This article, endorsed by Mark "Brain the Size of Utah" Pilgrim, says that XHTML 1.0 is pointless. Pilgrim goes on to mention than XHTML 1.1 is just as bad, and XHTML 2.0 is shaping up to be worse.
The solution? Mark Pilgrim, master of all things new and hip, has decided to go back to HTML 4.01. Every web designer on this board should think about this deeply. I know all my stuff is still XHTML 1.0 Strict, because I listened to Zeldman and decided it was the "correct" way of doing things... but now I'm not so sure.
All very interesting - XHTML 2.0 sounds a nightmare I think this from Zledman is apt though:
quote:Today I agree with Mark in having no desire to use XHTML 2. (Which I canâ??t anyway, since itâ??s not a final spec and no browser supports it or will for years, if ever.) But I disagree with Markâ??s conclusion that standards are bullshit. Funny, XHTML 1, CSS1, CSS2, ECMAScript, and DOM are working just fine here. The problem is not standards but this proposed XHTML 2 standard, and the problem, I think, is one of nomenclature rather than hubris or stupidity.
I think we need XHTML - we need to move to more of a XML style markup and structure so we can weave metadata into and around our documents (with things like RDDL, RDF, RSS 1.x, etc.). We are going to increasingly see The Semantic Web pay off soon and going back to HTML 4.01 strikes me as a solidly bad idea.
I think this is interesting:
quote:If you are one of the few authors who understands how to avoid the issues raised in this document and does validate all their markup, then this document probably does not apply to you
i.e. the problem isn't really with the technology its with its implementation (mainly in the people writing the code but also in the browsers/UAs) - if so we need to get educated not revert to older standards.
Granted the situation isn't ideal but when has it ever been? Lets be honest the situation today is far better than it has been and the future looks increasingly rosey
Lets not fool ourselves though these are important issues and by raising them we can address the problems (I've got a lot of reading ahead of me clearly) so.............
From: Rochester, New York, USA Insane since: May 2000
posted 01-16-2003 04:20
I found that most of Mark's bitching had to do a lot with him not being able to code appropriately. There are freely available code validators so that you can make sure your markup is clean. You are not suposed to mis a single closing tag. HTML and XHTML are not suposed to allow you to do things wrong.
The vendors are the ones who allowed you to break things not the code. It was a blessing that it allowed a lot of people in the door, but we are moving towards a far more structured enviornment where it is not just "human" me mingling in the web, but it is web robots of all kinds, crawling searching storing, repeating, even displaying.
We have openned the flood gates and now we have to slowlying bring them back into a fully manageable enviornment. There is no doubt in my mind that we need to continue forward in the advancement of the markup, we can not stay static, and at the same time the code base for the UA's are not and should not have to handle all of the complexities of 10's or 100's of DTD's internally, allong with their rendering. The code bloat there is just enormous. There is no reason for a browser to need to be 10 or 20 mb's in order to show you a webpage. The extra features are where the code should be put. I know that I do not want, with the current standards of the web, to be put in the position to develope a browser to work with the different code bases out there.
I must also say that when I looked at the XHTML 2.0 specifications I was really, really impressed. It was like I got a new candy jar to play with. The address tags, and add and delete tags, finally moving over to an object model, and deprecating the br tag. It made my mouth water. There were some amazing ideas in there, with some awsome posabilities. Just the edit tag made me really happy. Those are the anti-1984 tags, you can change your content, but still perserve your old content, for those who would like to see it. It is like a built in versioning system.
I could go on and on about the object model, the recent ALA article on PNG would be obsolete as you could design your images to degrade to a useful version, all the way down to a textual display, which was really top notch, a win for accessability all the way around.
I could go on and on about the bonus to accessability in the XHTML2 documentation.
I can also see where many would not be happy with XHTML2 If you don't like change and do not want to adapt or look at new things, you will definately not like this. As XHTML2 is now, if the masses were to accept it, it would need a middle ground. A bridge between XHTML1.1 and XHTML2 would be needed.
Back to the rants. Do not bring it down on the standards, which tend to be well written, but poorly maintained. It is really a hard nut to crack. SInce everything is seperate but trying in some form or another working togeather. Microsoft does not help, by funding and contributing the W3, and then not following the standards and creating competing standards. There are a whole lot of issues, and many are with browser support, and the rendering engines.
That is why I don't want to be a part of developing a browser, the weight of the web is on their shoulders.
WM: Thanks for that viewpoint - very interesting XHTML 2.0 has been painted as the baddie by so many people I had started to dismiss it completely so I think it is time to have a deeper look. I suspect the main problem people have is that it isn't backwards compatible and that will be a real problem unless we have XHTML 1.3-1.9 which ease us (over the next 5 to 10 years) into this.
As a side note about XHTML 1.0 - I have now been through Appendix C and the 16 points are either things you should be doing anyway or rather obscure little tweaks so I can't really see much excuse not to produce good quality working XHTML 1.0 - its not the fault of the standard (or, shockingly, even the browsers/UAs) the downfall of XHTML 1.0 is poor implementation by the coders. I can't claim to have it all nailed down perfectly but I can see the issues.
However, in the end it is your call what you code in but I think it would be a collosal mistake to discard XHTML 1.x and remember XHTML 2.0 is only a proposal so you all get to have some feedback.
From: Rochester, New York, USA Insane since: May 2000
posted 01-16-2003 05:16
True, version2 is not to be implement, but to be debated strongly as it is now. You need to have full support of people supporting it as well as critizising it. It all works to make it better.
