Slashdot Mirror
W3C Announces XHTML As Its Recommendation
miester writes "Since I haven't seen anything about this on Slashdot I thought I might submit it. W3C has officially recommended that XHTML Basic be the next step for the World Wide Web. Just when I learned how to do tables ...."
Well, the W3 recommended the CSS1 spec, and IE, Netscape6, Mozilla, and Opera follow it religiously. Hell, most of them support some of CSS2. Most of the "standards" people talk about being unsupported are, in fact, unfinished. While the base XML standard was solid for a while, other standards that were needed to make it web viable, such as XSLT were not (although I think it just became "stable" recently). No browser vendor in their right mind is going to implement something that is subject to change.
Also, some of the standards are a bit "ambitious", and are therefore hard to implement without breaking backwards compatibility. Some are just a pain to code into a browser (like CSS2's drop shadow effect that should be able to affect any text it is applied to). Sure, someone like MS that has a "Microsoft Universe" to work with will implement silly browser specific stuff, but they will still support standards. Arguably, standards, in part, made IE come out on top in the browser wars. Sure, aspects of their monopoly put pressure on Netscape to make stupid mistakes, and also probably cut down on their overall workforce, slowing them down. The main aspect is that people will always want the web to work better, and standards do just that. The browser makers will cater to that desire or they will perish. Old Netscape died because it didn't move with the times fast enough.
"We obviously need a new moderation category: (-1, Woo-fucking-hoo)" --Mr. AC
Honnestly, look at W3C's own homepage and see for yourself what clean HTML means. Already since the first XHTML draft was released almost a year ago, they have simplified their presentation, but the important stuff is still there: the documentation about WEB Publishing standards.
Sure, there are colors and neat formatting tricks, but most of that is done using CSS. Oh, there are a few icons and logos, too. Is the absence of frames and nested tables an obstacle to the content's diffusion? Well, admit it, it is not. All we need to find from W3C is there: the standards.
Also, while the obvious intention of XHTML Basic is to reconcile WML and HTML into a unified common base, note that Web Content created using this barebone standard with linked CSS sheets (as opposed to embeded style rules within the text) also has another strong advantage: it obsoletes the very concept of forcing people to "upgrade" to whatever latest version of a specific browser, which increases accessibility of the content and makes it possible to use "deprecated" browsers without loosing anything significant.
My own appreciation of XHTML Basic, from an HTML comparision point of view, is that it is about half-way between HTML 2.0 and 3.2: basic text and images, plus forms and tables, with XML rigor as a bonus. If your Web Authoring really is about content, not Flash animations demo, then XHTML Basic is all you really need.
--
Software is not supposed to be about how to work around a useability issue. - Ken Barber
Look deeper -- Openwave (nee Phone.com), Ericsson, Sun, Microsoft and other key members of the WAP Forum have representatives on the XHTML Basic working group. Also of note is that a representative Access Co Ltd of Japan -- the developer of the microbrowser inside the popular i-mode phone, and another member of the WAP forum -- is listed as one of the editors of the spec. From this list of contributors, it should be clear that xHTML basic is going places.
And, unfortunately, it's not a problem that can be solved with "Well, we'll put up a syntactically and conceptually correct website that's going to be rendered problematically, or not at all, on most of our client's browsers, in the hope this will encourage the producers to actually get it right"...
Open Source. Closed Minds. We are Slashdot.
"But Lynx CAN display images. I run it in an X-window, and have it configured to fire up xv for each image on the page."
To be polite, why the fuck would you do this?
Open Source. Closed Minds. We are Slashdot.
Eh, my main problem is how they deal with CSS, especially NS4.7 and its refusal to let CSS persist through table cells.
As for who is using NS4.7... well, my school, for one. Every flaming Mac and PC on the network has NS4.7 and, unless a student has installed somethign else, only NS4.7 I think I've talked them into including NS6 in the next network iamge, but we'll see.
-J
Karma: T-rexcellent.
I am psyched about XHTML, but especially the ability to MOD it for specific applications. That's what makes XHTML Basic so *VERY* cool. I also know that whatever "Basic" spec we come up with, there will always be at least one feature somebody can't do without.
The feature I can't do without is forms.
All I want to do is take 1980's technology and make it interact with the web. If I could put my entire address book on a calculator watch, why can't I use a similar watch to exchange addresses, phone #'s via the web? Back then, I was doing a lot with a little four line LCD screen.
I think the spec is missing a very important piece of functionality -- a little interactivity.
