validator.w3.org

[kgun]

Originally Posted by idansh

I don't understand why anybody needs it ,it just maketing toy, there is no real usage for it

Originally Posted by weslinda

idansh...really? that's what you've taken away from this entire thread? it's a marketing ploy?

There is a fight between W3C trying to make standards, that is in the end to make life easier for people and important companies and writers. This

Originally Posted by mike

Reply #6

"Mike,
No problem - here's my quote.

In exhaustive testing from many, many experts in the field - from Dan Thies to Greg Boser to myself (and even the contentious but ruthlessly picky Michael Martinez) - there has never been any evidence whatsoever to suggest that a page that validates the W3C standards would be considered more valuable or rank higher at the search engines.

Furthermore, search engine representatives like Matt Cutts & Priyank Garg (Yahoo!) have publicly stated on search industry panels that they believe W3C validation does not inherently add to a user's experience or satisfaction with a search result and thus do not include it in their algorithms.

The logic behind this is somewhat infalliable - great content doesn't always validate (in fact, often it doesn't) and perfect validation is no indication that the content matter is high quality.

Spammers can actually validate their material much more easily than large companies, media organizations or other "trusted" sources (whose programming, as many know, is often FUBAR), so why would the search engines use it as a quality signal?"

-Rand Fishkin

My emphasis again

is a typical example.

"Spammers can actually validate their material much more easily than large companies, media organizations or other "trusted" sources (whose programming, as many know, is often FUBAR), so why would the search engines use it as a quality signal?""

Never be an advocat of spagetthi coding / markup nor spamming.

I will launch a new concept:

Spagetthi spamming.

May be I should end my posting in this thread with the following Norwegian saying.

I give up so you can stay wrong.

[incrediblehelp]

I am glad Mike offered some email responses to help forward his position.

I have never thought valid code is needed to rank as that is easy to see with the naked eye in the SERPs everyday. One area I was gray on is if valid code at least "helped" websites rank better and that seems to be untrue from the responses here. Of course I am not talking about crawlability, but simply one website having valid code and one not having valid code with all other items being equal.

Still I will look to limit validation errors just from a design perspective as Webnauts suggests.

[Webnauts]

Mike thanks a lot for sharing the mails above. But I think you did not pose the issue as I previous mentioned.

One last time:

Which for example are the most important html tags in terms of ranking? Lets say, the title tag? Or maybe also headings?

So, I had a client who had the the following on his pages (lets take the homepage here):

<title="Northern Cyprus Hotel Manolya"></title> instead of <title>Northern Cyprus Hotel Manolya</title>.

Oh and by the way they had for a first heading tag:

<h1="North Cyprus Hotel Manolya"></h1>, instead of <h1>North Cyprus Hotel Manolya</h1>

So, can you please ask the employees of Google and Yahoo, how do their bots interpretate the above? As a title and heading tags? Or just plain text?

If they tell us that their bots extract and interpretate the above as title and heading tags, then I will admit that I am wrong. But in terms of software engineering and development, I cannot believe that they will tell that I am wrong.

Can you please give another try? Or did you mentioned this example already?

Me and my team have developed a server side solution, to strip off any piece of code and content on our pages. If you are really right, we can strip off all our pages html code, and serve the search engines only text. If that is what they want. But would they be happy with that? In terms of crawlability maybe yes. But will be be ranked the same well?

This is a very honest question, and not an attempt to convince anyone that I am right.

[Webnauts]

Originally Posted by incrediblehelp

I am glad Mike offered some email responses to help forward his position.

I have never thought valid code is needed to rank as that is easy to see with the naked eye in the SERPs everyday. One area I was gray on is if valid code at least "helped" websites rank better and that seems to be untrue from the responses here. Of course I am not talking about crawlability, but simply one website having valid code and one not having valid code with all other items being equal.

Still I will look to limit validation errors just from a design perspective as Webnauts suggests.

Jaan do you want to say here that editing this <title="Northern Cyprus Hotel Manolya"></title> is the same as editing <title>Northern Cyprus Hotel manolya</title>, and that both cases would be be treated or weighted equal by search engines?

Then I guess I never understood software development and semantics. And if that is the case, then I feel ashame.

Embarrassed!

[bj]

Google and Yahoo! handle some of the thickest traffic on the market. As of this writing, neither site complies with the W3C validator. How about heavy-hitters CNET and eBay? Failed. Adobe’s and Macromedia’s websites? Failed. What about php.net and python.org? These are the home bases of open-source developers who themselves rely on evolving quasi-authorities like the W3C … but no; these sites fail, as well. How about netscape.com?

