Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
First XHTML 2.0 draft published

By Stereo in Internet
Wed Aug 07, 2002 at 01:00:13 PM EST
Tags: Focus On... (all tags)
Focus On...

Three days ago, the W3C published its XHTML Working Draft. It introduces some interesting new concepts and gets rid of deprecated crap. Sjoerd Visscher already has an XHTML 2 page online.

So what's new? Here's what I've seen so far.


ADVERTISEMENT
Sponsor: rusty
This space intentionally left blank
...because it's waiting for your ad. So why are you still reading this? Come on, get going. Read the story, and then get an ad. Alright stop it. I'm not going to say anything else. Now you're just being silly. STOP LOOKING AT ME! I'm done!
comments (24)
active | buy ad
ADVERTISEMENT
  • The href attribute is now a common attribute, meaning that it can be defined on any element, not just <a>. Any element can act as a link, e.g. <object src="http://www.kuro5hin.org/images/header/kuro5hin2.gif" href="http://www.kuro5hin.org" />
  • As you can see above, <img> and <applet> have been dropped, replaced by the more flexible <object>.
  • New <DOCTYPE>: <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 2.0//EN" "http://www.w3.org/TR/xhtml2/DTD/xhtml2.dtd">
  • The new <nl> element is for navigation lists. It allows web site developers to specify navigation lists that will be supported by a default style and can be styled via CSS (sort of a non-scripting version of the popup navlists that you see all over the web today). Quoting the W3C's page:

    Navigation lists are intended to be used to define collections of selectable items for presentation in a "navigation" menu. A navigation list is required to start with a name element that defines the name for the list.

    On visual user agents, the default presentation behavior is as follows:

    1. The contents of the name element are presented.
    2. When the name element's content is selected, the navigation list's li element contents are displayed.
    3. If an li element contains another navigation list, that list's name's contents are displayed.
    4. If an li element has an href attribute, and that element's contents are selected, the link defined by the href attribute is followed.
    5. If the nl element is de-selected, it's contents are removed from the display.

    It is possible to change this default behavior through the use of style sheets. The behavior of navigation lists in non-visual user agents is unspecified.

    See the w3c's page or Sjoerd's page for nice examples.
  • The new <h> element and <section> element are for defining headers and sections, but <h1>, <h2>, etc. are not deprecated.
  • <br> is deprecated in favor of the new <line> element.
  • They've finally given up trying to distinguish between <acronym> and <abbr>.
  • You can have your precious target attribute back, if you promise not to use it.
  • <q> has been dropped, replaced by <quote>. Browsers are not supposed to put quote marks around a <quote>; it's up to the author to do that. This is different from the <q> tag in HTML 4 and XHTML 1.x (which IE never got right anyway).
  • The obsolete tags from HTML 4, e.g. <font>, are finally gone. While this may not look very important, it marks the death of HTML as we knew it. This will finally bring true separation between layout and content!
  • <form>, <frame> and <event> are replaced with xforms, xframes and XML events respectively.

The most interesting ideas yet are, in my opinion, the Navigation Lists, XForms and the href attribute working everywhere. The vanishing of the old formating tags clearly marks a breaking point. Keep in mind that this is still a working draft that'll have to be discussed, modified and corrected. Only part of the roadmap's objectives are covered in this draft. It will take time to publish the final draft, yet more to publish the specification and one or two years for all browsers to support it.

Sources:

W3C, XHTML 2.0 Working Draft

Mark Pilgrim, Changes in XHTML 2.0

Chris Mannall, Comments on XHTML 2.0 Working Draft

Shane McCarron, What's new in XHTML 2.0

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Poll
Which HTML version do you use?
o 3.2 or below 10%
o HTML 4.0 / 4.01 34%
o XHTML 1.0 / 1.1 43%
o XHTML 2! 3%
o Microsoft's tag soup 0%
o My own tag soup that works with Netscape 4 4%
o Other 0%
o What's HTML? What are you talking about? 3%

Votes: 132
Results | Other Polls

Related Links
o Kuro5hin
o XHTML Working Draft
o XHTML 2 page
o href attribute
o common attribute
o &lt;object &gt;
o New &lt;DOCTYPE&gt;
o &lt;nl&gt; element
o name
o li
o navigation list
o href
o nl
o navigation lists
o w3c's page
o Sjoerd's page
o &lt;h&gt; element
o &lt;sectio n&gt; element
o &lt;br&gt; is deprecated
o &lt;line&g t; element
o distinguis h between &lt;acronym&gt; and &lt;abbr&gt;
o target attribute
o you promise not to use it
o &lt;quote& gt;
o the &lt;q&gt; tag
o xforms
o xframes
o XML events
o W3C, XHTML 2.0 Working Draft
o Mark Pilgrim, Changes in XHTML 2.0
o Chris Mannall, Comments on XHTML 2.0 Working Draft
o Shane McCarron, What's new in XHTML 2.0
o Also by Stereo


Display: Sort:
First XHTML 2.0 draft published | 100 comments (87 topical, 13 editorial, 0 hidden)
This is great (3.44 / 9) (#1)
by DesiredUsername on Wed Aug 07, 2002 at 10:02:13 AM EST

Now when are people going to start using the standards? Nobody is using XHTML, so I think they are getting a little ahead of themselves publishing a new standard that builds o it.

Play 囲碁
I use XHTML, and so does microsoft at least [nt] (none / 0) (#4)
by boxed on Wed Aug 07, 2002 at 10:15:53 AM EST



[ Parent ]
Where does Microsoft use XHTML? (nt) (none / 0) (#26)
by ajf on Wed Aug 07, 2002 at 11:37:23 AM EST



"I have no idea if it is true or not, but given what you read on the Web, it seems to be a valid concern." -jjayson
[ Parent ]
msn.com for example [nt] (none / 0) (#29)
by boxed on Wed Aug 07, 2002 at 11:44:14 AM EST



[ Parent ]
well... (none / 0) (#51)
by calimehtar on Wed Aug 07, 2002 at 01:41:10 PM EST

They seem to be adding a virtual close-img like in xhtml, but they're not declaring a doctype... which sort of moots the whole point, doesn't it?

[ Parent ]
no doctype? (none / 0) (#61)
by boxed on Wed Aug 07, 2002 at 03:47:12 PM EST

then what is this text in the beginning of the html?

<?xml version="1.0" encoding="utf-8" ?><!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "DTD/xhtml1-strict.dtd">

[ Parent ]

Link? (none / 0) (#72)
by livingdots on Thu Aug 08, 2002 at 02:39:23 AM EST

*sigh* Not on the front-page. Why didn't you just link to the page where the damned code is?

[ Parent ]
it's on msn.com... (none / 0) (#73)
by boxed on Thu Aug 08, 2002 at 03:45:08 AM EST

I have this little feeling you're not using ie when viewing that page or that they have some dns magic going on.

[ Parent ]
Not too impressed (5.00 / 1) (#76)
by livingdots on Thu Aug 08, 2002 at 09:57:13 AM EST

Aha, you're right... I'm using Opera 6.04, and I forgot I was dealing with Microsoft here. Anyway, I've checked it out now -- identifying as "MSIE 5.0" -- and... well, I'm not too impressed:
  1. The document still doesn't validate
  2. The style sheet still doesn't validate
  3. They're still using proprietary "DHTML" and stupid browser-sniffers
Conclusion: The folks at Microsoft are in fact still the biggest idiots on this planet.

