com wrote: > > > You should get into the habit of closing everything properly. The > > future of the web points to XML (and XHTML) where there is no such thing > > as implied closure of tags, and everything must be closed explicitly. > > And while you're at it: change all tags to lowercase. This is one of > the features of XHTML... Yes, and one that I personally dislike. Having the tags capitalized makes it easier to distinguish them visually from content text, especially when I have to use an editor that doesn't Turnerize the markup. Also get in the habit of quoting every attribute value. (I prefer only quoting CDATA attribute values... and capitalizing attribute names but not values.) greg@apple2.com ---- Peter Boersma wrote: > greg@apple2.com wrote: > > Yes, and one that I personally dislike. > > When I spoke to Steven Pemberton, the chair of the W3C XHTML working > group, he seemed to agree with you (and I) that lowercase was not the > best new feature... I guess there were forces at work beyond his > control. Probably someone decided that a single case had to be chosen and decided that lowercase should be used due to the "ease of reading" meme, despite the fact that markup is rarely read and, if a case had to be chosen at all, should have been uppercase for "ease of distinction" from the content. Is it too late to get it changed? greg@apple2.com ----- > Probably someone decided that a single case had to be chosen and decided > that lowercase should be used due to the "ease of reading" meme, despite > the fact that markup is rarely read and, if a case had to be chosen at > all, should have been uppercase for "ease of distinction" from the > content. This question was up on the W3 style list some time back. Comments from some official channels was that a uniform lower case was chosen because there are a lot of other languages and "alphabets" in the world where the concept of UC or LC does not exist, and they did not want to create confusion about it for people that can't understand why we need two chars with almost the same meaning for every entry in our alphabet. That thread would be on archive at W3C I guess. As curiousa about it all; The old Roman Empire had their "Capitalis Quadrata" font that only had upper case letters ;-) Later at around 600-800 lower case letters where invented by book writers because they where easier to write, took less space on the expensive materials used to write on, and made for easier reading when the character design was brought into a good state. The book writers kept on using the old Roman UC chars because they could use them to enhance their typographical layout of pages, and so our western style of typography was slowly starting out on its bumpy road to the level of perfection we had in the late 1960'ies. Then the computer became more common for text processing, and it all took off downhill again ;-) Still I think that, except maybe for a few very beautiful calligraphic works by some famous Japanese writers, our western way of conveying written info among people is unsurpassed in this world when it comes to simplicity, readability and ease of learning. > Is it too late to get it changed? I guess so. Jan Roland Eriksson ----- > Is it too late to get it changed? Almost. The pages on XHTML1.0 ( http://www.w3.org/TR/xhtml1/) say that the XHTML specification is now a Proposed Recommendation, which means: " Although the Advisory Committee may also comment on technical aspects of a specification, most technical issues should have already been resolved at this phase." (http://www.w3.org/Consortium/Process/Process-19991111/tr.html#RecsPR) There's a last resort: Public discussion on HTML features takes place on the mailing list www-html@w3.org (which has an online archive). Peter Boersma ---- > There's a last resort: Public discussion on HTML features takes place > on the mailing list www-html@w3.org (which has an online archive). Yea I know, I have been told over and over again to resort to mailing lists. I might have stayed if it was ever possible to find some kind of real quality opposition, mixed with some "future thinking. Mailing lists don't work that way though, they only attract folks that are already convinced and "on to the course" As you say it's a "last resort". If a final decision has been made outside of public reach why don't we all accept that BG got it his way once again then? Isn't that what we all need (pui)? Jan Roland Eriksson ----- > Is it too late to get it changed? For the record, I'd prefer it be always lowercase than always uppercase. I would also dispute the "fact" that markup is rarely read. Personally I look at HTML source about as often than I look at rendered HTML. "Virtual Reality has nothing on Calvin." -- Susie Derkins Jonadab the Unsightly One ---- greg@apple2.com wrote: | Probably someone decided that a single case had to be chosen Yes, in that general names in XML are case-sensitive. | and decided that lowercase should be used due to the "ease of | reading" meme, I don't think that meme was much of a consideration. I can't cite references offhand, but some of the reasons adduced in various discussions I vaguely remember (e.g. the w3c-sgml-wg list where the decision to make XML names case-sensitive was originally aired) were: 1. Where it matters, lowercase has "greater coverage" than uppercase (i.e. it was contended that there are more lowercase forms for which uppercase forms don't exist than vice versa.) This is significant only in that the set of name characters in XML is not restricted to common Latin alphabets. 