Hey, you! Have mercy with the poor operators of that little site! Now thousands of people will follow the link, and the small Slashdot server will go down on its knees. It's Christmas, damnit!
The XHTML Pascal working group will be crushed.
Treatment, not tyranny. End the drug war and free our American POWs.
See my user info for links.
For table cells, can it inherit font specs? Or do I stil have to specify the font you want to use for each table cell?
I'm not sure if a move toward XML is necessarily a good thing. I hope I don't sound like a modern Luddite, so please hear me out.
/>" and "<HR />"), but we shouldn't use it for <P> or various other tags which normally had an end tag in HTML because it might screw up some browsers...? Can they even get the XML-ization of HTML right?
HTML and most browsers allow a large "fudge factor" -- your tag names can be uppercase or lowercase, you can quote your tag clause values if you want or not (like <P ALIGN="RIGHT"> or <P ALIGN=RIGHT>), you can use the header tags (<H1>...<H6>) in any order, etc.
In HTML, you can develop your own coding style.
I prefer uppercase tag names and tag-clause names, quoted tag-clause values for strings, nonquoted tag-clause values for numbers, free arrangement of headers.
Along comes the W3C, the "trusted" authority on browser standards and forces XHTML on us which has incredibly rigid syntax. Now it's all lowercase tags, all quoted tag clause values. We're supposed to adopt the single-tag XML syntax for <BR> and <HR>, etc. (like "<BR
There is even an "ISO HTML" standard which demands the header elements always be in "proper" order. I find it's much easier to whip out <H1> or <H2> when I need big text than going <FONT SIZE=+2> or defining a new style sheet class. Besides, the W3C doesn't want me using <FONT>, anyway...
There's an ANSI C standard, but it doesn't dictate that you always use the "one true brace" style, does it? I was taught code differently. But at the same time, I respect those who use the "one true brace" style.
Also, while we're talking about small, less-capable devices like PDAs, etc., how about people with lower bandwidth. I always try to make my HTML as compact as possible within reason. The damn picky Validator told me I had to add a TYPE="text/javascript" clause in my <SCRIPT LANGUAGE="JavaScript1.2"> tags because the LANGUAGE clause was deprecated. That's stupid. First, due to differences in implementation of JavaScript between the various versions, it's necessary to say LANGUAGE="JavaScript1.2" to tell the browser "execute this code using JavaScript 1.2 rules". I don't think there's such a MIME type as "text/javascript1.2" or "text/javascript1.1". Secondly (and this was the point I wanted to make), they're just asking for more bytes to be added to size of the page. I have many <SCRIPT> tags on this one page I'm working on. I want it to load quickly for dial-up users, and I'm supposed to make it larger? The TYPE clause above and separating space is only 23 bytes, but with many tags, that adds up.
Someone else was talking about <B> and <I>, etc. being dropped and being replaced with style sheets. (I haven't read that much of the specification so I don't know for sure.) Apart from the fact that style sheets are harder to learn, this (once again) slows down loading times because of the extra request from the client to fetch the style sheet.
And, with XHTML, every tag clause value needs to be quoted -- even numeric ones? That's unreasonable!
I realize we can't have anarchy and a standard is needed, but I don't think the W3C is totally in touch with what the Web has become and the business around it. If I didn't know better, I'd say they were trying to bring us back to 1993 with all pages being really basic and simple and all <H1> and <H3> elements having <H2> (not <h2>!!) elements between them.
What if a company has a standard coding style of uppercase elements, nonquoted tag-clause values? Are they forced to change?
Sure we use tables, <FONT> tags, and transparent GIF spacers like nuts. I'm sure it upsets Lynx users. But perhaps we're not catering to Lynx users. Most visitors to my pages are using IE 5.5, 5.0, or Netscape 4.51 or higher. Less than 0.1% are using Lynx. In business, you want an attractive page, and if that means making it difficult for one or two people so you can get more business from several thousand, so be it.
My apologies to those who prefer to code with lowercase tags, etc. We're all entitled to our own styles...until we get to XHTML.
Note that XHTML Basic is a stripped-down version of XHTML for phones, etc. It's meant to be the future of WML, not the future of Netscape (uh, IE, Opera, whatever).
As for tables, XHTML Basic includes tables, but simplified ones as necessary for reduced screen real-estate devices, not tables as are used to layout complex graphic designs in HTML.
Give XHTML a try -- as far as web authors go, it's pretty much just using lowercase tags and closing them all. Or try HTML Tidy with the "-clean -asxml" options to convert your HTML pretty effortlessly to XHTML. Current browsers will work fine with it.