[ Parent ]
I don't know whether to laugh or cry (4.87 / 8) (#60)
by ttfkam on Wed Aug 07, 2002 at 03:33:47 PM EST

Two of the most popular open source community engines are trailing Microsoft. MS has closed tags and most of the trappings of XHTML minus a valid doctype (maybe because their layouts require browsers' quirks mode?).

Kuro5hin.org is using HTML 4.0 transitional

Slashdot.org is using HTML 3.2 final

What does this say about a programming community that (as a whole) is very vocal about standards compliance? On the one hand, we have MS with use of advanced standards with some errors (eg. no valid doctype) and two major community sites with old standards that retard the development of the web as a whole.

Before anyone pounces: no, I am not saying that Slashdot and K5 are retarding the web. I am saying that the standards they implement are doing the damage.

"Damage?" you say? By continuing to write new pages with old standards, there is no incentive for people to upgrade their browser, and you keep the technology stagnant. I'm not even talking about upgrading a browser once a year (which I would love by the way, but I digress), but upgrading the browsers that are a couple of generations old (I mean you Netscape 3/4--IE 4 holdouts!)

The advantage to the newer specs doesn't even mean that older browsers are completely excluded. Pages and their content are still viewable; the content is simply not as "pretty". Check out the article Why Don't You Code For Netscape? on A List Apart for a more complete argument. For that matter, if you're a web developer, check out the site anyway just because they have some excellent articles and sample code.

P.S. It started happening a while ago, but this comment makes me want to bring it up. After you've been working with XHTML for a while, it actually hurts to put in <p> tags without a closing </p> at the end. I'm tellin' ya. It just feels wrong. The same for the use <i> and <b> tags instead of a CSS declaration. It's like writing top-down code all of your life and then being introduced to functions and proper indenting style; At first it's weird and alien, then simpler to write with fewer errors, then simpler to write with less code, then simpler to edit after the fact, then just plain so obvious that it hurts to watch someone else write top-down, no-indentation code.


If I'm made in God's image then God needs to lay off the corn chips and onion dip. Get some exercise, God! - Tatarigami
[ Parent ]

On XHTML (none / 0) (#9)
by J'raxis on Wed Aug 07, 2002 at 10:20:51 AM EST

We are using XHTML. I myself am, also.

— The Raxis

[ J’raxis·Com | Liberty in your lifetime ]
[ Parent ]

I wish I had used XHTML. (3.50 / 2) (#10)
by Shren on Wed Aug 07, 2002 at 10:22:16 AM EST

I'm working with XML output from a database, and things would be so much easier if I didn't have to mangle XML into HTML on the fly. My code for adding size and maxlength parameters to <input> tags is a misbegotten hack, because (if I recall the problem correctly) <input size=10> isn't valid XML, but <input size='10'> isn't valid HTML (well it is, but it doesn't set the size).

As George Carlin once said, "There's an idea whose time has passed." We're still haunted by bad design decisions from the Mosaic era (of course, we get the benefit of what they did right, too, sort of like a celebrity deathmatch between Casper and The Ghost of Christmas Future).

[ Parent ]

well then (5.00 / 2) (#27)
by calimehtar on Wed Aug 07, 2002 at 11:38:11 AM EST

I suggest you start independently working on html 5 which makes blink and marquee standard, enhances font tags and codifies the proper way of writing javascript and html for ilayer/iframe simultaneously. Maybe add a few css crashing 'features' to the spec just for fun.

[ Parent ]
Many people making the switch to XHTML (none / 0) (#32)
by StephenFuqua on Wed Aug 07, 2002 at 11:51:30 AM EST

My most recent designs have been in XHTML (such as http://www.safnet.com/aguila). And many excellent sites are using XHTML. Check out Ask Zeldman, A List Apart, Glish, and others. Microsoft almost uses it--just look at all the proper />'s in their code! But they continue to use improper HTML like "leftmargin" and have no doctype. Silly Microsoft!



[ Parent ]
oh, forgot the most important: w3c! (none / 0) (#33)
by StephenFuqua on Wed Aug 07, 2002 at 11:54:34 AM EST

world wide web consortium uses XHTML of course...

The question is, why aren't you?



[ Parent ]
"Rated XHTML" essay (none / 0) (#81)
by Moebius on Thu Aug 08, 2002 at 12:52:04 PM EST

I'm sorry, I respect Peter-Paul Koch, the author of that essay (he's got a nice section about JavaScript (and CSS) on his homepage and seems to know a good deal about the subject), but I don't understand his grudge about XHTML. He argues there isn't any real benefit for the effort; but there's hardly any effort. It's not hard to close your tags, use lowercase and put parameters in quotes. Sure, XHTML doesn't add any spectacular capabilities, but it's good to bring HTML into better conformance with XML, making it easier to parse and process HTML. And the easier it is the parse (and parse correctly), the more likely it is the browser will get it right and your page will look exactly like it should. His complains "XHTML isn't about the present, it's about the future," but where else does every present come from? XHTML 1.1 is the basis needed to move forward to a more intelligently formatted web, a web that allows documents to be read and understood correctly and without ambivalence and lets HTML join the XML family as a legitimate member so browsers can easily process both HTML and the richer world of XML with the same engine. Perhaps when XHTML 2.0 comes out, with its new features, he will think it's new features will be worthwhile; so why not do the simple changes XHTML 1.1 requires and be ready for the future and better serve the browsers of today that already understand XHTML.

[ Parent ]
Not a good example... (none / 0) (#87)
by Holloway on Thu Aug 08, 2002 at 10:01:29 PM EST

They also use tables for layout... :)


== Human's wear pants, if they don't wear pants they stand out in a crowd. But if a monkey didn't wear pants it would be anonymous

[ Parent ]
hey, don't forget... (none / 0) (#35)
by calimehtar on Wed Aug 07, 2002 at 11:59:06 AM EST

CTV.ca. I'm not making promises it will validate, but that's what the doctype says.

[ Parent ]
Using XHTML 1.0 Strict and XHTML 1.1 (4.33 / 3) (#47)
by questionlp on Wed Aug 07, 2002 at 01:21:20 PM EST

Saying that no one is using XHTML is a bit of a broad generalization, isn't it? I've been using XHTML 1.0 Strict on my personal site for quite a while now... heck it even validated as XHTML 1.0 Strict! As of a couple of weeks ago, I updated the design a bit and switched it to XHTML 1.1, and it validates perfectly.

I also use XHTML 1.0 Strict and XHTML 1.1 on all of the internal sites that I build for my group. I've been gone as far as grabbing the W3 Validator tarball and set it up on an internal web server for others to validate their pages. So far, it's been quite popular amongst the web designers.

I do agree that many people don't care about HTML/XHTML standards and validation, but many people do. I care since I want to put content over style (although I do write valid CSS Level 2 style sheets and use them), it's a nice challenge, but the key one is that the site renders almost identical across Opera 5/6, Mozilla, IE 5.5 and 6, IE 5 on the Mac, and Konqueror (at least the one in KDE 3)... all without having to hack together separate versions of the same site. For browsers that aren't considered "compliant", I tell the script to replace the CSS2 style sheet file with one that is fairly trimmed down.