LOL! Every damn site you pointed at is on legacy software that took massive investment to create. And most of them were waiting for the final version of IE7 to be released, which happened a few short months back, before they'd start on a very expensive rebuilding campaign. One of the guys who is working on Yahoo's rebuild is on one of the lists I subscribe to. It'll be close to a year before these new sites are completely finished, tested and deployed. It's not like any of those sites runs on static html created in Dreamweaver, guys!

I repeat-- Just because things are fine with sloppy crappy code now doesn't mean they will always be. So you head in the sanders keep doing what you're doing, and I'll be laughing all the way to the bank when the boom drops.

[Narasinha]

Originally Posted by mike

Google wants correct html -- that doesn't specify in any way VALIDATED html.

They don't say your HTML needs to be validated because they don't need it to be validated.

There is nothing 'incorrect' about using a lot of tags in your HTML that the W3C is going to fail you for.

Okay, maybe I'm just a little dense, but I really don't follow your reasoning here, Mike. What is "correct" HTML if not HTML that conforms to the published standards? If the validator finds that there is an error (not simply a warning) in your HTML, then how is it still "correct"?

[incrediblehelp]

Originally Posted by Webnauts

Jaan do you want to say here that editing this <title="Northern Cyprus Hotel Manolya"></title> is the same as editing <title>Northern Cyprus Hotel manolya</title>, and that both cases would be be treated or weighted equal by search engines?!

Nope not saying that at all. I use <title>hello</title> on my pages as the title of the page and that works fine. Should we be doing somehitng else, lol?

I have no idea why Manolya was using the =" ". Glad you changed it and saw + results.

[Webnauts]

I posted my point at Google's Webmaster Help Group:
Can invalid code impact rankings? - Crawling, indexing, and ranking | Google Groups

Maybe you would like to bookmark that for the possible coming responses.

[Webnauts]

I think you also would like to follow up possible coming responses to my feature request at Google's Discussions > Suggestions and feature requests > Revamping the Webmaster Tools Help Center - we need your ideas!

Lets see what Vanessa Fox might come up with.

[Webnauts]

At last I would also like to add here:

Don't you think this article of Wikipedia should be updated? Search engine optimization - Wikipedia, the free encyclopedia

Any volunteer editors? Me not.

[jtracking]

Originally Posted by weslinda

Accessibility, Useability, Cross Browser Friendliness are just a few to begin with.

it's funny when you find that you've validated your documents but when it finally comes to cross browser friendliness, well, there isn't any...

things look nice in IE and bites in FF....things like that.

[Webnauts]

Originally Posted by jtracking

it's funny when you find that you've validated your documents but when it finally comes to cross browser friendliness, well, there isn't any...

things look nice in IE and bites in FF....things like that.

What are you talking about? If your document validates doesn't mean that you are cross-browser. If it would be like that, every one who learned HTML would also a web designer.

[Webnauts]

Originally Posted by Webnauts

I posted my point at Google's Webmaster Help Group:
Can invalid code impact rankings? - Crawling, indexing, and ranking | Google Groups

Maybe you would like to bookmark that for the possible coming responses.

Got some responses. You might would like to have a look.

[weslinda]

Well, I think that most people simply won't put the time in to validate their web sites. No matter the issues with quality, accessibility and cross browser friendliness.

Personally, I will work to do things the right way, and hope to encourage others to follow suit.

[Webnauts]

Wes, I do not think the discussion is over yet. I am waiting for Mikes response to my last posts.

[Webnauts]

Can someone explain me what does the point 21 of this article published at SEOmoz suppose to mean? SEOmoz | 24 On-site SEO Checkups For Clueless Developers / Marketers

[paul_foss]

Appears not to be working. The following message is posted:
Looks like the W3C blocked the Multipage Validator. We are trying to install it locally to fix this issue. Please contact us if you would like to help.

[mike]

<title="Northern Cyprus Hotel Manolya"></title> instead of <title>Northern Cyprus Hotel Manolya</title>.

Oh and by the way they had for a first heading tag:

<h1="North Cyprus Hotel Manolya"></h1>, instead of <h1>North Cyprus Hotel Manolya</h1>


Neither one of those will work, no. There is no text where the text is supposed to go. Likewise you can't do <body="here's the body of my content"></body>

But that kind of thing won't even display in a browser, so I'm not sure that is an argument for anything other than 'make sure your pages load'.