Disclaimer: I am a member of the XML Forms committee, a descendent of XHTML, but I had nothing to do with XHTML Basic.
It's definitely a step towards device independant web browsing, but I think a better approach would be two streams. Some might think this an evil idea, but if this is the way W3C is going, with all its member companies dreaming of their own little 'devices', perhaps a stream for "feature rich" browsers and another for "feature poor" is appropriate...
Open Source. Closed Minds. We are Slashdot.
XHTML 1.0 is the current W3C recommendation for regular web content (ie the stuff we use HTML for now). XHTML Basic is basically a subset of the XHTML 1.0 functionality and is "designed for Web clients that do not support the full set of XHTML features; for example, Web clients such as mobile phones, PDAs, pagers, and settop boxes".
Basically, XHTML Basic has about the same feature set as HTML 3.2: images, forms, simple tables, etc.
Portable versions of Firefox, GIMP, LibreOffice, etc
Wow, that was one of the lamest first posts I have ever seen. You should try harder. It's 3am eastern, it's not like you have a lot of competition.
XML is an effort from the SGMLers to retake the lead and bring HTML back to the purity of markup languages.
Originally XML was a simplified SGML aiming to replace HTML. So much for that one. Many of the original SGML problems were still there (in a nutshell too powerfull and general for the average web page).
However SGMLers learned in the process, split the original XML proposal into many subcomponents, including XHTML. They also learned that rendering/formating is important and should be part of a tagging standard (thus XSLT)...
We'll see if the world finally follows them in this third attempt to take over text management...
No. XHTML is not aimed at PDA's. XHTML is an XML-ified version of HTML 4. It does a lot of cleanup and makes the whole thing much saner. It also divides up HTML into various profiles, of which one is very "correct" in the sense that it doesn't have FONT, CENTER etc. tags. Instead, all styling is done with CSS, which is *very* good.
What XHTML has to do with WAP is that the WAP Forum has announced that WAP 2.0 will essentially be XHTML. So it's WAP that is adapting itself (thank God!!) to XHTML. It's not XHTML which is for PDA's.
WAP is one of the worst cases of re-inventing the wheel ever, so this is a good thing - that I agree on!
Weren't computers supposed to make our lives easier? To me it doesn't make sense that we are forcing what used to be a flexible, tolerant Mark-up into the spend-half-an-hour-looking-for-the-missing-forward -slash nightmare that C and (sometimes) Perl present to us 'less precise' developers.
I personally think it's swell that i don't have to finish tags (<br/> - come on!) or make sure they're all in the right case - let the computers work it out for themselves - they don't mind! really.
I'll admit browsers sometimes go too far - IE is infamous for ignoring MIME types and searching the content of a document to 'work out' whether it's html, xml, or japanese (and let's not talk about it accepting back slashes - just go to http:\\www.microsoft.com\ie\) but come on - we've already been robbed of the meal-in-a-pill, silver body suits and personal jetpacks - let's not start spending our days working for machines!
--
Rare Window - free your photos
XHTML Basic is just a subset of XHTML 1.0 which already achieved recommended status last January.
http://www.w3.org/TR/xhtml1/
It is meant to be used for smaller applications such as PDA's and such. Personally I believe this set will go the way of the dodo. Designers will not be too keen on creating yet another version of their sites just to accomodate another markup.
I do plan on updating my own sites to xhtml transitional within the next couple of months.
-Rod!
No, I honestly don't believe that a stricter language is the next step anywhere. Remember, in the internet, a step that helps stupid people do more is key. Like AOL and IE. Its almost common now for people to leave out html tags and the browsers don't care. The other thing that seems to drive a lot of people is the way it looks. If they cant have browser specific custom scroll bars or annoying DHTML/flash they get bored. So I think the next step, if XHTML is even in the equation, are programs that write XHTML or some type of action that just makes XHTML more like HTML.
"Compared to the rich functionality of HTML 4, XHTML Basic may look like one step back, but in fact, it is two steps forward for clients that do not need what is in HTML 4.."
Is that saying that a simple HTML may be better for people that don't need everything in HTML 4? Id understand if it was "for an audience that cant read full featured HTML 4", but if you dont need something in HTML 4, you just dont use it?
A lot of Professional web designers already pay a lot of attention to the standards. When you write, you have to pay a lot of attention to the platforms available, but new standards tell you what you can look for on the latest software, and what to keep a heads up for in the next version. Most browsers have some CSS support now, and that's very useful, for example.