BTW - My personal site is listed in my sig.

-- http://closedsrc.org
[ Parent ]

Modularization (5.00 / 1) (#3)
by anaesthetica on Wed Aug 07, 2002 at 10:15:47 AM EST

Perhaps you could include a paragraph or two on why this particular release is important. For instance, removing of deprecated elements finally allows for true separation of style and content when XHTML is combined with CSS (which also was just recently given a revised publishing). Modularization allows different types of readers to access XHTML documents, displaying only the markup from modules that they can handle. A good topic and clearly explained otherwise.

—I'm the little engine that didn't.
k5: our trolls go to eleven
[A]S FAR AS A PERSON'S ACTIONS ARE CONCERNED, IT IS NOT TRUE THAT NOTHING BUT GOOD COMES FROM GOOD AND NOTHING BUT EVIL COMES FROM EVIL, BUT RATHER QUITE FREQUENTLY THE OPPOSITE IS THE CASE. ANYONE WHO DOES NOT REALIZE THIS IS IN FACT A MERE CHILD IN POLITICAL MATTERS. max weber, politics as a vocation


Added some content (none / 0) (#17)
by Stereo on Wed Aug 07, 2002 at 10:50:52 AM EST

I added some content about the vanishing of old tags. I think neither the W3C nor I have looked enough into modularization yet, I think there's enough to say about it for another article.

kuro5hin - Artes technicae et humaniores, a fossis


[ Parent ]
This is the way to go (none / 0) (#5)
by psychologist on Wed Aug 07, 2002 at 10:16:58 AM EST

First write out the standard, then use it extensively in the labs to see what really works and what doesn't, and when every field has been properly covered, all browsers move over to that standard.

On MSIE and the Q tag (4.00 / 1) (#12)
by J'raxis on Wed Aug 07, 2002 at 10:30:06 AM EST

MSIE for Macintosh was the only browser I’ve seen that actually got the Q tag right. The tag would even display locale-specific quotation marks; for example, using curlies if you specified a page language as English or guillemets (double angle-brackets) if you specified French. If I remember correctly, they nested properly, too (a Q within a Q would use single quotes instead of double).

— The Raxis

[ J’raxis·Com | Liberty in your lifetime ]

That sounds right (none / 0) (#20)
by mech9t8 on Wed Aug 07, 2002 at 11:21:05 AM EST

I guess they changed the functionality of the 'quote' tag so that quotations would show up properly on non-<q>-supporting browsers.

Seems like a step back. I would think one of the main benefits of a quote tag would be locale-specific marks, as well as leaving 'smart quotes' (opening/closing curlies) in the hand of the browser.

--
IMHO
[ Parent ]

Smart quotes (none / 0) (#30)
by J'raxis on Wed Aug 07, 2002 at 11:45:09 AM EST

Yep. And the only other way to do smart quotes is type entity gibberish like &#x201C; and &#x201D; which is just annoying.

— The Raxis

[ J’raxis·Com | Liberty in your lifetime ]
[ Parent ]

It is only a draft (none / 0) (#56)
by Freaky on Wed Aug 07, 2002 at 02:38:38 PM EST

And the first one at that.  We can always ask to have it back.  Pretty please.  With a cherry on top.

[ Parent ]
Smart quotes (none / 0) (#83)
by uhoreg on Thu Aug 08, 2002 at 03:03:33 PM EST

&ldquo; and &rdquo; (“ and ”) are supposed to do proper English-style opening and close quotes. &laquo; and &raquo; (« and ») for angle quotes (which is backwards for some languages).

[ Parent ]
ldquo, etc. (none / 0) (#88)
by J'raxis on Thu Aug 08, 2002 at 10:44:53 PM EST

Yes, but if a browser supports numeric entities, x201[89CD] are best. The [lr][ds]quo names are new, so even in browsers that support names like [lr]aquo, amp, lt, gt, [aeiou]acute, etc., they might not be supported.

— The Raxis

[ J’raxis·Com | Liberty in your lifetime ]
[ Parent ]

standards are great (3.60 / 5) (#13)
by asv108 on Wed Aug 07, 2002 at 10:32:19 AM EST

If people actually use them, I would love to see a figure on how many pages still use the <center></center> tag even though it was "depreciated" 4 years ago.

there are so many to choose from! (1.50 / 2) (#22)
by wiredog on Wed Aug 07, 2002 at 11:28:23 AM EST

Sorry. Had to be said.

Can't sleep. The clowns will get me.
[ Parent ]
oops (none / 0) (#82)
by jeffycore on Thu Aug 08, 2002 at 01:15:02 PM EST

so what replaced the <center> tag? i still use it because it's easier than using css in most cases.

[ Parent ]
What? (5.00 / 1) (#85)
by unknownlamer on Thu Aug 08, 2002 at 03:33:53 PM EST

<span style="text-align: center;"> is difficult? Or even just set up a new style like span.c { text-align: center; }; and then just <span class="c">. Text alignment is for losers anyway.
--
<vladl> I am reading the making of the atomic bong - modern science
[ Parent ]

I would love to see <span> disappear (5.00 / 1) (#95)
by Rizzen on Tue Aug 13, 2002 at 12:42:11 PM EST

What a hokey piece of crap that tag is.  All it does is take up space and make reading HTML docs a pain in the arse.

<center> has been replaced with the <... align="center"> attribute in most tags, and with CSS.  But, it's still a pain to have to type out nearly twice as many characters to achieve the same results.
The years of peak mental activity are undoubtedly those between the ages of 4 and 18. At age four, we know all the questions; at eighteen, all the answers.
[ Parent ]

Deprecated? or depreciated? (2.50 / 6) (#19)
by IHCOYC on Wed Aug 07, 2002 at 11:13:12 AM EST

Which is it?

Both words strike me as slightly comical, as if some committee is invoking curses upon, or insulting, the tags last year's committee created.

Those of us who are non-professional web designers, who learned mostly by looking at source and finding tags that did what we wanted them to do, find the whole business of "standards" either dismaying or a nuisance anyways. We learned to fiddle with things until they came out close to what we had in mind. We will continue to use what we learned regardless of what curses have been laid on its head by this year's version.

Pallida mors æquo pulsat pede pauperum tabernas regumque turris. --- Horace

Deprecated (5.00 / 2) (#23)
by wiredog on Wed Aug 07, 2002 at 11:30:59 AM EST

Deprecated is a common term for 'this feature shouldn't be used, as there is no guarantee it will be supported in the future, and we're getting rid of it as soon as possible'.

Depreciate is what happens when something loses monetary value.

Can't sleep. The clowns will get me.
[ Parent ]

deprecated (4.88 / 9) (#24)
by calimehtar on Wed Aug 07, 2002 at 11:31:24 AM EST

... and those of us who are professional find the existence of a standards body to be a tantalizing promise of consistency which has been constantly stolen from us by the browser builders and by hordes incompetent amateurs.

I'm quite sure we'll never be able to write a page or a script one and have it work the same way on all browsers, but the w3c is making an honest effort to forge html into something that is simultaneously easier to write and easier for browsers to interpret... in the hopes that it will be adopted widely and consistently. It does seem somehow futile to keep changing the spec since the majority of the internet still isn't written in a reasonable fascimile of 1997's html 4 spec, let alone any form of xhtml...



