Forum Moderators: phranque
Why do I feel like the only one in the world besides the W3C doing this. I checked a ton of google page rank 10 websites, with the only ones I could find that even met compliance being the W3C. As for creating usefull outlines, macromedia (who doesn't validate either) has a meaningful outline. Even webmasterworld doesn't validate or have outlines. I went through and checked websites of users on webmasterworld with over 1,000 posts. Albiet, most are personal sites and probably not their main site, almost none meet compliance and if they do, they have no outlines.
I found one website talking about xhtml validation, and it did validate but it's outline was gibberish. They said they used HTML Tidy to convert their pages, but they didn't even bother to fix the heading tags.
With all the time the people at the W3C spent to make these validators and how usefull and easy they make creating a validated website why do I feel like the only one on earth that is using them to their fullest potential. I seriously have not found one website that takes full advantage of the 'outline' feature.
It seems that all top level websites including google, msn, yahoo and so on could care less at validating so why have I wasted so much time on it? Google, it seems, could easily validate their pages but they obviously choose not to. My question is WHY? Why doesn't google or many others bother to validate?
Google does not need to be spidered and ranked by any search engine. It doesn't need to care if its HTML or XHTML trips up spiders.
It does care (I guess) that its HTML renders in a gazillion different browsers. Chances are they've done the research (or acted on a zillion feedback messages) and are happy that their HTML renders okay in all platforms they care about.
The only moral you can draw from that is:
The validation errors inserted by Google do not harm rendering. They may seriously harm spidering and indexing, so replicate these errors at your own risk
Without doing a vast amount of research, it would be dangerous to insert other errors into your X/HTML.
As I said, these things are not symmetrical. In the same way, Google doesn't need to spend anything on advertising, but makes zillions from Adsense. Doesn't mean you should not spend on advertising.
It just goes to show: the potential is there, and it is taking off rapidly. Don't right off standards compliance yet: it is just starting to hit the mainstream big time.
A point about validation too: it is very difficult to impossible to ensure validity on large dynamic websites which bring in content from a variety of sources, not all under the site owner's control (especially including user input, which is the case with this forum - whose basic templates all validate). In order to evaluate, you have to analyse the type of errors, and the overall structure of the page and site, not get bogged down with the validator messages.
Maybe I would feel better if someone could list a website(s) that has a page rank of 10 that passes any type of validation and has even a decent attempt at a outline that isn't W3C.
Although Encyclo is correct in pointing out that standards compliance is gaining popularity, so long as your site can look right on browsers with tag soup, most of the sites you see won't be standards compliant. It's just a lot of work to convert a large site from tag soup to valid code. Trust me, I've done it with a medium sized site. While their sites aren't broken, they're not going to invest large sums of money and time in fixing them.
There are lots of good reasons to write valid code, but Google rankings isn't one of them. I have my main site validating to XHTML 1.0 transitional at the moment and planning to tighten that up with XHTML 1.1 in the near future. But the reason is that my audience is likely to be interested in those kinds of things. If my audience didn't care, I don't think I can see myself putting the effort forward.
I prefer HTML4 to XHTML, but we're talking about much the same thing
Not really... I know you're talking about validation in general but there is a major difference between HTML and XHTML that shouldn't be glossed over. If you search for "Why not to use XHTML" you'll find an article by Ian Hickson that details some important things to consider about XHTML. I don't agree with everything he's written, but there are a lot of good cold hard facts that you should be aware of if you're using XHTML.
Why do I feel like the only one in the world besides the W3C doing this?
Hey I know this feeling. It seems that clients aren't even aware that there is a standard, and the competition is happy when their client likes it in IE. For me valid is the only way because that's just my nature and it's not that hard. It helps me find errors sometimes when validating along the way. I'm usually satisfied with 4.01 though.:)
The best known large site with standards-based markup and CSS is wired.com. There are more and more large corporate and government sites being built or redesigned with standards compliance in mind.
However, don't get too hung up on what the validator says: it is just a tool and not an ultimate judge of how the site stands up to standards compliance. You can have a totally inaccessible page full of
<font>[edited by: encyclo at 3:00 am (utc) on Dec. 5, 2004]
validation doesn't matter as much as ensuring that your page renders on all web browsers.
I guess my main point of starting the thread was this, is there something going on that I am not aware of? Is there some new type of universal web standerd just around the corner that all these companies are waiting for? If all web browsers coming down the line will support xhtml and all web browsers are free (that i'm aware of) and people can keep downloading the newest browsers for free what other reason are all these companies holding back on converting?