This point of view is really interesting... I am making a website in XHTML 1.1 (mostly in order to learn how it works) and this is true that there are things that appear to me really...humm "obsolete", that is to say useless to my point of view, and useful things that are alot harder to implement. I will read those "items" in order to be more informed and see if I come back to HTML 4.01...
Ultimately all things are known because you want to believe you know.
-Zensunni koan
From: Rochester, New York, USA Insane since: May 2000
posted 01-22-2003 15:43
What types of things do you see as obsolete?
I found the transition from HTML 4.01 to XHTML 1.1 transitional to be really painless. There was not much involved in that change. The hard part is turing the transition into Strict, which is a little goal I try to work on in small steps.
Well, if you try to validate your page as strict XHTML, you will see it's not so easy and painless...
I will explain a example I had which "angered" me... In XHTML strict, you can't anymore use the property "align= center", you have to use styles instead. This is a good point, I think using css is far more better than HTML code. So you create a css property where you use : "text-align: center". You try to applicate it to your object... Personnaly, I wanted to align in the center a table. The width of every row of the table had to be defined exactly because I used images and text in it. So I applied this property to my table. And... That didn't do anything. The property wasn't working on tables. I tried it on layers, images and text : it worked perfectly. But it didn't worked on tables. SO I included my table into a layer and applied the property to a layer : it didn't worked either. Finally I had an idea : I added another property to my table : I set the width to 100%. Now my rows were larger and the graphism was killed, it didn't worked. So, to prevent the width of my rows from being increased, I created a new empty row on the right with the property "overflow: hidden". I thought this row would be the only one to be increased. I was wrong. The result was still the same. Finally, after thinking about it a bit more, I added another empty row with this property but to the left of my table. And it worked ! Now, the width of the table is always 100% and the real content is always centered because the two empty rows expand on the sides. It works on all browsers (Uh at least IE 6.0, Opera 7 beta 2 and Netcape 7) in any rsolution.
I am quite proud of what I made. But frankly : do you think the transition between HTML 4.01 and XHTML 1.0 Strict or XHTML 1.1 in this case is painless ? Maybe I am not a great webmaster, maybe there was an easier answer to my problem, but as far as I know XHTML is much more hard than HTML 4.01.
Ultimately all things are known because you want to believe you know.
-Zensunni koan
From: Rochester, New York, USA Insane since: May 2000
posted 01-23-2003 01:29
You don't need to go to Strict you should probabally go to Transition, there is a lot you can not do with Strict. The solution to the afore mentioned problem might have to be that you need to add the "text-align:center;" to the td attribute as opposed to the table. If you apply it to the table you would get a centered table. And not the text, since the text is held in either the td element and not the table element.
I think you would find that XHTML at its root is not nearly that complicated Briggl's tutorial over in the coding section of the GurusNetwork is a good place to start it will get you into using XHTML far faster than many other tutorials on the same subject will. XHTML is really not that hard, just try using Transitional DTD instead of the strict, you page will as if magic validate a whole lot easier. I don't know many people who use the Strict DTD just because it is very very Strict.
First off, I don't think people really understood Mark's objections here. Dismissing his view on the grounds that he probably doesn't know how to code 'appropriately' is not only fatuous, it's wrong, as once glance at the source-code of his site ought to tell you. His complaint about application/xhtml+xml being oversensitive to missing tags wasn't the beef of his objections.
It was twofold: the fact that XHTML2 appeared to be touted as the immediate successor to XHTML 1 despite alienatingly parsimonious removals, arbitrary renaming conventions and the flagrant incompatibility that results; and the fact that XHTML1 was (according to the objections presented in the original article) impossible to serve up in true adherence to XML spec and had no apparent benefit. Personally, I think he took the potential logistical nightmare of XHTML content-types way too seriously, but only because the UA vendors don't take it seriously enough.
Secondly, XHTML 2.0 hasn't moved to an object model in the commonly-understood (programming) sense of the word. It simply removes <img> in favour of <object>, which has been around since HTML 4 with the same functionality and accessibility benefits - and the same implementational hazards. It still relies on the UA to know whether it supports a feature/format "well enough" for what the author intended, which in IE's PNG case is the problem.
And guess what else has been around since HTML 4, but never used? How about <address>, <ins> and <del>? This stuff isn't new, guys. XHTML 2 has put the spotlight on a lot of the stuff we've had for years (and apparently never realised): the real additions occur with the <nl> list structure, the <h>/<section> pairing, and the href attribute applied to all elements.
Some of the stuff in XHTML 2 offers great markup potential, and some of it improves accessibility, and some of it is just designed to beat authors over the head into bettering their semantics. But hey, at least the debate has drawn attention to the XHTML2 working-group and made the language have to justify itself for once.
Moon-shadow: the problem you're facing is that the style "text-align: center" applied to a block-level element (like a <table> or <p> ) means that all the inline content (text, images) inside the block is centered, not the block-level element itself. So when you apply "text-align: center" to a <table>, the browser thinks what you mean is "make the content in all the cells inside this table centered." Which then usually gets ignored due to arcane UA quirks with inheritance rules and tables.
Having empty filler cells on either side of the table isn't really the 'right' answer, though it works. In order to 'properly' center a block-level element (like a <table>, or a <p> whose width is less than 100%) using CSS, you need to set "margin-left: auto" and "margin-right: auto". This will tell it to allocate equal margin-space on the left and right side of the element, given the specified width (which is the case of tables depends on their content; most other block-level elements are 100% width by default), thereby centering it. Unfortunately, this technique isn't supported in IE < 6, and Opera 6 has some bugs regarding background-images and auto margins.
[This message has been edited by Vigil (edited 01-24-2003).]
posted 01-15-2003 00:48
Ultimately all things are known because you want to believe you know. -Zensunni koan