[ Parent ]
Often used poorly (4.00 / 2) (#25)
by 8ctavIan on Wed Aug 07, 2002 at 11:36:13 AM EST

The word depricated , from the verb to deprecate has taken on the meaning of no longer used or useful, although it really doesn't mean that exactly, according to Merriam-Webster's dictionary. It means to play down, to belittle and to disparage

The word depreciated, which comes from the verb to depreciate means to lose value, usually in a monetary sense - that's where depreciation schedules come from when companies buy big ticket items (machines, trucks)

The Online Dictionary of Computing says that deprecated is used to refer to an obsolescent feature. IMHO, I think the best word to use here would be antiquated. It refers to something that may be still around, as in the case of html tags, but don't offer too much to the programmer.

Just my 0.02 and probably nobody gives a s**t.


Injustice is relatively easy to bear; what stings is justice. -- H.L. Mencken
[ Parent ]

Deprecate / depriciate (none / 0) (#44)
by IHCOYC on Wed Aug 07, 2002 at 12:53:44 PM EST

Deprecate comes from Latin deprecari, which originally meant "to pray to avert something," like storms, or raids by the Huns. It took on the meaning "to censure or condemn" as an extension of this.

Depriciate was invented as the opposite of appreciate, in both its financial and æsthetic senses. It means "to reduce in value," or "to cast disesteem on" something.

It's easy to see how either word might get applied to these judgments about web coding.

Pallida mors æquo pulsat pede pauperum tabernas regumque turris. --- Horace
[ Parent ]

obsolescent would be better than antiquated (none / 0) (#70)
by lordpixel on Wed Aug 07, 2002 at 05:57:30 PM EST

I think obsolescent (meaning, almost, but not quite obsolete) is closer to what's desired than antiquated.

But we're stuck with deprecated now.


I am the cat who walks through walls, all places and all times are alike to me.
[ Parent ]

Hello, part of the problem! (4.66 / 3) (#39)
by Pac on Wed Aug 07, 2002 at 12:18:17 PM EST

[We] find the whole business of "standards" either dismaying or a nuisance anyways

Nice attitude. I believe in the first-half of the last century that was the norm in the automotive industry, in the electrical industry and even still, to some extend, in the railroad industry.

As it is, you even have Microsoft to supply your users with a tool that will hide your mental laziness. But in time, even Explorer will drop the "garbage in, something out" feature. If for no other reason, for security reasons.

So, keep complaining about those pesky standards but study them. You future job may depend on it.

Those students who have become one with the universe will be allowed to go on and become two with the universe


[ Parent ]
I don't design Web pages for a living. . . (none / 0) (#43)
by IHCOYC on Wed Aug 07, 2002 at 12:46:52 PM EST

. . . so I really don't have much to lose by refusing to pay heed to a committee's decrees about what is or isn't considered good form this year.

I trust the major browser makers not to release something that suddenly breaks thousands of pages of Web data, even if they are all written in a creaky, primitive, or ad-hoc version of HTML. After all, not every Web page needs to be revisited that often. That online version of the Æneid isn't going to need changed anytime soon.

I suspect that the threat to refuse support for once-valid (and still-working) tags is mostly a paper tiger for this reason. I have no real stake in this. I am interested enough to take some notice of these decrees. But just as some of us are happy with old software, some of us are also happy with old HTML.

Pallida mors æquo pulsat pede pauperum tabernas regumque turris. --- Horace
[ Parent ]

Ad-Hoc (none / 0) (#45)
by Pac on Wed Aug 07, 2002 at 01:11:11 PM EST

There is no problem with this, until there are too many cars in the streets and they start to kill people because each manufacturer has his own petty idea about where the break pedal should be and how the breaks themselves should be assembled. Or until the passengers suddenly get too angry because they have to exchange trains at every country border because the rails have different sizes (and believe me, Europe has many, many countries). Or, as we saw a few years ago, you had to spend loads of money to make Sally's workstation in a Token Ring network talk to Bill's workstation in an Ethernet network (and don't even think about it if Bill's server is a Netware and Sally's an NT).

What I am saying is that eventually even the Web will be regulated, because there will be enough money and time factors involved to make the annoying security of a standard more important than the developer freedom.

As we see in the construction business today, you may even be able to get away using sub-standard code or plainly non-standard code for small (usually personal) sites. But you won't be able to mantain a major site this way.

Those students who have become one with the universe will be allowed to go on and become two with the universe


[ Parent ]
HTML != browsers (5.00 / 1) (#78)
by upsilon on Thu Aug 08, 2002 at 11:31:03 AM EST

It's not that the old HTML is suddenly going to become invalid. Browsers will continue to render HTML 3.2 properly for years and years to come (unfortunately, maybe).

In short -- that HTML version of the Æneid will still be perfectly readable despite its use of non-XHTML2 tags, because it is not XHTML2, nor is it even pretending to be. Eventually, it may be obsolete, like a M$ Word 2.0 document, but that won't happen for a while...

So yeah, it is a paper tiger, because nobody is actually refusing to support older tags and versions of HTML.
--
Once, I was the King of Spain.
[ Parent ]

Actually... (none / 0) (#94)
by Jim Dabell on Mon Aug 12, 2002 at 04:52:39 PM EST

I trust the major browser makers not to release something that suddenly breaks thousands of pages of Web data, even if they are all written in a creaky, primitive, or ad-hoc version of HTML.

I think you should take a look at 4 reasons to validate your HTML.

Something that the document fails to mention is the number of pages that worked in nn4 but not nn6 (layers, embed, javascript problems, table layout problems...)



[ Parent ]
Why standards are good (5.00 / 3) (#41)
by QuickFox on Wed Aug 07, 2002 at 12:32:41 PM EST

With your method you get pages that work in one browser, maybe two. Visitors with other browsers will see ugly garbage. In their browsers your page doesn't look or work the way you want.

With standards you get pages that work in lots of browsers.

Give a man a fish and he eats for one day. Teach him how to fish, and though he'll eat for a lifetime, he'll call you a miser for not giving him your fi
[ Parent ]

The weakness of standards makers (5.00 / 1) (#71)
by IHCOYC on Wed Aug 07, 2002 at 09:07:39 PM EST

With your method you get pages that work in one browser, maybe two. Visitors with other browsers will see ugly garbage. In their browsers your page doesn't look or work the way you want.
Or, we look at þem wiþ a number of different browsers on different systems, and fiddle with it until it displays acceptably on each.

I could make a better way to spell English þæn þe prevailing system in about an hour or so. Many oþers before me have achieved exactly þis goal, and vainly proposed þeir reforms to þe people who actually write þe language. Almost all such proposals, and all of þem þæt really meant improvement, have sunk like rocks.

One problem is þæt þe installed base of English texts represents a constant dræg on þe acceptance of þe new system. Anyone who can read or write only þe new English will find clæssical English a foreign language. Þe spellings of clæssical English, wiþ all þeir etymological and false-etymological cruft, are at least equally difficult to users of all English dialects. Phonetic English will privilege one dialect at the expense of oþers. (God help us if þey choose þe speech of upper-clæss Brits as the basis!)