For example, macromedia is pushing their newest products including flash. What webmaster is going to drop hundreds of dollars on expensive software while browsing the Internet on his Windows 95 machine running IE 4? Is Macromedia concerned with just such a thing happening? The same company asking people to download their latest plugin on a daily basis is afraid to make its website visitors use a post-1995 browser? I mean, there supposed to be on the leading edge of web development technology yet they keep their website in tag-soup format? Dreamweaver MX 2004 has a checkbox asking if you want to make your document XHTML compliant when creating a new page! I guess its the irony that gets me. Adobe is no better. XHTML 1.1 has been out for years now, many of these websites have had complete makeovers since then.
I hate to keep beating a dead horse here, something just doesn't seem right about all this. There must be some bigger reason these companies are steering clear of xhtml for their websites.
The best known large site with standards-based markup and CSS is wired.com.
*sighs in relief*
If you search for "Why not to use XHTML" you'll find an article by Ian Hickson that details some important things to consider about XHTML. - karmov
"Unfortunately, IE6 does not support... XHTML at all" - Ian Hickson
Is this accurate?
Yes. And no. (It's a mime-type thing, and you could argue for years and not get anywhere).
XHTML is here to stay, but not necessarily in the form that the original creators expected. I prefer HTML, but I have one large site (a forum) which uses XHTML. Why? Because I'm using phpBB, and that tool has XHTML syntax hard-coded into it. I could change all of that, but it would make patching much more difficult and it would make no difference to the forum users, so what's the point? I'd just be wasting my time. More and more tools are producing XHTML syntax by default, so the presence of XHTML is only going to increase.
Seeing as we're off-track anyway, a quick word about XHTML 1.1. Don't use it. Ever. Seriously. XHTML 1.1 is not the successor to XHTML 1.0, it is an offshoot designed for XHTML Modularization - ie. for use mixed with other markup languages such as MathML or SVG. There are debates over whether it can correctly be served as
text/htmlXHTML 1.1 is not the "latest standard" either - XHTML 1.0 Second Edition came out afterwards, and includes many improvements which are not present in XHTML 1.1, such as allowing an
idhtmlBut much more importantly, XHTML 1.1 is inappropriate because of its sparseness: it is actually less accessible than XHTML 1.0 because so many attributes (especially
lang
So why would companies, even google, not turn their homepages, at the very least, into xhtml compliant pages.
I'd suggest that there's no real point to validating one part of a website without validating the rest. If you do then you have two different codebases to maintain.
But again I would expect the issue to come down to the fact that XHTML won't render properly on something. I'm not talking major browsers that everyone uses. Most likely it's something obscure that we've not used or possibly even heard of before. But if there's something out there that only has an HTML parser, it will choke on XHTML.
The real beauty in Google is its simplicity and its consistency. Google looks identical on every single browser I've ever had the pleasure of using.
The W3C standards were never meant to guarantee that a site would look the same on all browsers. They were designed with different goals in mind. The goal of the standards are different than the goals of Google, plain and simple. For some companies, they're not mutually exclusive, but for others it would appear that they still are.
So before anyone continues whining about why your valid, semantically correct (X)HTML website has a lower page rank then an invalid, semantically incorrect (X)HTML, perhaps you should check into who is linking to you first and how you can get more people to link to your pages.
Seeing as we're off-track anyway, a quick word about XHTML 1.1. Don't use it. Ever. Seriously. XHTML 1.1 is not the successor to XHTML 1.0, it is an offshoot designed for XHTML Modularization - ie. for use mixed with other markup languages such as MathML or SVG. There are debates over whether it can correctly be served as text/html at all, and whatever your stance on that question, why deal with the uncertainty?
This is directly from the W3C website,
The purpose of this document type is to serve as the basis for future extended XHTML 'family' document types, and to provide a consistent, forward-looking document type cleanly separated from the deprecated, legacy functionality of HTML 4 [HTML4] that was brought forward into the XHTML 1.0 [XHTML1] document types. This document type is essentially a reformulation of XHTML 1.0 Strict using XHTML Modules. This means that many facilities available in other XHTML Family document types (e.g., XHTML Frames) are not available in this document type. These other facilities are available through modules defined in Modularization of XHTML, and document authors are free to define document types based upon XHTML 1.1 that use these facilities
As far as I can see XHTML 1.1 just removes the legacy features from XHTML 1.0 that were carried over from HTML 4.
XHTML 1.1 is not the "latest standard" either - XHTML 1.0 Second Edition came out afterwards, and includes many improvements which are not present in XHTML 1.1, such as allowing an id attribute on the html element.But much more importantly, XHTML 1.1 is inappropriate because of its sparseness: it is actually less accessible than XHTML 1.0 because so many attributes (especially lang) have been removed. If you prefer XHTML, then aim for XHTML 1.0 Strict as the optimum standard.
