...and the way search engines see them. Though the Google Web Authoring Statistics study is few days old, I had the time to read and comment it only now. The study provides very little valuable or even new information related to SEO but, it does have several valuable points for anyone interested in web site development in general. There are few issues I think are worth addressing in SEO sense:
* The study shows how Google has the capacity to detect and analyze patterns in document structures in very large scale (a billion documents in this case) and use that information to produce something useful... Identifying these patterns is crucial when talking about advanced IR methods, and one can only imagine the use search engines can come up with wide scale page structure analysis.
* The Semantic web is still far, far away... As the saddest part of study commented, page owners (developers etc) widely use structures that have been dead for a decade... With the way we (or most of us) build pages it is very unlikely that the semantic web and, in that sense, advanced search features provided by the semantic web, will ever occur in full scale.
As a summary I can say the study is a very impressive, but also very depressing to read if you know a thing or two about web development and standards.
2K - As a Dreamweaver 8 user, I can assure you that DW couldn't make a page that validates if it's life depended on it. You've got to spend a great amount of time in the code for all but the most basic pages in order to create valid XHTML with it. Frontpage 03, as I understand it, is even worse.
Beyond that, pages made from a lot of older blog templates or layout templates bought online or even Yahoo! storefront templates don't validate...
The web has to allow for the freedom for anyone to create simply and easily. So far, standards and validation are hinderances to those processes, thus the conflict.
Here,here for the first phrase...
The second one though amazes me... Currently all pro-level web design tools (dreamweaver, frontpage2003, vs.net etc) provide recommended and easy-to-apply workflow and methods that enable production of valid and well structured web applications (with well known exceptions like object/embed element usage in dw). Entry level tools (like wordpress) have relatively good inbuild support for standards...
All it takes is opening the help/manual and studying how to efficiently use the tools provided. But I know, only geeks (like me) read the docs... ;)
It was recently pointed out to me that using tables or even stting absolute positioning is bad form. It prevents mobile users from seeing the whole picture.
rmccarley, when targeting mobile devices / user agents there is much more than screen size to worry. For example, bandwidth/download times, memory consuption etc. limits are something that are a major topic in all mobile development. As for tables, absolute positioning... You might want to try Opera (and set it in Small Screen mode). For more info, see https://www.opera.com/products/mobile/smallscreen/
rand, I'm a longterm dw user too but I have no problems whatsoever making pages to match standards. Same applies to FP... Most likely the difference comes from different approaches and working methods. DW & FP are tools and do only what you tell them to do. Correctly used they are very efficient, but otherwise used they can cause lots of headaches.
In any event, the standards people lose because they don't have the weight of majority use behind them. I don't think so, those who loose are those who aren't able to take advantage of the possibilites, and see this in a bigger perspective. By creating semantic web pages you can build extra value into the web site without extra work, which increases the customer satisfaction. Using yesterdays methods will only work as long as the customers aren't enlightened, underestimating the customers knowledge is never a recommended strategy. This is simply good, old, marketing.
All the ranting and raving in the world isn't going to change the fact that the Web is not going to rely on all the extra coding that the standards community wants to impose on people. Extra coding?
Well, of course, those who oppose the use of tables "for technical reasons" are wrong, but I'm not going to repeat the obvious details.
Please tell me the obvious Michael. I guess table designs don't increase the possibility for keyword fights or burning the tree vs a semantic web page either?
Still noticing you are using a quite interesting debating techniques Martinez. Instead of being able to discuss objective and focus on the case, you are using well known prevailing and social evidencing techniques well known in debates where the focus is on position.
"..but I'm not going to repeat the obvious details.." A quite common prevailing technique, make the other part look less smart than you (you are the expert and they don't know the obvious).
Another prevailing and social evidencing technique: "You're in the minority, you've collectively so far proven to be an unpersuasive minority.."
Social evidencing technique is the same as saying "We are the truth" (no one wants it, we the web community, people are not interested etc.).
I will now step down to your level of debating before I leave the discussion, and make Walter Lippmanns words about social evidencing to mine: "If everybody thinks the same, nobody are thinking very much".
Mr. Martinez,
I agree with your observation, that most people are not concerned with standards. Google openly admits to it. Less than 1% of the W.W.W. is W3C compliant.