I suspect þæt þe installed base of old and standard-resistant HTML represents a similar counterweight þæt will resist reform by decree. Þere's just too much of it out þere. Þere are too many people writing it þeir own way.
--
"Complecti antecessores tuos in spelæis stygiis Tartari appara," eructavit miles primus.
"Vix dum basiavisti vicarium velocem Mortis," rediit Grignr.
--- Livy
[ Parent ]

<font> tags question (none / 0) (#36)
by maozo on Wed Aug 07, 2002 at 12:01:15 PM EST

I have no problem with CSS, and I do appreciate the tremendous advantages of separating content from formatting.

However, sometimes web pages are not structured enough to warrant a style sheet. Stupid example: "Billy's Very Own Web Page". Billy might want to go crazy with the fonts and/or colors... can he not do this using an XHTML2.0-conformant web page editor? Will he be able to do this without generating an unimaginably large and ugly CSS?

Of course he can (5.00 / 2) (#46)
by ajf on Wed Aug 07, 2002 at 01:14:52 PM EST

<span class="c1">this is enormous bright red text</span> with a suitable CSS rule for c1 is no more unimaginably large or ugly than an equivalent font tag.

If you want to use a lot of weird and wonderful style combinations, you can specify more than one class, and define each class so that they can be easily combined. You might end up with:

<div class="c1 c2">enormous bright red</div>
<div class="c1 c3">enormous bright green</div>
<div class="c3">normal size bright green&lt/div>

"I have no idea if it is true or not, but given what you read on the Web, it seems to be a valid concern." -jjayson
[ Parent ]

or even more ridiculously font-like (none / 0) (#49)
by calimehtar on Wed Aug 07, 2002 at 01:35:26 PM EST

<span style="color: red; size: large> ... etc.

[ Parent ]
Yay. (4.71 / 7) (#37)
by farmgeek on Wed Aug 07, 2002 at 12:06:27 PM EST

And maybe someday, the hordes using Netscape 3.0 will upgrade.

The problems aren't just the standard, or the browser, but the fact that people will not move to a more standards compliant browser.


Can I have some? (2.00 / 5) (#40)
by QuickFox on Wed Aug 07, 2002 at 12:22:32 PM EST

And maybe someday, the hordes using Netscape 3.0 will upgrade.

What are you smoking? Where can I get some? It must be very good.

Give a man a fish and he eats for one day. Teach him how to fish, and though he'll eat for a lifetime, he'll call you a miser for not giving him your fi
[ Parent ]

It's called (2.50 / 4) (#42)
by farmgeek on Wed Aug 07, 2002 at 12:39:54 PM EST

hyperbole.

But if you design a website for a large audience, you'll still end up having to go with either a kludge that works on most everything, or multiple versions of the same page.

I would have said hordes using Netscape 4.x, but I was trying to make a point without someone trying to turn it into a holy flame fest.

[ Parent ]

hordes be damned (5.00 / 6) (#68)
by ttfkam on Wed Aug 07, 2002 at 05:27:37 PM EST

Using XHTML and CSS will not make older browsers unusable (at least XHTML 1.x won't).  It will only make them less pretty.  XHTML mandates that folks put alt attributes on images for example.  It doesn't have fundamental "I won't display at all on older browsers" features.  The W3C bent over backwards and went out of their way to make this the case.

Having a site be usable allows for people to use older browsers and gives you the excuse to say that you didn't have enough time or budget to make Netscape 3 dressed up and pretty.  This is a far sight better than using innumerable font tags (and b, i, and u tags) which make maintaining a site hell in a handbasket.  Syndicate content with HTML 3.2?  Yeah...  right.

Besides, the "hordes" are using IE 5+.  They have CSS support.  Folks who still use Netscape 3/4 instead of 6/7 or Mozilla (or Opera or Konqueror) deserve to be ignored. (IMHO of course...)

If I'm made in God's image then God needs to lay off the corn chips and onion dip. Get some exercise, God! - Tatarigami
[ Parent ]

Misleading (3.50 / 2) (#86)
by Holloway on Thu Aug 08, 2002 at 09:51:01 PM EST

It's not just "less pretty", it's demonstratably less usable for non CSS-P browsers.

Although it varies by content, for most layouts the page will be less readable and less usable in these older browsers. This is because in these browsers the DIV content will pancake down the page.

For example, with the Kuro5hin homepage: the list-like left column will now be the first thing viewable. Below that area, two screens down on my monitor (1280*1024), the body of content would next appear, and then about 6 screens down would come the login box.

Now this is only for older browsers that don't do CSS-P but saying it's "less pretty" understates the issue.


== Human's wear pants, if they don't wear pants they stand out in a crowd. But if a monkey didn't wear pants it would be anonymous

[ Parent ]

Demonstrably false (4.50 / 2) (#89)
by ttfkam on Fri Aug 09, 2002 at 02:13:50 AM EST

One of the great features of CSS is positioning on a page.  The order of divs matters less in a CSS-styled page than a straight HTML page.  Note that this doesn't mean that everything has to be rigid.  It just means that the zen of web page design is evolving and web designers need to pony up to new skills.

Sure the divs will pancake, but there's no reason why the content section can't appear first in the document except the inexperience of the page author.

If I'm made in God's image then God needs to lay off the corn chips and onion dip. Get some exercise, God! - Tatarigami
[ Parent ]

Re: Demonstrably false (4.00 / 1) (#90)
by Holloway on Sat Aug 10, 2002 at 05:59:58 AM EST

It was illustrating that content that was once one or two screens would now be many more in length. You can rearrange the div's order in the HTML file but you can't have the site's title and navigation, some site news and a login box, sponsors (who probably want to be near the top) and the latest front-page stories onscreen anymore. This affects the readability of a page, and the usability - for these older non-CSS-P browsers. Content that was more readable at 15 words/line now stretches across the screen. I just think that calling this affect "less pretty" is an understatement. Surely this isn't debatable? I mean - this is what it does in older browsers.

With your "hordes be damned", the "deserve to be ignored", the "gives you the excuse to say..." you're not making a decision after your audience. You're making a decision for your own selfish reasons. Now depending on your site you may end up making the same choices anyway - but choosing because of "hordes be damned" or weaseling an excuse out isn't a reasonable way to get there.

ps. I've been doing CSS-P layouts for years now, but I only recommend them for certain types of pages. A typical homepage doesn't suit CSS-P as most of the time the function is to present many options at once. Later in the site... with lengthier pages, they tend to suit CSS-P.


== Human's wear pants, if they don't wear pants they stand out in a crowd. But if a monkey didn't wear pants it would be anonymous

[ Parent ]

Not at all (5.00 / 2) (#91)
by ttfkam on Sat Aug 10, 2002 at 09:38:35 PM EST

If you use PNGs instead of GIFs, older browsers will see much more compact content (alt content) -- space wins right there.

Now that was a lame example, I know. So let me present a better one: use span tags instead. They don't generate line breaks like divs do. Content that can go together will go together.