As for Guillaume's post on SEOmoz, I'd just say that it's not a lot different that some of the posts made by people here. All he says is run your site thru validation to "cover the major issues you might not have thought of"...

I don't have a problem with that -- It falls well short of endorsing validation as an SEO factor. Just citing the validator as a resource for troubleshooting your HTML - it's great for that.

I really don't have anything else to say on the subject. You guys that want to hang your hat on Validation as an SEO factor -- go right ahead. Be my guest, good luck with it and I hope it serves you well. It's summertime here, have an extra cup of that validation Kool Aid and wear plenty of sunscreen.

That said, I have pretty unambiguous, straightforward information directly from some of the preeminent minds in the SEO business as well as directly from the search engines themselves saying: VALIDATION IS NOT AN SEO FACTOR. Take it for what you will.

Personally I'd say you're better off using your time for something like link building than piddling around in the validator -- but that's just me. You do what you like.

[weslinda]

Why do the messages have to be negative Mike? For an Administrator and a Moderator to be this negative, it's a little suprising.

And we're now so far off topic of the original request, we are simply circling like little kids at a playground getting ready to fight.

Validation is easy, if you know your craft, it's hard if you don't want to take the time to learn HTML, XHTML or CSS. That is when you'll get in trouble. Unfortunately, too many people have had to work around the issues of browsers for so long, no one really knows how to write clean and quality code. This is the main issue the engines won't use valid code as an SEO factor, how could they? 1 out of 1 million sites might validate, and that is a sad state of affairs, but it is a fact of life, and the reason that direct validation won't be a factor in SEO. The results of validation do in turn offer cleaner, more readable code to the search engines which will affect your rankings.

But hey, I'm really hoping that for the sake of the forum, our Administrator and Moderator can kill off the pissing match and be adults and professionals.

[mike]

What's negative? I have no problem with validation proponents advocating validation.

I do have a problem with validation proponents advocating validation as an SEO factor. I have a problem with it because it is untrue.

This has been a dead horse issue for years. I have made post after post after post explaining that search engines do not care whether or not your site validates. This time, I went out and had search engine representatives and SEO experts comment specifically on the issue -- so you no longer have to take my word for it.

It's not a pissing match and I'm not being negative -- it's just a matter of fact. W3C Validation has absolutely, positively no effect whatsoever on Search Engine Optimization.

How is it off-topic by the way? The OP is specifically about this exact issue.

If you want to talk about or extol some other virtues of validation, be my guest. I'm going to draw the line on obvious misinformation though and W3C Validation being an SEO factor = obvious misinformation.

[Webnauts]

OK. This is my last try.

I guess I can not express myself in English so well, but...

Manolya Hotel had on their page:

<h1=North Cyprus Hotel Manolya></h1>

Obviously that could not be displayed in any browser. So it could also not be read by a bot, or?

So, how did I find that invisible thing which if it would be correct coded would have an influence on the page ranking? With the W3C validator. Maybe Mike you do that with a naked eye, but myself I have no time for that.

Is a heading tag one of diverse ranking factors? From my experience: YES!
Don't start here with stories that the factor is too little or so ever please. If you think it is factor value="0", then go on.

Could the search engines recognize that it was a heading? NO!

Did the search engines give Manolya credit for that? NO!

End of story.

[incrediblehelp]

Mike I think I see what John is saying now. His point is that he might have never seen the improper title and H1 tags unless he ran Manolya's website through Validator. For me the title tags and H1 tags are something I am going to always search and fix, regardless of page validation.

[mike]

Well. OK, yes you have to have tags and pages that will render and crawl ( I thought we established that a long time ago -sure feels like I've typed that phrase a lot).

Yes. Absolutely, if you're failing validation because you don't have a title tag -- or a bad title tag -- that's certainly something you should fix.

If W3C is the tool you use to look for those kinds of things, then there you have it... We have a purpose for W3C I agree with. I'll be sure to send Eric Ward an email and let him know we've come up with a problem that W3C is a good solution for. I'm sure he'll be thrilled.

That said, I'm going to have to stick to my guns a bit as I have maintained all along that you need to have a page that loads in browsers and is crawlable. Loading in browsers and being crawlable has nothing at all to do with failing validation for things in my Google example of : "there is no attribute "WIDTH"." and "document type does not allow element "STYLE" here." and "required attribute "TYPE" not specified."