Unfortunately, after slowly learning about "most" people, the masses, I see more clearly why the Forefathers of the USA chose to establish a Republic over a Democracy. They were correct in their assessment and following action, the Constitution of the United States.
The masses WILL set a "standard", of sorts. Not good and I now understand why more clearly.
Since over 90% of the W.W.W. is out of compliance, I can site you many, many examples of "standards-not" and its relationship to "SEO" and the end-users of the web pages.
Ms. Whalen's website's righthand column blows up under certain "standards". Though, I think that she has finally fixed that.
Target Dot Coms drop down menu dropped down behind their flash under certain "standards". I emailed them shortly before Christmas and they complied to the "standards" on or about January 8th.
Ethical SEO's website's text flies away, way past their righthand border, the last time I checked.
An established SEO firm recently touted their acquisition of a large account. I visited their website only to be confronted with their text over some of their other text on their home page. I couldn't read their table-based website under certain "standards". Good picture.
This web page has over 40 warnings. I just checked. No big deal. Doesn't impress me. However, I'm not one of the 99.9% of the masses.
I'm seeking web page optimizers for my web design clients. I'm not having too much success in finding any. If there were a "SEO" standard, my search would be over. Not going to happen. The masses "standard" dictates that!
If there isn't a "standard" set in sentence structure, the sentence, "I'm seeking web page optimizers", probably would appear unreadable. Then again, is there a "standard" for "web page optimizers" such that the writer and the reader of the sentence can communicate?
My industry (electronics) recently set a "standard", no lead in electronic components.
Do "most" people know about this standard? Do they really care? Probably not.
If this standard was not established, these very same people would continue to tell their off-spring, "eat lead". Not a big deal. It's only their kids.
Though the Republic of the U.S.A. is by far free from flaw, now that I understand "most" people, I can only envision a disaster, it instead being a Democracy.
The same applies to the W.W.W. But then, this is how I view the necessity of "standards" in website design and promotion and is not that of the masses. And the W.W.W. confirms that.
It was recently pointed out to me that using tables or even stting absolute positioning is bad form. It prevents mobile users from seeing the whole picture. This is a growing market that is getting more important and brings things back to usability again...
benj - no worries about starting a war - we love a good discussion.
I agree with all of your points (although I'm not putting up a "screw off" sign on my site just because we use tables).
The thing about validation and code is, it's just as important to the web's development overall that creating web pages is easy and hassle-free as it is that standards exist to make browsers and reading the web easy.
Sadly, my "coding" skills (by which I mean my editing of what Dreamweaver creates for me to make it workable) could never achieve full validation and do everything I wanted them to do. Are there people who are locked out of SEOmoz due to this - maybe... But, millions of sites and pages couldn't exist if only the skilled coders who could fully validate, use successful CSS and still include the full functionality of the design were the only ones building.
The web has to allow for the freedom for anyone to create simply and easily. So far, standards and validation are hinderances to those processes, thus the conflict.
The standards obsessed folks in webdev have always maintained that the search engines value proper code and W3C standards (including my favorite, most ridiculous bit of code - the need for a separate tag identifying any uncommon acronyms and spelling them out anytime they're used on the page).
Maybe this will help to open eyes and let people know that validation does not equal rankings, nor is it even worthy of a boost. As long as the search engines can properly index the text on your page, you're fine.
Of course, having pretty markup can help you get linked to by the standards-obsessed communities on the web, so I suppose, in a roundabout way, that could help rankings :)
Frames? What are frames? ;)
Standards no one wants to use are not standards at all. They're just meaningless fluff. Are you everybody, and are you the Web community?
The study really shows just how out-of-touch the standards writers are. They need to get their noses out of the air and work up some standards that reflect how Web page designers put out their product. So, what about those with disabilites, do you feel that you have the right to dictate something that will make the web less available for them? What about usability? Frames goes against that. Claiming other are using it isn't an argument, it's just an excuse.
Perhaps time for some Web designers and SEO's to wake up from their narrow world built on yesterdays solutions, and start to think about all of their visitors. It's so easy to make something that will make the life easier also for those with disabilites, and it will also make the life easier for you as a web developer.