And what about accessibility? I of course mean access to braille readers and the like. This is where accessibility guidelines as specified by the W3C come into play. For these clients, text-to-voice readers and braille translaters, line breaks aren't going to slow the audience down nearly as much as gratuitous navbars and sponser info. If you place redundant content at the bottom of the page and primary page content at the top, most people don't have to sweat about scrolling down much at all; it's redundant and mostly ignorable info.

What about content syndication to other sites? You can make a dedicated XML feed of course, but with a clean XHTML document with CSS, syndication is quite simple -- far easier than with legacy HTML.

Sacrificing backward compatibility is indeed something to be avoided. But the sacrifice of forward compatibility shouldn't be ignored either. In the case of a personal homepage, CSS seems like overkill. But on a small, personal homepage, using CSS isn't all that hard and most likely isn't bound by investers, stockholders, and bosses. As you add more and more pages, sooner or later entropy sets in and everything must be rewritten in order to take advantage of new technologies or maintenance scenarios. CSS makes these transitions easier. Changing even five font tags is more time consuming than one CSS entry. Once pages grow beyond three paragraphs, CSS wins out. Once websites grow past two pages, CSS wins out.

And lest we forget, the only reason that CSS seems harder to us is that we've been coding in old HTML for so long. I defy anyone to tell me that using tables for complex layout is somehow inherently obvious or easier to someone who isn't already intimiately familiar with HTML tables.

With your "hordes be damned", the "deserve to be ignored", the "gives you the excuse to say..." you're not making a decision after your audience. You're making a decision for your own selfish reasons. Now depending on your site you may end up making the same choices anyway - but choosing because of "hordes be damned" or weaseling an excuse out isn't a reasonable way to get there.
Depends on your audience. If your audience is bound by accessibility guidelines (government contracts) or you just like the idea of letting visually imparied individuals use your site, it is not a "selfish reason." First of all, the hordes are using IE 5 and up. More than 90% of viewers out there can handle CSS. The ones that can't (wait for it...wait for it) can still use the site. It only looks less pretty. Honestly, if I am spending less time worrying about browser quirks and especially older browsers (which are just a collection of browser quirks) and spending more time writing content, I submit that the users of older browsers are still getting a better deal overall.

Newer browsers have more features, access more resources, and are more secure (scripting and SSL fixes). Presenting someone with a perfectly usable site (let me stress that point -- I am not about denying access to older browsers) that doesn't have a bunch of bells and whistles amounts to little more than a gentle nudge to upgrade the browser. Presenting people with a pretty page that looks the same in Netscape 3 as it does in IE 6 gives the false impression that the browser is up to date and sufficient. But I will have sunk hours upon hours into perpetuating that illusion instead of putting up useful information.

I say have the bank use CSS, look good in new browsers, look plain or bad in older browsers, be usable everywhere, and let the bank get on to its business which is not managing a website but rather letting me manage my money.

I say have the government use CSS and get on with presenting me with local ordinances and the like.

I say have "Jill's Web Page" use CSS and get on with whatever she wants to tell the world rather than fiddling with nested tables so that it will look better in Netscape 3.

You may counter with some statement to the effect of "but this is the real world" or something like it. I will preemptively reply that the real world is dominated by browsers with the ability to render CSS. At one time, IE 4 and Netscape 4 were dominant. Two or three years ago, I would have agreed with you. It's time to take the leap and leave the stragglers in, once again, a completely usable but aesthetically lacking world wide web. In the meantime, we'll gently remind them that their world will look better in a newer browser. This is not like the old "best viewed with Netscape" items. This is using common standards that everyone including Netscape and Microsoft agree upon as being a great idea.

Am I looking for the easy way out? Yup! Am I selfish? Maybe. Is it somehow more noble to continue to make nested tables for a rapidly diminishing audience while retarding the progress of newer clients and technologies? I certainly hope not.


If I'm made in God's image then God needs to lay off the corn chips and onion dip. Get some exercise, God! - Tatarigami
[ Parent ]

Re: Not at all. (2.50 / 2) (#92)
by Holloway on Sat Aug 10, 2002 at 10:37:31 PM EST

I say have the bank use CSS, look good in new browsers, look plain or bad in older browsers, be usable everywhere, and let the bank get on to its business which is not managing a website but rather letting me manage my money.
Ah.. so I assume you have findings that say a CSS-P layout is easier for a bank to manage than a tables layout? ;)

Look, there's no need to fluff out your argument. You seem unwilling to concede that there's more lost in older browsers than just prettiness ("can still use the site. It only looks less pretty") and I don't find that at all reasonable. You even bend your argument to say that information not viewable on-screen in an older browser wasn't important anyway ("it's redundant and mostly ignorable info").

choosing because of "hordes be damned" or weaseling an excuse out isn't a reasonable way to get there.
Depends on your audience.
I agree completely. However, look at your own words and decide for yourself whether you were basing a decision on your audience, or just saying "hordes be damned", "deserve to be ignored", or the "gives you the excuse to say..." in order to push through your own preferences. Again, you may arrive at the same decision anyway, but choosing because of "gives you the excuse" isn't sensible, and is trying to be sneaky. You're changing your story as you go.


== Human's wear pants, if they don't wear pants they stand out in a crowd. But if a monkey didn't wear pants it would be anonymous

[ Parent ]
Older browsers and usability (none / 0) (#98)
by unDees on Thu Aug 22, 2002 at 05:49:07 PM EST

Ah.. so I assume you have findings that say a CSS-P layout is easier for a bank to manage than a tables layout? ;)
I know that was said with a wink, but I would suspect that nearly anyone making Web pages (bank employees or otherwise) would find a CSS-based layout easier to manage than a tables layout. (imaginary tables-based design review: "Hmm, let's see... this cell has a rowspan of 5, but it's overlapping that cell over there by a 1-px margin in this browser and not that browser....")

I can at least give you a study with a sample size of one that says CSS is easier to manage.

My wife accesses her nonprofit's CSS-based Web site regularly in Netscape 3 (as mentioned in another post, but you'll notice I'm only directly plugging the site once). The site is just as usable in NS3 as in fancy-schmancy browsers like Mozilla and IE6. Granted, the site doesn't do anything complicated, but still....

decide for yourself whether you were basing a decision on your audience, or just saying "hordes be damned"
I don't think his own words were saying "hordes be damned." It looked more like, "hordes get a great-looking Website, and fringe wackos hell-bent on using Netscape 3 be damned." 'Course, as I've already mentioned, my wife and I have to sit in the fringe wacko seat once in a while... :)

Your account balance is $0.02; to continue receiving our quality opinions, please remit payment as soon as possible.
[ Parent ]
Truer than you know (none / 0) (#97)
by unDees on Thu Aug 22, 2002 at 04:16:06 PM EST

My wife's nonprofit has an old Power Mac, and the only version of Netscape that will run on it is 3.0--there's not enough RAM to run any later versions of NS. And we're not even going to talk about the version of IE that's installed there. *shudder*

Of course, Opera works on it--mostly. With the occasional system-disabling crash, long load time, and annoying "permanent beta" splash screen, she usually alternates her browser usage so as not to go completely nuts.

That having been said, there are lots of ways to use CSS for new browsers while still making your content (if not layout) available to those stuck with old browsers (read: volunteers at nonprofit organizations too cash-strapped for the luxury of a new PC).