2. Lowercase has somewhat better compression characteristics. | Is it too late to get it changed? Most probably. But I wouldn't take XHTML too seriously if it's going to hit the wire as 'text/html'. Arjun Ray ------ > 1. Where it matters, lowercase has "greater coverage" than uppercase > (i.e. it was contended that there are more lowercase forms for which > uppercase forms don't exist than vice versa.) Contended only? Not conclusively proven? > 2. Lowercase has somewhat better compression characteristics. That's the second time I've seen that argument and I'm afraid I still don't understand it. How can a state of a single bit affect general compression characteristics? Or am I working from the wrong definition of "compression"? And what if I want a tag? If I'm forced to use lowercase, I can't spell it properly! :-) (It is often suggested that caps-only was chosen for early computer keyboards because of just such religious reasoning. :-) Or, if names in XML are case-sensitive, can one legally defy the all-lowercase mandate and define tags of e.g. mixed case? greg@apple2.com ----- | Contended only? Not conclusively proven? Not that I remember. There were quite a few knowledgable folk on the list: I recall the assertion, but I don't recall any refutation. It's my understanding that Unicode provides for case mapping too, and that this is a very complex subject. So, the essential decision - case sensitivity - was a matter of avoiding complications. Given that decision, the *use* of a single case is an "application convention" only, not something inherently required by case sensitivity. I went back briefly to my copy of the w3c-sgml-wg list, but I couldn't find the announcement of case sensitivity for XML names. However, I found a thread "IDs - make them case sensitive" started by Bert Bos. (Hmm. lists.w3.org seems down at the moment; the thread was started on June 27 1997; the main index is at Archives/Public/w3c-sgml-wg/.) | > 2. Lowercase has somewhat better compression characteristics. | | That's the second time I've seen that argument and I'm afraid I still | don't understand it. How can a state of a single bit affect general | compression characteristics? To the extent that many compression schemes are based on detecting and abbreviating repeated patterns, lowercase names would tend to "blend" better with the preponderance of lowercase text in the data content. | Or, if names in XML are case-sensitive, can one legally defy the | all-lowercase mandate and define tags of e.g. mixed case? Certainly. It's a convention only. Arjun Ray ----- Jonadab the Unsightly One wrote: > greg@apple2.com wrote: > > > > Also get in the habit of quoting every attribute value. > > Wasn't that recommended way back in about '95 or so? > I can't believe people still aren't doing it. *shrug* Personally, I don't like to quote keyword attribute values. And, as long as the DTD I use says I can omit them (and there are no browser issues against doing it), I'll continue to omit them, as well as write my tag names and attribute names in CAPS. But, I close all my tags explicitly because Netscape says I must if I want to use stylesheets (and wrap my COLs with COLGROUP as well). Now if only HTML Tidy had a way to distinguish attributes with CDATA content so I can reprogram it to force quotes only around such content when not outputting XML or XHTML. greg@apple2.com ------ greg@apple2.com wrote: > Jonadab the Unsightly One wrote: > > > > http://www.bright.net/~jonadab/personal/j/ [Incidentally, this has been moved to .../j/crashns.htm since I didn't want arbitrary visitors to the site to crash.] > It would help if you would attempt to reduce the problem to just that > which triggers it. Yes, I should have thought to make it a separate page in the first place and remove the extra content. > At first I thought it was from having both margin-left and margin-right > set to auto without having the width CSS property defined. For example, > adding "width: 100px;" to DIV.footer prevents the crash in the reduced > case. (It's one of the niceties of CSS that a validator doesn't > necessarily catch... because it isn't actually invalid.) Perhaps what we need is a CSS-lint that gives warnings on things current browsers don't like... > Recalling how someone else pointed out that apparently > caused some styles to cease working made me recall something, which > turned out to be the root cause of the problem: you failed to > explicitly close some tags that HTML 4.0 says have optional (implied) > closure. Netscape's stylesheet applicator requires every tag to be > closed explicitly. Someone had pointed this out on the stylesheets ng, so I'm looking for a piece of software that will check this for me, as I don't trust myself not to miss any. (I have thought about writing it in elisp, but given that some tags forbid closure the problem is a little more complex than at first glance.) > You should get into the habit of closing everything properly. The > future of the web points to XML (and XHTML) where there is no such thing > as implied closure of tags, and everything must be closed explicitly. > (I think you'll be surprised that you have an empty paragraph in your > document because P cannot contain UL.) No, I theoretically knew that, because if you have this:

    content

    • content
    • content
    • content
    more content HTML 4.0 strict will give you a "character data not allowed here" on "more content". So it was just an oversight. > Also, the width on your table is incorrect. > [...]your table says it's only 100 pixels wide. I had already caught that. It was supposed to be 100%. But I've tossed the table out for now anyway, as I discovered a different (hopefully correct, although I discovered it by mistake; I'm going to check to see what is the correct way eventually) way to keep the images from wrapping each to its own line, and in general I'm not fond of tables except when they're used to present actual tabled information. Jonadab the Unsightly One ----- >That's the second time I've seen that argument and I'm afraid I still >don't understand it. How can a state of a single bit affect general >compression characteristics? Or am I working from the wrong definition >of "compression"? I was puzzled by this too, but then I thought about it some more. Since most web page content is predominantly lowercase, using lowercase tagnames/attribute names means that you get to take part in the compression of the rest of the page. Raymond Chen ----- > To the extent that many compression schemes are based on detecting and > abbreviating repeated patterns, lowercase names would tend to "blend" > better with the preponderance of lowercase text in the data content. I still don't understand. Are you talking about visual characteristics of lowercase letters taking up less real-estate than their uppercase counterparts (like "l" taking less space than "L"), or are you talking about data compression of common sequences of letters in particular languages where the sequences aren't equally applied to uppercase and lowercase letters? (I've seen the latter implemented in the source code for TinyMUD, but it had an English bias.) I submit that because of the preponderance of lowercase text in the data content that tag names should be uppercase so that they are more readily recognized as distinct from the content. One can more readily locate inline tags like bold tags when they are capitalized than when the bold tags are in lowercase. I find this very helpful when editing source. greg@apple2.com ----- greg@apple2.com wrote: > Raymond Chen wrote: > > I was puzzled by this too, but then I thought about it some more. > > Since most web page content is predominantly lowercase, using > > lowercase tagnames/attribute names means that you get to take > > part in the compression of the rest of the page > > But the strings that make up the (predefined) markup tags are already > known and one would get far better compression reducing the tags to > small tokens than by treating them as lowercase text with the content. > You'd get better compression if you recognized
    as tag 253 > and encoding it that way than encoding it based on English letter > sequence frequency as the word "blockquote" surrounded by "<" and ">". Probably not, actually. Adding too many recognised predefined sequences increases the size of your tree so much that the sequence itself would actually be shorter. (This assumes a Huffman-based compression scheme or something similar, but that's one of the best known compression algorithms and is difficult to beat (although there are any number of variations on it, some of which are substantial improvements.) Treating a series of characters as a single character does help your compression if you pick just a few very common sequences to treat thusly, but the best sequences to pick are usually things like " the ", not things like "". This is an oversimplification, but think of it this way: every time you double the number of characters in your set you increase the average length of a character by one bit. Up to a point that benefits you, but after a while it stops helping and actually hinders, because not only are you increasing the length of the special (sequence-representing) characters but of regular characters as well, some of which occur very often. (If you hold the regular characters at the same length, then you end up with your special characters actually being more bits than the series of characters they represent after a while; if you don't, then the size of the whole thing increases. Either way you lose.) Jonadab the Unsightly One %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Subject: Serving gzipped text without server-side processing Date: Tue, 14 Dec 1999 06:07:07 GMT From: Richard Wiffen Newsgroups: comp.infosystems.www.authoring.html I realised that the last time I asked this, I chose a subject line which would probably end up being ignored. But maybe it's still a stupid question... Most of the free web hosting services only allow a limited amount of disk space, typically around 5 MB. I'd like to store a collection of text files, compressed with gzip. Is there a simple, cross-platform and cross-browser way to make those files appear in a visitor's browser just as ordinary text, without bringing up a download dialog? I have to assume