This is also from the W3C site,
1. On every element, the lang attribute has been removed in favor of the xml:lang attribute (as defined in [XHTMLMOD]).
2. On the a and map elements, the name attribute has been removed in favor of the id attribute (as defined in [XHTMLMOD]).
3. The "ruby" collection of elements has been added (as defined in [RUBY]).
I am not looking for a argument but I do use xhtml 1.1 and saying that a person should never use it is quite intimidating. I can't seem to find anything on the W3C that really points out where xhtml 1.1 is somehow less superior than xhtml 1.0. If you have some sources I would love to read them. This is very important to me.
Just how much is a webpages page rank related to valid, semantically correct (X)HTML? From what I know of Google's page ranking system, it is mostly based on the quantity and quality of inbound links then the actual content of the page. Page rank and valid, semantically correctness doesn't appear to be related to each other at all.So before anyone continues whining about why your valid, semantically correct (X)HTML website has a lower page rank then an invalid, semantically incorrect (X)HTML, perhaps you should check into who is linking to you first and how you can get more people to link to your pages.
I have no idea how the search engines do their work but I do understand that back links are very important. I am just curious what they think about correctly configured pages. It's not a like a human is looking at your design and layout it's just a computer reading your code. So, besides backlinks, if the only way a computer can judge the quality and build of your website is in how easily it can decipher your code then wouldn't having computer freindly code be handy?
My main point would be adding a outline using the <h1> through <h6> tags. We can format are pages so that humans can grasp headlines and sub content but this does nothing for a computer. Using the heading tags we can tell the computer the level of importance of the text in the code. It seems like this would be a big bonus to people who write search engine algorithms. It would let the computer reading your code know that your site has valid content that is well structered.
<h1>Website name
----<h2>Page title
--------<h3>Category 1
------------<h4>Sub category
--------<h3>Category 2
------------<h4>Sub category
----------------<h5>Side note
It seems like the W3C came up with the idea of structuring websites according to heading tags, whether or not this was an idea when they wrote the html specifications I haven't a clue, but seeing that almost nill websites use this simple structure is quite a mystery. I know this is non-XHTML related, sorry.
I have no idea how the search engines do their work but I do understand that back links are very important. I am just curious what they think about correctly configured pages.
It's not a like a human is looking at your design and layout it's just a computer reading your code. So, besides backlinks, if the only way a computer can judge the quality and build of your website is in how easily it can decipher your code then wouldn't having computer freindly code be handy?
If you are creating a semantically correct webpage expecting to increase your PR value, then you are chasing a mirage. It's fairly clear that the two issues are completely unrelated to each other.
I checked a ton of google page rank 10 websites, with the only ones I could find that even met compliance being the W3C.
For a PageRank 9 that almost validates: check out the Open Directory Project, DMOZ.org
(Normally, it would validate; and their stated intention is to generate standards-compliant code. Looks right now that a couple of bugs have slipped through)
[webmasterworld.com...]
[webmasterworld.com...]
[osnews.com...]
To keep the discussion of the issues somewhat short:
1) there are two mime-types that can be used when serving web-pages: text/html and application/xhtml+xml
2) xhtml 1.0 in html compatible mode is the *ONLY* kind of xhtml that ought to possibly be served as text/html
2) IE *ONLY* understands text/html. When sent application/xhtml+xml IE thinks it's an executable program and acts accordingly.
3) xhtml is most correctly served as application/xhtml+xml
4) IE6 sent xhtml as text/html with an xml prolog behaves like IE5.x
5) *any* browser that receives xhtml served as text/html treats your carefully validated xhtml as plain tag-soup html 4.01
Problem is, most web developers and web designers don't know about the differences between text/html and application/xhmtl+xml and thus get all of their pages served and treated as plain old html 4.01 in quirks mode.
See the links in the above forum articles for documentation.
I seem to see quite a few opposing views and some contradicting comments on both sides of the issue.
I have IE6, firefox, mozilla, and opera. My web pages are validating as xhtml 1.1 and are being sent as text/html. They all look the same in all the browsers, nothing is breaking are acting weird in any way. I used to have IE 5.0, 5.5 installed before reinstalling my O/S (forgot where to download them from but maybe someone will remind me) and am sure I tested my pages in those also. I would be happy to test them again.
I can't seem to find anything on the W3C that talks about these problems or that xhtml 1.0 is their newest standardization.