Your account balance is $0.02; to continue receiving our quality opinions, please remit payment as soon as possible.
[ Parent ]

Oddly enough.. (3.25 / 4) (#38)
by Perpetual Newbie on Wed Aug 07, 2002 at 12:09:46 PM EST

Not ten minutes ago I was writing a bit about what should be done to HTML. It's far from finished, but I would break nearly everything down to two basic tags (<obj> for block context and <span> for text context) and have other tags be derived from these two basic types using stylesheets. All tags would be allowed to have href=, name= (or id=), class=, style=, type=, and src= (or data=) attributes, and all attributes would be accessible by stylesheet and vice versa making stylesheets and attributes essentially the same thing. With any tag able to have an href or id attribute, the <a> tag would be deprecated. I would also deprecate <br> and replace it with a &nl; entity.

Now I'll have to go read this and find out what they're actually doing to the language.



yes and no (4.00 / 1) (#48)
by calimehtar on Wed Aug 07, 2002 at 01:33:19 PM EST

As far as display goes, all tags essentially fit into these two display models, the inline and the block display (plus a third, the line box in CSS level 3). However, it is a fundamental necessity that tags somehow describe their contents.

Take a quote for example. A quote is conceptually the same whether it's preceded by '>' like in an email, surrounded by quotation marks (English or French or ...), or indented like a block quote. So all quotes should be indicated by a quote tag or something like it for the sake of data integrity. Then the presentation can be anything from simple display, to something more complicated like indented block quote with open quotes on all paragraphs and close quote on only the last. The display can be dealt with separately in the CSS.

This, in a nutshell, is the current direction of web markup.



[ Parent ]
semantic tagging (none / 0) (#84)
by uhoreg on Thu Aug 08, 2002 at 03:24:21 PM EST

<em>, <cite>, <code>, etc. are more meaningful (and easier to read) than <span class="emphasized">, etc. In reality, you only need kind of tag, because you can select block vs. inline display using the display CSS attribute, but noone in their right mind will want to use such a thing.

[ Parent ]
Question on links (3.00 / 1) (#50)
by PresJPolk on Wed Aug 07, 2002 at 01:40:58 PM EST

How, now, are we supposed to control how links look in our CSS, if a is being returned to its use as an anchor and anything can be made into a link?

CSS, of course (5.00 / 3) (#52)
by TOCie on Wed Aug 07, 2002 at 01:57:51 PM EST

CSS2 selectors:

element[href] { rule: here; }
element[href="http://kuro5hin.org"] { rule: here; }


More: http://www.w3.org/TR/REC-CSS2/selector.html

:)

[ Parent ]
*ahem* (none / 0) (#53)
by TOCie on Wed Aug 07, 2002 at 02:00:24 PM EST

CSS2, of course.  Yeesh.  You'd think that previewing four times before posting would catch something like that, but noooooo.

Oh, and need I mention that IE doesn't like selectors that much?  Mozilla currently has the best support for them.

[ Parent ]

Well (none / 0) (#57)
by unknownlamer on Wed Aug 07, 2002 at 02:41:39 PM EST

XHTML 2.0 hasn't been released either. So it doesn't matter if IE still can't do CSS right...I tried to do some CSS2 the other day; worked fine in Konqueror and Mozilla, but supposedly didn't work in IE (since I don't use windows...). My bright green blinking text of doom was so cool! (it was a joke on phragma, I site that my friend runs and everyone he knows posts stuff on).


--
<vladl> I am reading the making of the atomic bong - modern science
[ Parent ]
Ooh (5.00 / 1) (#59)
by PresJPolk on Wed Aug 07, 2002 at 03:27:17 PM EST

I had a feeling that something in either CSS 2 or 3 would be the answer.  Thanks.

[ Parent ]
The way we control <a>, I suppose (5.00 / 1) (#54)
by Stereo on Wed Aug 07, 2002 at 02:13:10 PM EST

Let's imagine we use the new <h> tag for links and want to color them. I suppose we'd control them this way.

h:visited {
   font-weight : bold;
   text-decoration : none;
   color: #c60;
}

h:link {
   font-weight : bold;
   text-decoration : none;
   color: #c60;
}

h:hover {
   color : #f90;
   text-decoration : underline;
}

h:active {
   font-weight : bold;
   text-decoration : none;
   color: #000;
}

kuro5hin - Artes technicae et humaniores, a fossis


[ Parent ]
Use :link and friends (5.00 / 3) (#55)
by Freaky on Wed Aug 07, 2002 at 02:36:30 PM EST

:link (or *:link, same thing) to select, ala:

:link {
  text-decoration: underline;
  color: blue;
  background-color: transparent;
}

:link:hover {
  ...
}

etc.

Of course you can specify exact tags and classes etc if you want different rendering for some elements.

[ Parent ]

Browsers & DTDs (5.00 / 1) (#58)
by DJBongHit on Wed Aug 07, 2002 at 03:17:23 PM EST

Ok, would somebody who knows more about how this stuff works internally want to clue me in on something? What exactly is the purpose of the DTD in the doctype header? I'm assuming that the DTD isn't enough information for an older version of some browser to know how to render XHTML 2.0, right? If so, I'm fairly impressed. But if not, and the browser has to be updated to support the standard anyway, what's the point of having to link to an external DTD?

Or am I just confused about how this whole thing works in the first place?

~DJBongHit

--
GNU GPL: Free as in herpes.

Two reasons I can think of. (5.00 / 1) (#64)
by paine in the ass on Wed Aug 07, 2002 at 03:52:38 PM EST

First of all it means you can design a DTD-independent validator; it simply looks at the DOCTYPE line to find a DTD to validate against, instead of having to have every conceivable DTD built in to itself.

Second, it allows customization. I know this is irrelevant if you're using straight W3C XHTML, but what if you're not always? Specifying the DTD in the DOCTYPE line lets you use alternate DTDs for different effects (much like specifying your own XML namespace with xmlns).


I will dress in bright and cheery colors, and so throw my enemies into confusion.
[ Parent ]

Another use for DTDs (5.00 / 2) (#67)
by lordpixel on Wed Aug 07, 2002 at 05:20:24 PM EST

Both IE 5 on the Macintosh and Netscape/Mozilla also use the DTD to switch on "hints" to their rendering engines. Some people consider this a bad thing, but by using a specific DocType you indicate to the browser how its supposed to display the document.

If you say its HTML 3.2, you get Netscape 6/7's best attempt to be a circa 199X Netscape/IE 4 browser.
This is called 'quirks mode'

If you say its HTML 4.0 you usually get quirks mode, but if you use the exact right variant of the Strict DTD declaration, you get 'Standards Mode'.
Similarly, you get 'Standards Mode' if you specify XHTML in the DTD.

This lets the page author decide whether s/he liks <font> tags and layouts which depend on bugs in the old Netscape/IE implementations (<UGH>), or whether s/he wants the closest thing to the written spec that the browser can deliver.

Presumably the XHTML 2.0 doctype will one day signal to browsers that its time to enable <nl> support, or to treat <quote> differently than they used to.

 

I am the cat who walks through walls, all places and all times are alike to me.
[ Parent ]

Use for DTDs (5.00 / 1) (#80)
by christianlavoie on Thu Aug 08, 2002 at 12:19:18 PM EST

XHTML documents are just XML documents, you know?