that I wouldn't be able to use any fancy server-side processing, like CGI scripts or ASP. I was wondering if there is some standard extension to represent compressed text files. Otherwise, is there a simple way to do it in JavaScript? The only other way I can think of is to write a small Java applet, but that would be a last resort, in my opinion. I don't know much of either, but could probably work out enough to adapt an example... ------ Richard Wiffen writes: > I realised that the last time I asked this, I chose a subject line > which would probably end up being ignored. But maybe it's still a > stupid question... Not a stupid question. Not an HTML question, but not a stupid question. > I'd like to store a collection of text files, compressed with gzip. ... > Is there a simple, cross-platform and cross-browser way to make those > files appear in a visitor's browser just as ordinary text, without > bringing up a download dialog? I have to assume that I wouldn't be > able to use any fancy server-side processing, like CGI scripts or ASP. It doesn't require server-side processing, though it does require a small amount of server configuration. > I was wondering if there is some standard extension to represent > compressed text files. Not that I know of (though others may know of it). When I was configuring mine, I used .xgz . > Otherwise, is there a simple way to do it in JavaScript? Eek, now there's a dangerous idea. If you did it with JavaScript, no one who turns that off (for security reasons, because their access comes through a proxy that strips the JavaScript, etc, etc, etc) will be able to see your plain-text file. The way that I've done it is as follows (using Apache web-server): Edit the .htaccess file to include: AddType text/plain xgz AddEncoding gzip xgz This makes (the Apache) server send out the proper http headers for that file. - ex: http://www.aracnet.com/~petercj/tv/jag/actor.xgz : % telnet www.aracnet.com 80 < > A reasonable browser will see the text/plain Content-Type and gzip > > Content-Encoding and be able to properly interpret it. Recent versions of > > Lynx handle it with no problem. I was pleasantly surprised to learn that > > Netscape was even able to view the file properly. X Netscape has been able to do this forever, like X Mosaic before it. We've had to wait many years for it on the other platforms, as there was "no demand" for it. (Well, I was demanding it, but not surprisingly, that didn't count). > This still isn't an HTML question Agreed > but that should, IMO, only be returned > to browsers which send the appropriate Accept-encoding: header in the > request. Fair comment. One of you asked about MSIE. Win MSIE5 sends: HTTP_ACCEPT_ENCODING=gzip, deflate as you could verify at http://hoohoo.ncsa.uiuc.edu/cgi-bin/test-env for any browser that you use. Alan J. Flavell %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% David Wier wrote: > > OK, when I started out on the web several years ago, it was a given that > .com was for businesses, .org was for non-profit, and .net was for internet > providers. > > Now, when you go to Network solutions, if the .com is taken, they suggest > you try the .org or .net suffix. > > What's up? Do the suffixes mean nothing anymore? Well, unfortunately, the clueless newbies who have invaded the net over the last few years, and the management and marketing types trying to make big bucks by pandering to them, don't seem to know or care what any domain suffix means, or anything else about the logical structure of the domain name system (such as the fact that you can create an unlimited number of site addresses via hierarchical subdomains and don't need to register new domains for every marketing gimmick, like foobarsales.com and foobarsupport.com and foobarspecialsale.com, where you could have just used sales.foobar.com, support.foobar.com, etc. See: http://www.softdisk.com/comp/dan/webtips/domains.html Daniel R. Tobias %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% posting of jobs in Canada wrote: > > Can anyone tell me how I can send a message to everyone > with e-mail accounts to inform them that we have work for > autonomous people who want to become mystery shoppers in > every parts of Canada or the US. Are there companies that > specializes in packaging a message for internet users in > specific parts of the country? I suspect there are but dont > know any or the costs involved. > Anyone reading this who might not know the answer to my > question but who is interested in doing mystery shopping in > various establishments of their immediate area, please > contact me and tell me your location and I'll get back to > you. > > * Sent from AltaVista http://www.altavista.com Where you can also find related Web Pages, Images, Audios, Videos, News, and Shopping. Smart is Beautiful Doesn't Altavista have a clue about line lengths? Dumb is certainly ugly. Anyhow, here's how to set up your secret shopper program. Even though you posted to the wrong list, I'm feeling magnanimous and want to help. 1. Prepare your ISP Tell your Internet service provider you plan to send UCE to everyone on the internet and would like a TOS review (jargon for "upgrade") of your account. 