RE: [WSG] SEO, Semantics, and Web Standards
It's probably safer to say at this point that semantic coding certainly can't hurt your search engine ranking :) Chris -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jono Sent: Thursday, March 03, 2005 4:49 PM To: wsg@webstandardsgroup.org Subject: Re: [WSG] SEO, Semantics, and Web Standards I've always been fascinated by the search results for for certain words on Google. Sometimes it is hard to tell - by viewing page source - why the top result is in fact the top result. For exampel, try searching for Fried Chicken on Google. Take a look at the top two results, and then take a look at their code. You'll discover that Number 2 is much worse than number 1, which is not very good to begin with. If clean code has anything to do with SEO, this is definitely a good case study. The third result has the worst coding of them all... I'm not sure how it even made it to number 3? There is also a lop-sided brand competition between the top two ( see 1 2 in Google search results for Fried Chicken ), which probably has a lot to do with it as well. On Feb 24, 2005, at 7:48 PM, heretic wrote: So can I hear it from the experts (ie: you guys) what the truth behind SEO really is. Are semantics worth anything? Well, it's hard to find the hard data to back this up... but what does come up consistently is this: semantically-correct markup will improve rankings. I can't say definitively how much it will improve, but it does seem to work :) ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] SEO, Semantics, and Web Standards
I've always been fascinated by the search results for for certain words on Google. Sometimes it is hard to tell - by viewing page source - why the top result is in fact the top result. For exampel, try searching for Fried Chicken on Google. Take a look at the top two results, and then take a look at their code. You'll discover that Number 2 is much worse than number 1, which is not very good to begin with. If clean code has anything to do with SEO, this is definitely a good case study. The third result has the worst coding of them all... I'm not sure how it even made it to number 3? There is also a lop-sided brand competition between the top two ( see 1 2 in Google search results for Fried Chicken ), which probably has a lot to do with it as well. On Feb 24, 2005, at 7:48 PM, heretic wrote: So can I hear it from the experts (ie: you guys) what the truth behind SEO really is. Are semantics worth anything? Well, it's hard to find the hard data to back this up... but what does come up consistently is this: semantically-correct markup will improve rankings. I can't say definitively how much it will improve, but it does seem to work :) ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
[WSG] SEO, Semantics, and Web Standards
'Lo again I recently paid a visit to a certain SEO forum and had a look at the forums there. Whilst reading the threads, I couldn't help but be shocked and appalled at the FUD being spread there. To quote one (altered slightly, so you guys can't reverse-google for it ;) ) ...yeah, it also helps to copy your meta keywords into a h1 tag at the top of your page and hide it wiv jscript Naturally, that same person also seemed to have Mozilla and Firefox confused. The site had a few articles on how ethical seo 'doesn't work', and activly promotes bad practices, non-semantics, and praises HTML4.01 Transitional because it lets you do anything you want! There are claims that sites manually submitted to Google reduces your pagerank by a few points because it didn't find your site by being linked to it. It also has some kind of comparison between ethical SEO (a website that complied with spec (but not semantics, it still used the font element and tables for layout, but it did validate), verses a normal SEO site (IE-DOM eat your heart out) and claimed that the latter site would get higher rankings in Google So can I hear it from the experts (ie: you guys) what the truth behind SEO really is. Are semantics worth anything? ...Is it worth sinking so low as to use 302s on Googlebot and display: none; on keyword paragraphs, what about mini-linkfarms (a table style=display: none; (not a ul) full of hyperlinks to other pages on other sites full of similar content) just to get a slightly higher pagerank? IMHO, ranking is more dependant on your brand strength, rather than dirty and underhand tricks. Besides, I thought Google was a semantic web bot. Perhaps we should petition Google to produce a Semantic Cralwer that looks for a special HTTP Response Header or page meta telling the spider that the page is semantic and doesn't use any tricks. Comments? -- -David R ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] SEO, Semantics, and Web Standards
hi, I recently paid a visit to a certain SEO forum and had a look at the forums there. Whilst reading the threads, I couldn't help but be shocked and appalled at the FUD being spread there. Most SEO seems to be either complete FUD or ideas with very questionable sources/backup info. Separating fact and fiction is hard, since not all search engines are as helpful as Google (http://www.google.com.au/webmasters/seo.html). So can I hear it from the experts (ie: you guys) what the truth behind SEO really is. Are semantics worth anything? Well, it's hard to find the hard data to back this up... but what does come up consistently is this: semantically-correct markup will improve rankings. I can't say definitively how much it will improve, but it does seem to work :) Specific tags which are apparently weighted quite highly: - heading tags, esp a page's H1 - the link text in a/a tags - the contents of a page's title/title The other general principle I have heard many times is that search bots will see pages much the same way as a text-only browser like Lynx. So, anything requiring JavaScript (including js-only popups) probably won't be indexed, convoluted URLs (ie. really long nasty ones from dynamic apps) are often truncated, any media object with no alternative content won't be indexed. Flash content is invisible, at least anecdotally (my favourite coffee supplier doesn't turn up in Google at all for that reason). I don't have hard data to back this stuff up - but from what I've seen I do believe it :) ...Is it worth sinking so low as to use 302s on Googlebot and display: none; on keyword paragraphs, what about mini-linkfarms (a table style=display: none; (not a ul) full of hyperlinks to other pages on other sites full of similar content) just to get a slightly higher pagerank? Theoretically speaking you can get caught out using such techniques and find yourself blacklisted. In reality I don't know of any specific cases, but personally I just wouldn't take the risk. Besides that, you're better off spending that time and effort creating compelling content which people will link to :) Perhaps we should petition Google to produce a Semantic Cralwer that looks for a special HTTP Response Header or page meta telling the spider that the page is semantic and doesn't use any tricks. ...which would only work for bots/searches which honoured that tag. MSN apparently doesn't honor nofollow or robots.txt instructions, based on a conversation I had with a colleague yesterday (their aging server is being hammered by MSN, using up valuable sessions). Hope this helps... h -- --- http://cheshrkat.blogspot.com/ --- The future has arrived; it's just not --- evenly distributed. - William Gibson ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