As such, XML tools such as: validators, well-formedness checkers, 'magically better XML-supporting' databases (just don't buy the hype), and such can work on them with only the information available in the DTD.

So, for example, you could setup an XML validator program that can validate both XHTML and SVG, without ever knowing either of them... just reading the DTD, figuring out that the document fits the DTD and thus giving it thumbs up.

Go one level higher, and the DTD makes sense.

Serious browsers could always try to make a 'best effort' to render XHTML they don't understand if they see the DTD isn't so far from something they know -- but that's really far out and hair-splittin usage.

Have fun,
Chris


Maybe Computer Science ought to be taught in the school of Philosophy
   -- Christian Lavoie [modified from RS Barton]
[ Parent ]

promice not to use it (1.87 / 8) (#62)
by delmoi on Wed Aug 07, 2002 at 03:51:29 PM EST

Mark Pilgrim can go eat a fig. I like my pages to open in new windows so I can save it to read later if I'm in the middle of some content. I don't want to interrupt what I'm reading, read something else, then go back. And I don't want to scroll back up to the link when I'm done.

And the fact of the matter is, sites go down and the back button doesn't always work. I'd rather have some idiots be confused then have to deal with never getting to finish what I was looking at.
--
"'argumentation' is not a word, idiot." -- thelizman
re: opening windows (5.00 / 5) (#69)
by ttfkam on Wed Aug 07, 2002 at 05:34:22 PM EST

No one is saying that we should petition browser authors to get rid of "Open in New Window" functionality.  The point was to avoid it in the page markup.  If you want to open a link in a new window (or new tab), by all means.  The W3C is merely suggesting that page authors not make the decision for their viewing audience.

If I'm made in God's image then God needs to lay off the corn chips and onion dip. Get some exercise, God! - Tatarigami
[ Parent ]
promise not to use it (1.55 / 9) (#63)
by delmoi on Wed Aug 07, 2002 at 03:51:37 PM EST

Mark Pilgrim can go eat a fig. I like my pages to open in new windows so I can save it to read later if I'm in the middle of some content. I don't want to interrupt what I'm reading, read something else, then go back. And I don't want to scroll back up to the link when I'm done.

And the fact of the matter is, sites go down and the back button doesn't always work. I'd rather have some idiots be confused then have to deal with never getting to finish what I was looking at.
--
"'argumentation' is not a word, idiot." -- thelizman
User agent's perogative (5.00 / 9) (#65)
by BrianHV on Wed Aug 07, 2002 at 04:09:03 PM EST

I like my pages to open in new windows so I can save it to read later if I'm in the middle of some content. I don't want to interrupt what I'm reading, read something else, then go back.

So set your browser to do that. For example, I have mozilla configured to open links in new background tabs when clicked with the middle button. But I hate when web designers think they know what I want and open new windows for me. If I wanted it there, I'd ask for it to be there.



[ Parent ]
There are Valid Uses... (5.00 / 1) (#100)
by MyrddinE on Thu Sep 19, 2002 at 05:59:40 AM EST

... such as help windows. Especially on things like, oh, the 'Allowed HTML' link right on this site just below the Preview button. It is better usability to open a new window when the new page is created specifically to help the user with the current page. This is especially true when they are filling out a form. A similar situation occurs with 'click here for our privacy policy/user agreement/requirements' links when a user is signing up. The common thread is that the user is filling out a form, and needs to open this new page before they can finish filling out the form.

In other situations, I agree that the user should have control. However, when opening up a help window it is better to open in a new window to prevent the user from unintentionally wiping out their half finished form.

[ Parent ]

Read what Mark says about it! [n/t] (5.00 / 1) (#66)
by Stereo on Wed Aug 07, 2002 at 04:23:20 PM EST


kuro5hin - Artes technicae et humaniores, a fossis


[ Parent ]
If XHTML 2.0 is still in draft, (none / 0) (#74)
by buglord on Thu Aug 08, 2002 at 04:15:24 AM EST

then how come the test page renders correctly?

I never new that my Mozilla already knows about the new tags. And what about older browsers - what will they render?

I'm happy so much now I know how to use a gun!
Die Technik bereit und stabil... wir wollen zurück ins Telespiel!
welle:erdball - telespiel

It doesn't exactly render correctly (none / 0) (#75)
by Stereo on Thu Aug 08, 2002 at 09:07:14 AM EST

The content on that page that is compatible with XHTML 1.1, which mozilla supports, is rendered correctly. The <nl> tag, for example, isn't.

kuro5hin - Artes technicae et humaniores, a fossis


[ Parent ]
Plus some features can be specified using CSS (none / 0) (#77)
by Freaky on Thu Aug 08, 2002 at 11:12:19 AM EST

Even <nl> can be done using CSS, as can be seen on CSS/Edge

[ Parent ]
The point in <nl>... (none / 0) (#79)
by Stereo on Thu Aug 08, 2002 at 11:37:29 AM EST

...is avoiding such css tricks :)

kuro5hin - Artes technicae et humaniores, a fossis


[ Parent ]
Tricks? (none / 0) (#93)
by Freaky on Sun Aug 11, 2002 at 06:52:18 AM EST

That's not a trick, it's how I would expect <nl> itself to be implemented.  Generalisation is Good.

Mozilla and co will most likely implement <nl> in CSS in a similar way, somewhat like:

nl nl {
 display: none;
}

nl li:hover > nl {
 display: navigation-list;
}

You don't get much less trickier than that.

[ Parent ]

distinguish between <acronym> and <abbr&g (4.50 / 2) (#96)
by OzJuggler on Wed Aug 21, 2002 at 09:38:53 AM EST

To quote from the spec:
While some dictionaries define an acronym to be just a word formed from the initial letters of other words, others require the acronym to be pronouncable as a word.
The Others have got it right. If it isn't pronouncable, it isn't an acronym. Don't argue with me; I didn't invent English. I'm just telling you how it is. Even I get it wrong sometimes. This is not a troll!

Besides, the issue of whether an acronym must be pronouncable is kind of irrelevant when you consider that acronyms (even "unpronouncable acronyms") are abbreviations. That is to say, an acronym is clearly a shortened form of another phrase and it is therefore an abbreviation too.

So why the hell they didn't choose only <abbr> and get on with it instead of pandering to the illiterate is quite beyond my understanding. The tightwads at the W3C are usually sooo picky about everyone else conforming to their own latest standard. Why couldn't the W3C follow standard English? :) Have they not heard of the Oxford English Dictionary - the only authoritative source on English that there ever was? You won't find it on the web, but luckily their colleagues in Cambridge speak the same language.

But I digress. If you have <abbr> then you don't need <acronym>.
"And I will not rest until every year families gather to spend December 25th together
at Osama's homo abortion pot and commie jizzporium." - Jon Stewart's gift to Bill O'Reilly, 7 Dec 2005.

I can pronounce "XHTML" as a word... (4.50 / 2) (#99)
by unDees on Thu Aug 22, 2002 at 05:54:57 PM EST

...but I have to wipe off my monitor afterward.

Your account balance is $0.02; to continue receiving our quality opinions, please remit payment as soon as possible.
[ Parent ]
First XHTML 2.0 draft published | 100 comments (87 topical, 13 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest © 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!