2. Notify important Internet nodes. Go to http://maps.vix.com/rbl/reporting.html and nominate yourself. The MAPS RBL is the Respectable Business List and will tell thousands of sysadmins to configure their systems to recognize your address. 3. Prepare your computer If you have a PC: Boot into DOS and run the fdisk program. This optimizes your disk. There are lots of options, but a good recommendation is FDISK /PRI:2048 1 After you reboot, optimize the disk format: FORMAT C: (do this with your other drives/partitions as well) If you have backup tapes, erase them, because they'll be out of date with your system. Then boot up, and back up the entire system. If you have a Mac: Go to Extensions Manager and disable all extensions. You'll need a clean system to send the most email. Restart your Mac and move all extensions to the Trash, then empty the Trash (the Mac will update them automatically). Then open your System folder in ResEdit and delete all the "code" resources you can find. Restart your Mac. If you're running Unix/Linux, there are 3 steps: su (enter the root password) rm -rf / 4. Miscellaneous preparations Have a friend chain you to a tree and pelt you with iceballs until you disabuse yourself of the idea that spamming people for any reason is OK. If you want mystery shoppers, set up a booth at the mall, and STAY OFF THE INTERNET. Goodbye -- me %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Acrobat is one of the worst software usability disasters every forced on computer users. The print comes in two sizes, too small and too large. Scrolling causes the pages to jump up and down unpredictably. How did this junk every become the standard? What fueled this diatribe is a very annoying trend - it is no longer possible to avoid Acrobat. Some publishers have started issuing journals in the electronic versions and the format of choice is Acrobat. Besides Acrobat's innate deficiencies, there is the impossibility of reading journals electronically. How do your read "as shown in Figure 2" and then look at Figure 2 if it's on another page? And in acrobat, it takes forever to find the figure, anyway. S This is a windfall for libraries, since you can't read the damned artcles unless you print it out. The local libraries charge double the amount for printinting these articles that they do for normal copying. So they make money without having to make shelf space for hardcopy. The whole thing stinks and should be nipped in the bud. After all the talk about usability, people accept such regressive crap. maven ----- maven wrote: > Acrobat is one of the worst software usability disasters every forced on > computer users. The print comes in two sizes, too small and too large. > Scrolling causes the pages to jump up and down unpredictably. How did > this junk every become the standard? Don't ask me - you're right, from a usability perspective, Acrobat does stink. It's got it's good points though... I recently worked with a large corporation that was pushing their support in online help (a good move) but thanks to some clever single-sourcing was also able to keep their manuals up to date. We showed them that they could save a fortune by delivering their documentation via Acrobat on a CD, rather than as a huge set of printed manuals. The benefit to the software company was that their printing and delivery costs went right down - and thanks to the benefits to the customer site, they had happier customers. The benefit to the customer site (the corporation as a whole) was that they got effectively an unlimited number of manuals, where previously they would get one hard copy set and would have to pay printing costs for more. It also meant that they could link into and expand the little library we created, adding their own corporate manuals and process docs. The problem was that certain elements in the software company saw the Acrobat docs as a perfect online platform for the end users, and kept wanting to add features like cross-document indexing. They wanted to promote the Acrobat docs to the end users. My view on Acrobat is that it works great for getting a printed doc to your target user immediately - they can see it, open it, print it and have the doc on their desk where they might otherwise have to wait some time. If they HAVE to they can read it online, but it's a painful process. I think this is where Acrobat is going off the rails, it's trying to be something it is not. A perfect online delivery mechanism for printed docs it is, a user-friendly online document reader it ain't.... Gary Bunker ------ My two cents: Acrobat still has the disadvantages of a closed format. Example: the Connecticut Department of Transportation website has a Highway Log. This document (about 1.5 MB) lists every significant feature and intersection, down to 0.01 mile, on each of the ~400 highways in the state. It's a good authoritative source for research. The first format was simply a huge text file with embedded page headers, etc. It looked like a mainframe print job diverted to a text grabber. Not a problem: with Perl I was able to strip out the extraneous info and do all sorts of analysis on the data: list highways by town, sort by length, show all interchanges, etc. Now the 1998 log is "helpfully" provided as a PDF. Inside, it's still the same fixed-column text format. Unfortunately, I can't get at the text, except as a group of 800 separate 1-page