Jan
Now, I’m not one for trying to rehash and rekindle old arguments. Matt Cutts has already stated that currently valid code isn’t taken into consideration for rankings. He also made mention that 40% of sites have code errors, so it would be hard to simply discount that content from the rankings.
I will argue, however, that having Valid XHTML / CSS or close to it should be and in the future may be an indicator of trust to the search engines. I’m not saying valid code will ever be a huge indicator of trust, but maybe just a smaller piece of the trust puzzle, like having a privacy policy.
In the not-too-distant future, close to valid code will be a mark of trustworthiness, mainly because its an indicator of quality. If a webmaster takes the time to make sure his code validates, generally that webmaster takes an interest in producing quality content as well. As far as search engines are concerned, good quality sites should be trustworthy sites, if only for the reason that most spammers and arbitragers probably aren’t going to take the time to fix any validation issues. The name of the game with spam and arbi is to get many sites up at once with “auto content generation” (to put it politely) which is less likely to be clean code . This being said I don’t think any small validation errors should take away from your trust (people do make coding mistakes), but the WYSISWYG code generation or just plain sloppy coding is a far cry from the quality of code that I’m talking about.
My main interest in valid XHTML and valid CSS is accessibility, but hey, if I take the time to make my site accessible by search engines and humans, shouldn’t I be rewarded for that somehow? So how about it Google, can I get some more trust over here? 
» | Posted by Brian under 4 Responses to “Valid Code as an Indicator of SE Trust”
Leave a Reply
type keywords | hit enter
- Professional Search Engine Optimization with PHP - Review
- Ebizzsol - My Secret Weapon
- Advice from the Trenches: Advertising on Content Networks
- For Better Goals, First Define Success
- Andy Hagans is a Good Dancer
- Thats sooo 2006: Sensational Titles
- Domain Names & Usability
- NCAA Tournament Coverage has Poor Usability
- Usability Makeover Series - Twiddy Vacation Rentals
- 2 Copies of Pages Indexed in Google
January 19th, 2007 at 10:01 am
There is no correlation between the quality of a Web site’s HTML code (as measured by the W3C’s arbitrary and mostly ignored standard) and the value that other people place upon the Web site.
The idea that a search engine should weight or adjust relevance scores on the basis of how clean the HTML code just flies in the face of all sense.
At least PageRank has the poor excuse of assuming that people actually choose to link on the basis of perceived value and for no other reason.
General purpose search won’t benefit from incorporating arbitrary coding standards that most Webmasters are not even aware of into their ranking algorithms.
But where you may see some impact, if it’s not there yet, is in accessible search, like Google’s accessible search. There it makes sense to look under the hood because the purpose of the tool is to help people with specific browser needs to find content that is most compatible with those needs.
January 19th, 2007 at 11:21 am
Thanks for chiming in Michael. Like I said, I’m not interested in renewing the debate, alls I’m reallying saying is that I could see Google implementing this. I think there are a lot of ranking factors that fly in the face of all sense. After all its really about returning the most relevant content to persons query which may negate the “trust” factor all together. Really, what does trust have to do with deciding who’s content is more relevant than who’s? For now we just have to deal with the fact that there are arbitrary standards applied to Google’s ranking algorithm, and I can see this being one more on the list.
January 19th, 2007 at 1:10 pm
I think the trust factor is important because of all the manipulation that goes on. Microsoft’s Live Search actually empowers the user to filter on the basis of limited trust, which is a great feature of their service.
What I would like to see Google and the other services do, actually, is provide us with query operators that allow us to specify “W3C-compliant”, “Accessible-only”, etc. results because I think it underserves the community to relegate these types of queries to secondary (and underpromoted) tools.
Compliance advocates would, I am sure, welcome the greater visibility such enhancements to the query process confer upon their goals. But more importantly I think it would help educate the surfing public in the value of searching for content on the basis of other presentation factors.
I don’t really oppose the adoption of standards. I just feel that right now standards advocates aren’t looking at the real picture. It’s not about how much we would all benefit from standardization — it’s about how we really are not all in line with anything.
Ignorance rules the Internet. Isn’t that ironic for the Information Superhighway?
January 19th, 2007 at 1:18 pm
I’d turn your argument on its side and say the opposite. Who’s more likely to have perfect code, a professional spammer or a simple mom & pop site using a WYSIWYG editor? The professional spammer who’s written his own scraper script and auto-publishing to millions of subdomains will have a higher probability of even knowing what w3 code is than your average person actually writing content.