I guess I can easily agree that there are things you can fail validation for that you need to fix. Fixing everything you need to fix and making things validate are 2 different things though.

Fix what you need to fix and move on with your life and your business. There are bigger fish to fry than chasing stupidand meaningless "there is no attribute "WIDTH" messages the W3C is going to give you. That'sbecause there's really nothing at all wrong or harmful with "attribute width" or a thousand other tags that aren't W3C 'canon'.

[Webnauts]

Mike do you know what? WPW Rocks! WPW Rules!

Great post man. You managed to put in words what I am trying to explain for ages.

Thank you very much for that.

I guess we can go now and do some Marketing.

[kgun]

Originally Posted by bj

LOL! Every damn site you pointed at is on legacy software that took massive investment to create. And most of them were waiting for the final version of IE7 to be released, which happened a few short months back, before they'd start on a very expensive rebuilding campaign. One of the guys who is working on Yahoo's rebuild is on one of the lists I subscribe to. It'll be close to a year before these new sites are completely finished, tested and deployed. It's not like any of those sites runs on static html created in Dreamweaver, guys!

I repeat-- Just because things are fine with sloppy crappy code now doesn't mean they will always be. So you head in the sanders keep doing what you're doing, and I'll be laughing all the way to the bank when the boom drops.

Good points.

Originally Posted by mike

That said, I have pretty unambiguous, straightforward information directly from some of the preeminent minds in the SEO business as well as directly from the search engines themselves saying: VALIDATION IS NOT AN SEO FACTOR. Take it for what you will.

From the Forum Rules and Guidelines

"Font Restrictions:

* Do NOT use excessively large font sizes, as it is considered to be "shouting" and will be edited back down to the default size by the moderators"

Why do you as an admin shout?

Originally Posted by Webnauts

Mike do you know what? WPW Rocks! WPW Rules!

Great post man. You managed to put in words what I am trying to explain for ages.

Thank you very much for that.

I guess we can go now and do some Marketing.

Again from the rules and guidelines above.

"Signature Restrictions:

* Maximum length: 3 lines
* No images or cookie requests allowed
* You may link to your site, but no illegal, pornographic or otherwise objectionable links are allowed
* No more than 3 links per signature"

Why do you break this rule?

Originally Posted by kgun

Note in the XHTML, that will also say in an XML world, a page must be well-formed, but it need not be valid.

Who know's the future and dear say that valid markup / code is (will) not be (more) important.

I have an impression that people posting here do not know the difference between a well-formed and a valid document. Well-formedness is about crawlability. A valid document is a well-formed document that adheres to its DTD (Document Type Definition).

[jtracking]

Originally Posted by mike

Why can't I lay off of these threads?

Validated code makes no difference in your search engine ranking whatsoever -- unless it's just so jacked up that the spider can't crawl it.

That is the point blank direct answer. I have sat and watched the engineers mouths move and heard those words actually come out of their mouths at LEAST 10 times.

I'm relatively sure I even have it on video with some of them.

If you want to validate your code -- by all means validate your code. I am not saying -nor have I ever said -- that you shouldn't validate your code. But if you're doing it because you think it is going to impact your SERP... you're wasting your time.

see I don't know if that's true but I believe you - i think the main reason to have validated code is so that your site can be viewed properly on all the different web page readers in the world ie. handhelds, pdas, screen readers and so forth...

oh the structure MAY make it easier or more logical for the search engines but there must be tons of other factors besides validation when it comes to search engines.

[jtracking]

see the person who posted this topic:
http://www.webproworld.com/web-site-...ext-blind.html

wants to be able to hide certain links from users who use screen readers if their blind or have difficult seeing.

The screen reader will read out all the links on the page to the user and the poster only wants a certain amount of link text to be read out by the reader so as not to confuse the blind screen reading user.

If he can figure it out im sure he's going to write it according to standards set for accessibility and not mainly or at all for the search engines...

[kgun]

I think I would share this cite with you:

Character Data.

"Markup consists of XML declarations, document type declarations, elements, entity references, character references, comments, processing instructions (PI's), CDATA, section delimiters , text declarations, and any whitespace outside the document element and not contained within other markup. An example of whitspace that is considered markup is the line feed used between the prolog and the body. Character data, simplys is everything else that is not markup. It is the actual content of the document, which is being described and structured by the markup."