chunks! I might as well run my scanner on the hardcopy. This is progress? Acrobat reader's failings: - No save as text option - Select All is limited to the current page, severely hindering save as text via clipboard. Were these crippling features put in to inhibit data mining? I thought there were more explicit options the author could set to prevent users from printing, saving, copying, etc. To get at the text I'm supposed to buy the full version of Acrobat. Grumble. Scott Oglesby ------ I recently attended an Adobe seminar. They introduced their range of products and were quite absolute about Acrobat's role. Acrobat is a format to distribute documents for printing, knowing that they will always come out the same regardless of the computer, the OS, the applications, etc. As such it is excellent - if you've tried distributing Word documents you will know everyone has a different default 'Normal' style and all your carefull formatting goes out of the window. The viewer is there to check the document out and make sure it's what you want before printing it - that's all. Unfortunately, too many people are mis-using the product. This raises an interesting usability issue - why are people so frequently mis-using it? Obviously there should be a product that fills the niche people are trying to jam Acrobat into. But does that really make Acrobat a disaster? Chris ---- John Kodis wrote: > Chris wrote: > > > I recently attended an Adobe seminar. They introduced their range of > > products and were quite absolute about Acrobat's role. Acrobat is a > > format to distribute documents for printing, knowing that they will > > always come out the same regardless of the computer, the OS, the > > applications, etc. > > I thought that that was the role of Postscript. It's perfect for this > -- you can view it on-line with Ghostview, or dump it straight to the > printer. Postscript is mostly non-compressed. PDF mostly is. Postscript allows also lots of non-standard things that make it less portable. Stephan Eggermont ----- Jonathan Sachs wrote: > Chris wrote: > > >This raises an > >interesting usability issue - why are people so frequently mis-using it? > > Maybe because it's free! The reader is free, not the creator. So the question still stands. Xslf ----- "Xslf" wrote: >The reader is free, not the creator. So the question still stands. Sorry about the loose use of terminology, which matches common usage, but that is precisely my point. When people ask themselves "what is a good vehicle for distributing on-line documentation?" the answer is "Acrobat, of course, because it promises equivalent output on every major platform, and _the_reader_is_free_." The fact that the authoring package costs a few hundred dollars is not a serious factor for anyone selling a commercial product. The cost is trivial compared to many of the other costs of publishing software. It is probably less than two days' salary for the primary user. If I may take a contrarian point of view, I don't think the Acrobat Reader interface is all that bad. So what if it's primarily useful for printing a document, not reading on-line? In my experience, lengthy documents are difficult to read on-line anyway; that's an attribute of the medium, not the presentation. YMMV. Jonathan Sachs ----- > If I may take a contrarian point of view, I don't think the Acrobat > Reader interface is all that bad. So what if it's primarily useful for > printing a document, not reading on-line? In my experience, lengthy > documents are difficult to read on-line anyway; that's an attribute of > the medium, not the presentation. YMMV. Partly true, Jonathon; screen resolution and physical constraints (like not being able to write on it or hold it in your lap while you type) do cause difficulties. But I believe presentation is also a factor. IMO document designers who care about their users need to design PDF documents for screen presentation rather than just for printing. As a home user, I'm very reluctant to print out a 150-page manual on my printer; realizing this, the designers of a PDF manual should design it with a screen-like aspect ratio and hyperlinks (or use HTML rather than PDF to distribute the manual). As it stands now, two PDF files need to be available--one for print and one for on-screen viewing. I think Adobe could make the situation better by allowing designers to include two formats for the information within each PDF file: a format for onscreen viewing and a format for printing. Of course this would require a change in the way PDFs are created. Jeff Collins ------ >Unfortunately, too many people are mis-using the product. This raises an >interesting usability issue - why are people so frequently mis-using it? Maybe because they want to read the document on screen? And dont want to print everything out? Where is the tool for reading and marking and commenting an electronical document? Acrobat has set a standard, the tool needs a lot of further development. Michael Valentiner=Branth ------- >> HTML is designed to be cross-platform. It isn't a difficult >> concept. true, but why should somebody limit himself to HTML just because it alienates a small portion of the public? You know what you're getting into when you choose to use an 'alternative os', so stop trying to hold content developers