Interlinking related stories:

http://sphinn.com/story/144077

http://sphinn.com/story/144038

http://sphinn.com/story/143886

Since the opposite to this topic has gotten all the love, let's bring the discussion over here and talk about your reason for Desphinning, if you don't mind?

False. W3C compliance is not a factor.

Why do you think it is NOT a factor? And please, let us not get into the whole 100% compliancy bit, if you do it, you usually go 100%, typically not an a la carte proposition. It's an option but it comes with many disclaimers. I surely wouldn't recommend it.

Can we talk about your 171 combined errors across the 3 sites in your Sphinn Profile? On a serious level? I'd like to go through them line by line and show you which ones are an SEO factor, can we do that?

I'm offering my services free of charge as long as you submit to a public review of your home page document markup, we'll just look at the home pages, I'll keep it simple.

Surely anyone who states that W3C compliance is not a factor would be more than willing to be proven otherwise, correct? Would you mind if I did that?

Related: The OTHER topic. It's a REALLY good one too. :)

W3C Validation for SEO - Myth and Reality

On a side note, you have to appreciate Edward's use of HTML. He's only person (AFAIK) on this site to have handcoded his comment quotes in blockquote tags and found the hidden CSS style.

On a side note, you have to appreciate Edward's use of HTML. He's only person (AFAIK) on this site to have handcoded his comment quotes in blockquote tags and found the hidden CSS style.

Thanks Nick, I thought I saw others using it. This is a perfect opportunity to delve into semantics and SEO. Let's discuss the blockquote element and its purpose real quick.

Unfortunately the blockquote has not received its fair share of exposure when discussing SEO. Why, probably because many folks truly don't understand its meaning. WYSIWYG Editors offer the Indent feature (produces blockquote) which is named incorrectly. Sure, by default blockquote is indented but that is strictly a presentation aspect, what you see.

What I should be using for all of these single p quotes is the p/q elements. If I'm quoting someone and there are multiple paragraphs, then the blockquote is utilized. For single quotes, the p/q elements are utilized.

You could easily turn commenting here into semantic powerhouse. The challenge becomes getting folks to utilize the platform correctly.

Did you also know that blockquote and q typically work in conjunction with the cite element and/or cite attribute. We'd use that to cite the author of the quote if we really wanted to get down to the micro optimization of each byte in each comment. That's what you want to do.

Can we talk about your 171 combined errors across the 3 sites in your Sphinn Profile? On a serious level? I'd like to go through them line by line and show you which ones are an SEO factor, can we do that?

Now there's an offer I'd be foolish to pass on. :) Of those three sites, one is not under my control (State of Search), one I haven't even made a vague hint of an attempt to optimise (Adamus.nl), which leaves my semi-professional blog GreatWebsitesBlog.com.

By all means go through it line by line, I'm actually quite curious to see what you'll find that needs improvement. I'm always ready to learn and to be proved wrong, and you seem like the type of guy I can pick up a few tricks from. So, shoot. :)

That isn't fair, you appear to have fixed half the errors that were present last time I checked, how come?

Validation History for GreatWebsitesBlog.com

So, you pick the site with the least amount of errors, clean up half of them and then expect me to clean up the ones you can't figure out? Nah, I'm not falling for that ole trick. ;)

How about if I just go through those 3 sites and cherry pick the errors I believe will cause indexing challenges?

That isn't fair

Wasn't it Sun Tsu who said that it is only wise to engage in battle when your victory is already assured? 'Fair' has nothing to do with it. :)

'Fair' has nothing to do with it.

Okay, I was trying to be fair in my approach. So, that means I can cherry pick errors from the 3 sites and we can discuss them here, publicly? You're welcome to fix more errors if you'd like because that is my goal, to have you CHASE and FIX errors while I'm doing a public review.

Hey, you just said 'fair' has nothing to do with it. ;)

So, that means I can cherry pick errors from the 3 sites and we can discuss them here, publicly?

Sure. I'll pass on your tips for State of Search to Bas van den Beld, but I won't promise I'll do anything with the undoubtedly massive amount of errors you'll find for Adamus.nl - that site was never meant to rank for anything, personal venting space as it is.

P.S. a comment on your actual article - you quote text from the W3C website to show how important W3C compliance is. That's a bit of a circular argument, don't you think? A bit like the bible saying that the bible is the most important book in mankind's history. (Interpret that as you will.)

That doesn't leave us with much ROI does it? I mean, if the one site left you wish to discuss was never meant to rank for anything, it's an exercise in futility, wouldn't you say so?

I thought you were serious about the topic based on your comments in the other. I mean, you were ready to send me a scan of your Mansa certificate and all. I'd really like to discuss this stuff with folks who are serious and want to figure out what the hell it is that I'm on about. ;)

On Twitter you go on to say...

Here's some free advice: move on. Let it go. This battle isn't yours to win.

First, I'd like to thank you for that free advice but I'll have to pass. Second, this isn't a battle until I run into folks such as yourself and others with similar mindsets.

Not a battle for me to win? I'd rather not approach this from the battle perspective although as you've found out, I am prepared to do battle.

No, I'd rather approach this from an educational perspective, hence the many articles over the years about the subject at WebmasterWorld, my own sites and, other unnamed resources.

It's 2010, the rules have changed. Actually, they haven't changed, they've just become more focused, they have semantics now.

I'd like to provide one example of how adhering to markup standards has SEO benefits. I'd like to put a Donate button here because what I'm about to share with YOU is worthy of a sizeable donation. In my own mind anyway. I'm a legend. :)

Are you familiar with Fragment IDs? I've been using and writing about them for years. Per protocol, they are SUPPOSED to be dereferenced by UAs. Google changed all that last year when they started showing what some refer to as Mini Sitelinks. Do a search for SEO in Google and look at the 3 links under snippet for the Wiki, those are Fragment IDs.