Source:
Robert Richards (2006): "Pro PHP XML and Web Services". Apress.

In an ideal world with only Character data, SE Bots should rank, archive and index character data. In this ideal world, well-formed (valid code) is not a SEO factor.

[Webnauts]

Kgun do you want to say here that SE can not make a difference for example between <h1> and <p> tags?

[kgun]

In a nutshell, I say that in and ideal world, only content could / should rank. Content, though is more than character data, especially in a web 2.0 world with web applications.

In an ideal world it should not be possible to manipulate a SE Bot.

A very interesting scientific project:

• Select a list of (random?) KW's.
• Analyze the SERP's of different SE's.
• Is there a significant (by scientific statistical measueres) difference in the quality of the results judged by a group of professionals?
• Is there a significant difference for SE concentrating on on site versus those concentrating on off site factors?

P.S.

Have you ever seen an ideal world? In an ideal world, there is not a comma error.

[kgun]

Originally Posted by Webnauts

Kgun do you want to say here that SE can not make a difference for example between <h1> and <p> tags?

Of course they can. If one tag is more important for content, that should be a ranking factor. A h1 tag may be more important for human beings, since it is a heading. A p tag indicates a new section.

In an ideal world it should not be important wether the tag is closed or improperly nested.

Picture:
Galileo Galilei's message that the earth is spherical should be independant of whether a tag is closed or improperly nested.

Do you see the difference?

"Heng ham ikke vent til jeg kommer." = "Hang him don't wait til I arrive."

"Heng ham ikke, vent til jeg kommer." = "Don't hang him, wait til I arrive."

Heng ham ikke vent til jeg kommer Title of a Norwegian book.

[Webnauts]

Well, if my story about valid code is a myth, what about this one?

Bruce Clay, President, Bruce Clay, Inc. at the Search Engine Strategies in San Jose 2007 said:

Originally Posted by Bruce Clay

When I redid my site, I switched from tables entirely to CSS. I also made it W3C compliant. That may be something that emerges. I was moderating a panel at adTech and Google said that the cleaner the code, chances are the search engines will get a better idea of what your site is about. From that point of view, go to CSS because it's simpler.

More: Organic Listings Forum: SES, San Jose 2007 » Unofficial SEO Blog

Now what do have to say about this. Doesn't he know what he is talking about? Don't tell me again here that the horse is dead...

[weslinda]

Well, I think his major emphasis is on moving to CSS, with a minor emphasis on Validation. As he thinks it may be a future target, but it doesn't seem to me like he is implying that it is a current factor.

I do see that mentioning validation is a solid idea by Bruce, and I think that it makes life easier in the long run when you move to update a web site.

Good info, and I'd love to hear if there is more on this from him, or other industry "leaders".

[Webnauts]

Did you read the whole statement?

I was moderating a panel at adTech and Google said that the cleaner the code, chances are the search engines will get a better idea of what your site is about.

Come on man. If you had the slightest idea about Software Engineering, you wouldn't even come with the idea of making such a statement. Geez....

[weslinda]

Actually John I'm not trying to read any more into his statements than what was said. He shared in the note that it was perhaps a "future" target. Something clearly not in the forefront of SEO world and certainly not a major variable as yet. Please realize I'm an educated person, and can read quite well.

We all know how this conversation goes on whether clean code means perfect code or readable code. I think those are two different things. Of course I'm on the side of validation, and I think that any job done is worth doing right, but I don't think this is a Bruce signing off on validation being the next great holy grail in SEO or anything along those lines. I think that what his statement is sharing for all of us to plainly see is that if the spider can't read it, then it can't index it.

This isn't as deep as software engineering. I'm all for validation, and I'm all for using CSS, and I think much of the world is moving towards those things to some degree.

Think about what some "non-compliant" code might be. Such as aligning a paragraph within the document or adding a "border="0"" tag will throw a document out of "validation" do you really think that will drastically affect a document and it's placement in a search engine?

More importantly, do you think that's really what Bruce is saying?

[Webnauts]

What does W3C compliant mean? Well formed / clean code, or valid?

[weslinda]

Bruce specifically says, "I also made it W3C Compliant, that "MAY" be something that emerges."

That is the key word here. "MAY"

Bruce is not sharing that W3 Compliance is the latest SEO trick. Or a new ranking factor. Lets not get ahead of what he's saying there.

[weslinda]

I think this is going down the usual winding path...sooo...i'm out. I've shared my thoughts. Everyone can read them as they are.