This is a pretty cool experiment conducted by the international Web Developers Network. http://www.iwdn.net/showthread.php?t=5914 It gives pretty good insights regarding what gets picked up by the search engines and what doesn´t. Especially the part regarding the keyword spam was very revealing.. Check it out
To be honest I didn't see anything earth shifting.. some interesting indexing tidbits here and there... but not a whole lot... I didn't see a lot of interest in there on Spam either? In the end simply a SINGLE site experiment into indexation. Much of indexation/crawl rates is based backlinks, age/authority, freshness etc... there wasn't much covered on activities in these areas as well. One can't really speak to the efficacy of TITLE tags, as they have, without tracking rankings and estblishing link profile and content development profiles for the site. So many of the assumptions are based on a VERY limited experiment. Kudos for thinking of it, but poor marks on execution for them I would say...
Actually, that's what this article was meant to be. A exercise that explores what gets indexed and what doesn't, so Web developers can make better decisions regarding how and where they use keywords in the markup of a site. It's not particularly earth-shattering at all, but when one of our members on IWDN asked about whether certain items within HTML are read by search engines, I couldn't find any documentation that made it clear what did and didn't get indexed, and so the experiment was born. Someone with advanced knowledge in SEO will find a few interesting tidbits, and that's all. An SEO novice may find every test case is new information for them. You take out of it what you need, and leave the rest The experiment was posted on eight different sites simultaneously in order to ensure the test page was picked up (most SEs picked up 3-4 of the copies). We put the control word last on the list so we would be reasonably sure the 14 test cases were seen before the control word. Remember, this isn't an experiment to see how effective a method is. It is simply an experiment to see whether the vehicle for keyword placement is given any SE attention whatsoever. The experiment did not test TITLE tags in any way shape or form. We tested title attributes (there were three title attribute test cases - take a look at the source code and it should make more sense). Just wanted to make that clear, in case anyone else was confused on this point. Having a better understand of what took place in this experiment, does the execution seem at all more meaningful to you now? This was a baseline experiment meant to learn whether certain objects, attributes and content display mechanisms would allow or reject search engines from adding text to their indices. Search results display SE indices, so pulling results was that easy. We're very interested in seeing what types of offshoot experiments can come from this one, simple experiment, and if anyone wants to test this one themselves, the source is available to everyone. I'm working on the first offshoot now. I'm creating pages with a mixture of hexidecimal text and regular text (with a control page) to see whether Yahoo! and Wisenut will return search results with hexidecimal text removed, as we anticipate will happen after seeing the results of the current experiment. More suggestions are welcome! Thanks. I'm glad you got something out of this