What I find quite interesting is that both of you are attacking the standards and in some way making fun of them (and the people who find them valuable), but you don't present something better. Complaining doesn't solve anything or move the world forward..
Maybe you instead should focus on the case and not the "standards-people", and explain what favours your solutions?
sigh... yet another coder vs non-coder fight ;)
I agree with rand about validation part in SEO sense. Validation alone does not give any kind of boost directly - it's the structure (and mainly semantics) of content that matter the most. Valid code though is usually better organised/structured than invalid, and thus it has generally better changes in getting ranked well.... Instead of thinking valid markup as boring code, it is a simple and relatively efficient method for better copywriting.
There are of course broader, non-SEO views that defense the role of making valid, standard compliant pages like decreased development and maintenance expenses (something I think all agree to be very important; or show me a person who doesn't like money ;), and like mano70 said accessibility issues...
Also legislation is taking afoot especially in Europe (in US you have section 508). For example in Finland (and to my knowledge also in many other european countries) there already is /will be a recommendations/legislative iniatives that define minimum requirements for government/EU financed web sites. These minimum requirements are standards - sites that conform XHTML/CSS/WAI etc specifications. And this has effect also for SEO/SEM because especially in Europe it is hard to find any iniative that somehow doesn't involve EU / governement financing in some form. (Completely different story is how well this is managed/followed since most decision makers are non-tech people)...
Now, I don't want to start a war on a site I love so much, but Mr Martinez is just horribly wrong.
While coding to standards may not affect SEO/SEM in any way, they most definitely do affect the end user and the site owner.
If you are writing using tables, font tags, frames and other such junk:
a) You are producing more code, increasing download times. Maybe only slightly, but this can affect the end user.
b) reducing accessibility for text readers, browsers for the physically impaired, etc.
c) not guaranteeing cross browser compatibility.
Why does this matter?
Each of those points above is reducing your ability to reach the widest possible audience. If your site slows one person down, is unreadable to one visually impaired person, or fails in one browser, then you have immediately lost those people, and the cash they had to give you.
If it is a personal site, a fun site, or not designed to generate revenue, all very well. However, if it is a commercial site designed to generate any kind of return, be it direct sales, or an ROI from reduced support, customer service, whatever, then you need to:
a) Be aware that you are shutting out a percentage of your audience (if it is your own site)
b) Ensure your boss knows the loss of income caused by your actions (if it is for your company site)
c) Make sure your client is fully aware your work is below par and will cause a loss of income (if you are being paid to develop a site for a third party)
d) put a big notice on it telling visitors if the site doesn't work for them, you really don't care, you don't need their money.
Then there's the potential costs of someone else having to go through your tag soup in a year, when the site needs an overhaul, but lets not go there.
So, yeah, standards make no difference to SEO/SEM, but they sure make a difference to the bottom line.
Edit: Dammit, 2K got in there while I was typing this out....
As long as the search engines can properly index the text on your page, you're fine.
Fair enough but if your pages validate you know the SEs can read them. How do you know otherwise? I mean, personally I like to validate my pages for reasons beyond SEO and I'm glad I do. I'm human and make really dumb mistakes sometimes.
That aside, the point of the standards was to stop having to support one browser over another or building 2 versions of the same site to appease everybody. That has been mostly accomplished.
Standards no one wants to use are not standards at all.
This is a two-edged sword. It's the same as with legislation - some laws are worse than others and people disobey, but they are/should be still as bindable in all occasions. If people don't follow recommendations for common rules the output is chaos... take for example driving in red lights or other minor traffic violations; everybody knows what happens when enough people start behaving like that...
Another similar story, but closely related to webdevelopment would be SQL. As most know, it has an ANSI and ISO-standard status. In real life each database manufacturer has their own dialects of SQL which make complex cross-database solutions very inefficient to develop...
(as you might have figured from above rants, I'm a standards-geek ;)
One factor the study didn't tell was how old the documents were (publishing date) and how the structural trends have changed based on publishing date... I'm willing to bet the analysis contained lots of "graveyard files"... Current web editors / tools (dreamweaver 8, frontpage 2003,typo3 etc) are able to provide relatively valid and well structured markup/code even if the users were total dummies (assuming they know how to use the tool and don't mix with the default settings ;).