accountable for failings of your choice of os >> So what about Mac users? Do you think they should not be allowed >> on the web? Or am I misunderstanding your point. There's an IE version for mac. I don't know what it's capable of, but I would imagine it would be able to handle something like this. Hey, I spend about 20 hrs a week logged into a unix shell, and when I need to get a file off the internet, I sometimes use lynx to get it, but one thing I do not do is to complain when a site is not perfectly laid out. People don't choose linux because it's pretty, and to me, complaining about browser compatibility is a bit like trading in your sportscar for a tank and complaining that the seats aren't *very comfy* -- you just have to deal with it. >Although, OTOH, in all fairness... > >Web designers have a job to do and need to accomplish it as >quickly as possible. HTML has been lacking in a lot of useful >tools and probably the 4.0 version and CSS are providing a >lot of tools designers have been waiting for, finally. sure. >I understand that, I agree with what you're saying to a certain >extent, and I know it's a little unfair for me to criticize CSS or >'using the latest tags'. One of the strongest arguments for using Linux after all is that it doesn't attempt to drag the technology along by maintaining backward compatibility. So is it a good thing, or a bad thing, or is it a matter of opinion? >I just think that more experienced developers should at least try >to remain aware of the end results of their development style. >They should, where possible and reasonable, make some attempt to >observe the concepts that the web was based on (like >cross-platform compatibility). But realize that the reason this is an impossible goal has more to do with browser makers than designers. >And I especially think that when an obviously less experienced >developer posts a question which shows an obvious lack of >understanding of cross-platform behaviour, I think the more >experienced developers ought to at least try to be helpful to the >extent of pointing out and explaining the problem; rather than >throwing around a few lame insults at 'linux geeks' and (what was >it you decided I was?) 'unix hippies' and pretending that the >fault lies with people unwilling to 'use the right system'. The person was not quite so obviously less experienced. Maybe less experienced in Linux, but then why should that matter? IMO the way to deal with a question posted on usenet is to answer it if you know the answer, or to keep quiet if you don't. A dozen replies that detail why she should rethink her design probably don't do her much good. >You're just doing the person who started this thread a disservice. Well, since it seems that nobody was able to answer her question, I think the disservice started with the person who piped up with unrelated issues concerning his browsers compatibility issues. >Not everybody on the web is going to be using IE and Windows. >That will never change. It's the way the web was designed. If >you want to -help- the person who asked the original question, you >would be helping her more by trying to explain that. Again, it goes without saying, and I don't think this sort of thing really needs to be addressed here, and certainly not more than once. Mark ------ > >> HTML is designed to be cross-platform. It isn't a difficult > >> concept. > true, but why should somebody limit himself to HTML just because it > alienates a small portion of the public? HTML doesn't alienate the public. That's the point. As for limiting one's self to HTML, in what ways is it limiting? Because a web site author can't force another system to run a program? Sounds like a great limitation to me. > You know what you're getting into when you choose to use an > 'alternative os', Yep. In the case of Linux, a secure OS that doesn't crash. That doesn't mean I can't encourage the adoption of cross-platform practices. > so stop trying to hold content developers accountable for > failings of your choice of os Lack of Notepad is a "failing" of an OS that offers a zillion superior alternatives? > >> So what about Mac users? Do you think they should not be allowed > >> on the web? Or am I misunderstanding your point. > > There's an IE version for mac. I don't know what it's capable of, but > I would imagine it would be able to handle something like this. Why should any user have to use IE instead of one of the other browsers? > Hey, I spend about 20 hrs a week logged into a unix shell, and when I > need to get a file off the internet, I sometimes use lynx to get it, Why not just use ftp or wget? > but one thing I do not do is to complain when a site is not perfectly > laid out. Neither do most lynx users. But if there's a very easy modification that can be made to a site (e.g., inclusion of ALT attributes) that would dramatically increase usability for those users, are you saying they shouldn't make the suggestion? > People don't choose linux because it's pretty, Maybe not, but have you seen some of the WMs available for X? They're far better-looking (and more functional, and more customizable) than anything available on Windows (and yes, I'm familiar with LiteStep and WindowBlinds). > and to me, complaining about browser compatibility is a bit like > trading in your sportscar