I am not finding the documentation saying for a fact that you can't send xhtml 1.1 as simply text/html. My only basis for a opinion is what I see. I see my pages rendering the same in all the above browsers. My code is clean and readable, it renders the same across the board, it validates at W3C, and is future compatable (I hope).
From some website
The XHTML 1.1 specification, states that it should be sent as application/xhtml+xml [1] - sending it this way will cause IE to break; Sending it as text/html invalidates it. I suggest using XHTML 1.0 Strict instead.
So the only real true reason I am seeing to not use xhtml 1.1 is because technically because you are sending it in the wrong format, even though the browsers and validators don't care, it is technically invalid.
Well if the validator and browsers don't care, why should I? Is there some other way of viewing web pages that this will cause problems in? I guess I am missing the point, if it validates and renders correctly what exactly is the reason not to use xhtml 1.1?
*any* browser that receives xhtml served as text/html treats your carefully validated xhtml as plain tag-soup html 4.01
If my code is validating as xthml 1.1 using the W3C validator this means my code is in accordance with the formatting of xthml 1.1 which is a very strict set of rules, correct?
If my pages are all following a strict set of rules then even if the browser treats it as tag soup it is still perfectly formatted tag-soup and will look the same on all pages of the browser regardless of how the browser treats it. So if my pages all look like their are supposed to then I don't see the harm.
As for *any* browser, I don't quite understand that either, under page info in firefox it clearly states my page is application/xhtml+xml under content type, which is most likely the same for netscape and mozilla.
In fact it's very easy to make a page in xhtml 1.1 + css that doesn't show up exactly the same when served as text/html and application/xhtml+xml:
In xhtml you can - and should - apply a style to the <html> tag as under xhtml the <html> tag is the outermost tag - *not* the body tag as under html. So, when sending xhtml as text/html the style set for the <html> tag is ignored.
Do you use any javascript? And do you use document.write in them? If so, then your javascripts aren't xhtml ready as document.write is an no-no under xhtml.
But to quote w3c about serving xhtml 1.1 as text/html:
Why is it disallowed to send XHTML 1.1 documents as text/html?
XHTML 1.1 is pure XML, and only intended to be XML. It cannot reliably be sent to legacy browsers. Therefore XHTML 1.1 documents must be sent with an XML-related media type, such as application/xhtml+xml.
text/htmltext/htmlBut that is a minor problem. Here's a better example. If you want to mark up the following phrase:
The French word for hello is bonjour.
You can do it like this in HTML 4 or XHTML 1.0:
<p>The French word for hello is <span lang="fr">bonjour</span>.</p>The span marks the French word with the appropriate language for a screenreader.
In XHTML 1.1, there is no
lang<p>The French word for hello is <span xml:lang="fr">bonjour</span>.</p>There are two problems. Firstly, user agent (eg. sreenreader) support for the
xml:langtext/htmlapplication/xhtml+xmlxml:XHTML 1.0 is more complete, even in its Strict form, and is a much better choice, especially if you are serving it as
text/htmlFinally, whatever the W3C documents say, "deprecated" is pretty meaningless in the context of elements and attributes which are likely to be removed from future versions. The W3 has abandoned all development of HTML and XHTML 1.x in favour of XHTML 2.0 - and in that standard, even the
imgThe advantage of using a strict doctype is that you forcibly use fewer elements and attributes which have presentational aspects or are problematic in terms of accessibility. However, if you require some of those deprecated elements, you can use a transitional doctype in the knowledge that the markup is no more or less future-proof than a strictly conforming document.
Ok, we're really off-track now :)
Ok, we're really off-track now happy!
This has actually turned out into a much better discussion that I could have ever imagined.
Now here is where I stand:
On one side, I have a regional website that I don't really have to worry about other languages so that isn't a problem. In hopes of making my site accessable I only use javascript for keeping my page from being framed and another to replace target=_blank. This way if a visitor has JS disabled they wont even notice it's not there. It also seems when they refer to legacy browsers they are talking about older than IE5 and it is getting past the point of worrying about people still using IE4, at least in my opinion. So since i'm having no problems and I love the strictness of using 1.1 and how nice my code looks I might want to consider keeping 1.1.
On the other hand, what's the big deal about switching back to 1.0 strict?
I know I can go read a lot of documentation and re-check my web pages in the validator but is there any chance someone here can sum up what it is going to take to go back to 1.0 strict from 1.1?
sum up what it is going to take to go back to 1.0 strict from 1.1?
Just change the doctype at the top. If your page is valid 1.1, it's valid 1.0. The advantage comes if you ever need to use the additional attributes and such that XHTML 1.0 offers.