It's really good to extend this to less featured devices. Plus, there is a better chance that they will track the standards, since many of them have not been developed yet, and even the ones that are often need a software upgrade to work with the Internet at all.
There is also much less diversity in the feature set supported across each class of device, so that makes it easier, also.
Even Slashdot wants to hide some things
In which case a lot of people are "missing the point". Go into a newsgroup like comp.infosystems.www.authoring.html (which is a story unto itself), and you'll see endless and pointless competition on people converting their sites to XHTML, just because it can be done. Or in many cases, can't.
Open Source. Closed Minds. We are Slashdot.
Just when I learned how to do tables ....
/. ...
Yeah, really. We're moving pretty fast here, aren't we? Hells, most browers now aren't fully standards-compliant (how that annoys me!). Also, what about web designers who are learning the language now? I've known HTML for quite a while, but my mastery of tables and their relationships with CSS has been more recent.
Nevermind whether XHTML is better or worse, this will probably confuse everyone involved with web design, from coders like me to browser designers.
BTW(and OT), I've recently switched from IE and NS4.7 to Mozilla 0.6 and I'm quite satisfied with it. Though I still have to use IE to post to
-J
Karma: T-rexcellent.
I assume you're saying that Tim misunderstood SGML's goals and mangled them in the creation of HTML, but mightn't it have been a more pragmatic thing? He wanted something usable, and he created it, and it worked.
The software field is littered with theoretically pure technologies which aren't widely adopted because of a staggering variety of practical issues that arise when ordinary mortals are faced with the task of doing something useful with a computer. Every now and then, someone with a practical bent comes along and incorporates just enough of the most useful and usable pieces of one of these pure technologies into a working end product, which people then adopt en masse, largely because the theoretical underpinnings, while imperfectly implemented, significantly improve on what went before.
Think of computing science and the concepts it produces as a Platonic world which effectively has no real existence, because of lack of practically usable implementations. The real world can only generate weak shadows of the concepts in that Platonic world. Cutting-edge software designers attempt to translate theoretical concepts into useful and usable end products, but these attempts often fail.
One can't and shouldn't judge people like Tim and their achievements from a purely software-theoretic point of view. Rather, he made something usable for the rest of us, and raised the bar for those coming after him.
Beginners now, instead of using <b> and </b> will have to learn how to create style sheets or use extensions to the DTD! This new move, if actually adopted by web browsers, will make even MORE people use Netscape Composer and other front-ends rather than realizing how simple HTML is... or was... to use. XHTML Basic doesn't seem so "basic" to code, removing one of HTML's great strengths.
Wireless clients will now have to send out more GETs to display data. If they honor any sort of style sheet, they're going to be following at least one LINK per web page. That'll use up more air time; services charging by the page or even by the minute will cost more for wireless users.
Can't handhelds and other browsers just ignore markups that they can't display?
I know they aren't kidding when they say BASIC, but what do you want in markup that'll work on a wristwatch as well as my PC. I know there aren't even tables in the spec, but it'd be a great way to "page" your friends.
It seems to me that this will be a big hit for "Push" content (like PointCast -- remember PointCast?) I just wish it could have tables and forms for more interactivity. I guess if you're getting pages on your smart watch it's not an issue.
Still I'd like the ability to punch in my zip code to get some weather updates, or something.
I wonder if they will make XHTML Basic less basic when people start adding tables and forms with XHTML Modules. Interactive sells. Is XHTML Basic, too basic? (No forms, or tables -- can Mod it though)
It's not just about device independance, it's about a change in concept. Tables were not intended to be formatting elements, but to contained tabulated information. Previously, there was no other way to do the same kind of formatting that tables accomplished, therefore they were used as such.
Now we have Dynamic Positioning, which negates the necessity to use tables for formatting. It can do everything tables can do, and much, much more. Being able to look at positioning attributes in their own seperate file eases upkeep by an infinite amount when compared to wading through multiple levels of tables that will break if just one of the tables is off by a bit.
Dyanmic Positioning is a CSS element, therefore seperating Style from Content(tm). If everyone started using tables only for their intended purpose, the searching capabilities would expand tenfold. TD stands for Table Data, not "A place to to put yet another grossly overused left-hand nav buried within 10 nested tables".
"We obviously need a new moderation category: (-1, Woo-fucking-hoo)" --Mr. AC
Yes, XHTML Basic is for internet devices, as it is just a subset for XHTML 1.0, which is a recommended standard for the web as a whole.
"We obviously need a new moderation category: (-1, Woo-fucking-hoo)" --Mr. AC