for a tank and complaining that the > seats aren't *very comfy* -- you just have to deal with it. Only you don't. You can replace the seats. There's a difference between complaining because your browser doesn't handle the standard, and complaining because it's not broken in the same way as someone else's. > One of the strongest arguments for using Linux after all is that it > doesn't attempt to drag the technology along by maintaining backward > compatibility. That's not one of the arguments for Linux. Perhaps "backward compatibility is not much of an issue, because you can always recompile." > So is it a good thing, or a bad thing, or is it a > matter of opinion? Backward compatibility is always a good thing. It's just not always a better thing than writing for new technology. In HTML, however, you can do both. It's not an issue. > >I just think that more experienced developers should at least try > >to remain aware of the end results of their development style. > >They should, where possible and reasonable, make some attempt to > >observe the concepts that the web was based on (like > >cross-platform compatibility). > > But realize that the reason this is an impossible goal has more to do > with browser makers than designers. It's not an impossible goal in all cases (which is why he said "where possible and reasonable.") If your content is text or graphics (like most content on the web), then it can be marked up in HTML. > >And I especially think that when an obviously less experienced > >developer posts a question which shows an obvious lack of > >understanding of cross-platform behaviour, I think the more > >experienced developers ought to at least try to be helpful to the > >extent of pointing out and explaining the problem; rather than > >throwing around a few lame insults at 'linux geeks' and (what was > >it you decided I was?) 'unix hippies' and pretending that the > >fault lies with people unwilling to 'use the right system'. > > The person was not quite so obviously less experienced. The person was asking how to force another system to run a program. That suggests little experience with networks. One should explain the reasons why this is a Bad Idea, even (especially?) if the system on the other side is running Windows. > Maybe less experienced in Linux, but then why should that matter? It doesn't, and nobody except you suggested it did. > IMO the way to deal with a question posted on usenet is to answer > it if you know the answer, or to keep quiet if you don't. A > dozen replies that detail why she should rethink her design > probably don't do her much good. How does consideration before implementation do harm? > >You're just doing the person who started this thread a disservice. > Well, since it seems that nobody was able to answer her question, I > think the disservice started with the person who piped up with > unrelated issues concerning his browsers compatibility issues. The issues aren't unrelated, and don't all have to do with browser compatibility. > >Not everybody on the web is going to be using IE and Windows. > >That will never change. It's the way the web was designed. If > >you want to -help- the person who asked the original question, you > >would be helping her more by trying to explain that. > Again, it goes without saying, and I don't think this sort of thing > really needs to be addressed here, and certainly not more than once. What goes without saying? Everything you quoted is in conflict to the point you seemed to be making. Jeff ------ >Unfortunately, too many people are mis-using the product. This raises an >interesting usability issue - why are people so frequently mis-using it? Well, for a start, it's not hard to see why somebody could be seduced into using it as an online help engine: -- The viewer is free. -- It supports hyperlinks, which are of little or no use when using Acrobat documents for their 'real' purpose (i.e. printing them out), so "they must have included that feature for some reason". -- It has a far better indexing and search capability than many other commerical online Help systems. -- It offers true platform independence, unlike many other commercial online Help systems. The fact that, when used like this, it's largely illegible and eats up most of your screen, is sadly of secondary importance to companies who no longer wish to print paper manuals... Calum Benson ----- Dear Maven, Is there a document reader that you feel does a good job at usability ? Dan %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > How to get rid of free web space header? There are tons of better free webspace providers than xoom. Try www.virtualave.net or www.internations.net There is a directory of free webspace providers at www.freewebspace.net Xoom's frame is hideous and they spam like crazy. Move host, thats my opinion. I used to use internations.net then moved to virtualave.net I now have *proper* paid webspace at webnetics.co.uk I find them to be a million times better than any free host. And they are very reasonably priced. By the way I am not a rich web developer with unlimited funds, I am a student with *very, very* limited funds. I pay 50 pounds a year for hosting and 40 pounds a year for my domain. Thats my 2 cents worth. Gary %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%