What does this have to do with Validation you're asking? Well, one very blatant error that most SEO'd websites have is Multiple Instances of IDs. In some instances, they are IDs used to target specific sections in the document. If someone does a cut and paste of a element that has a unique ID assigned to it guess what happens? The doc is now invalid and that duplicate ID reference is now a cause for concern. But that is a simple example. Take that and multiply it by 20, 30 or 40 instances and now you have something that CANNOT be interpreted properly and you lose the SEO value of utilizing Fragment IDs.

Does the above example prove one of my points? Anyone?

That's a bit of a circular argument, don't you think?

Who else should I paraphrase? Who's above the W3? Wouldn't they be considered the authority on the topic? They wrote the instructions you're supposed to be following. I'm confused as to your comment. Let's keep Religion out of it, talk about off topic. Oh, keep that Sun character out of it too. :)

Who else should I paraphrase?

The thousands of SEOs with actual hands-on experience with getting non-compliant webpages to page one in the SERPs perhaps?

Who's above the W3?

I think this is the crux of the matter. You see the W3C as the governing deity of the world wide web, whereas most SEOs (myself included) see them more as evangelising priests spreading the gospel of an imaginary nirvana that differs starkly from the actual reality of things.

I do apologise for the horrendous religion-analogies, my brain is a bit fuzzy today and these contrived analogies are all I can come up with. I mean no offense.

P.S. I get a bit thrown off too by seeing your comments expand from when the email notification comes in to when I've replied. I should probably be a bit more patient and allow you to fully flesh out a comment before I jump back in.

I should probably be a bit more patient and allow you to fully flesh out a comment before I jump back in.

Unfortunately there is one larger reply that was busted in the process here so things are a little out of sync. And yes, I am known to go back and read my replies again to make sure there are no typos, grammar issues, something written incorrectly that may cause confusion, etc. So, you might want to wait just a few moments while I get things in order. I usually get it right the first time around. ;)

P.S. Looks like you and I are a in the handful of folks who actually use the HTML Source Editor, heh! That is pretty much the State of the Industry - NO STYLE! - NO SEMANTICS! :)

Now that your previously arrested comment has reappeared, let me respond to that:

When it comes to semantical interpretation, I fully agree that the code needs to be as tidy as you can make it. Whether it's anchor names (now fragment IDs), microformats et al, or proper quotations and citations, I agree that that code needs to be valid for it to convey the full amount of semantic value to the search engines.

I'm probably backtracking here from earlier statements where I said that validation isn't really necessary - as evidently it is - but to clarify, when it comes to stylistic validation errors that have no semantic value I stick with my guns and say fully compliant code is unnecessary.

I would go so far as to say that to get some advanced functionality to work properly in all browsers you may need to employ stylistic code shortcuts that actually break validation.

Your stance seems to be that any invalid code hinders semantic analysis. My stance is that that only applies to a minority of the cases.

The thousands of SEOs with actual hands-on experience with getting non-compliant webpages to page one in the SERPs perhaps?

Okay, I have some catching up to do. Hopefully I won't get arrested again in the process. :)

Your quote above is very amusing and the number one argument I run into every single time. Getting non compliant documents to page one in Google is fairly straightforward, they have error recovery routines in place to accommodate for crap code. The browsers do too but neither of them have thought about every potential error there could be. I know, after seeing thousands of 500 errors from an app we built to parse X/HTML, I know what the UAs have gone through so they don't choke on the non-compliant documents. ;)

I think this is the crux of the matter. You see the W3C as the governing deity of the world wide web, whereas most SEOs (myself included) see them more as evangelising priests spreading the gospel of an imaginary nirvana that differs starkly from the actual reality of things.

No wonder you're quoting Sun. We're talking about the governing body that has written the instructions for most of the web, that which you, I, and everyone else reading along contributes to. You can contribute crap or you can contribute quality. One, or the other, and unfortunately, all things inbetween. I believe that's changing rapidly.

When it comes to semantical interpretation, I fully agree that the code needs to be as tidy as you can make it. Whether it's anchor names (now fragment IDs), microformats et al, or proper quotations and citations, I agree that that code needs to be valid for it to convey the full amount of semantic value to the search engines.

Both Alan and yourself started off making sweeping statements about validation and SEO. Now both of you are cherry picking and back stepping, that doesn't work.

I'm probably backtracking here from earlier statements where I said that validation isn't really necessary - as evidently it is - but to clarify, when it comes to stylistic validation errors that have no semantic value I stick with my guns and say fully compliant code is unnecessary.

Not once has presentation been discussed. We've been talking mainly about markup. You SHOULD not have any stylistic errors, that is a given. THOSE ERRORS SHOULD NOT BE THERE. I can repeat it 100s of times but few will truly understand. The border="0" has been deprecated since the 90s and so have many of the other elements/attributes for those using specific DOCTYPEs. The [font] element causes some of us to feel ill but we understand the old school folks just can't let go of that one. The list goes on and on. All of that stuff should have been corrected pre-2k.

I would go so far as to say that to get some advanced functionality to work properly in all browsers you may need to employ stylistic code shortcuts that actually break validation.

You know what I found in most of the above instances? The original structural markup is broken and someone is attempting to do something to make it work through styling, those are really funny. :)

Your stance seems to be that any invalid code hinders semantic analysis. My stance is that that only applies to a minority of the cases.

Here we go, I'm being misquoted because folks DO NOT READ! I have NEVER, EVER said the above, EVER! I do believe I've done a decent job of separating content from presentation, or at least I thought I did.

False. W3C compliance is not a factor.

That's the reason we're having this discussion right now, let's not forget that. :)