For some reason the logic of using "backlinking" as "major points" by search engines baffles me. I see sites selling "backlinks". Isn't that like selling votes? If search engines were able to use them as a criteria to rank and no one knew that was a criteria, then I would think that may be effective. But once the cat is out of the bag, then it is time to think of something else. You have to many cheaters. Now that people know that this is of great importance, then it skews the results if they acquire them in an unnatural way. It seems like voting without anyone monitoring the polls. Won't it be abused more every year, and thereyby skewing results even further? And won't those who have worked so diligently to obtain backlinks try to maintain this system, despite the fact the search results become less relevant? Am I missing something? This is what all the PHD's in Mathematics give their Search Engine bosses?...a vote by a backlink? I could have come up with that...and that would be after my 1.7 GPA after my first semester in the fraternity. Come on...please explain to me how this will give the searcher better search reults and how those results will be improved in the future. jocknotech
And a better solution would be....? Figure that out and you can have yourself the next big seach engine ipo
read up about topic sensitive pagerank (TSPR). the algo does go beyone votes from links but at present I think it's heading in the TSPR direction or some adaption there of.
clustering and topic distillation are where things may start heading... Mike Grehan wrote a good article about Teoma & topic distillation. http://www.searchguild.com/topic_distillation.pdf
Wow! Thank you! I just searched for TSPR and well before I got to the algos in Greek I was quite humbled. Apparently we have some smart guys at Stanford on TSPR. Their math skills are much better than mine. Good. When I typed in "clustering and topic distillation" I tried to read an Abstract by a couple of guys from India. I should have studied harder. I tried to read it.."Topic distillation is becoming common in Web search engines, but the best-known algorithms model the Web graph at a coarse grain, with whole pages as single nodes." I do not know what it means but I can only think that it means that they parse through the content and identify the content the best they can. What does that mean to the common man-clustering and topic distillation?
It means they're simply applying a different set of rules to the same data (if not a subset). I'm only working with what you posted, but rather than word 'clusters', they're talking page clusters. Rather than thematic associations between word clusters (or to assess word clusters) they're deriving page clusters. It's pretty much the same old samo - theme concentration at the site hub, with associated/derivatives encountered as you progress outward through the link structure (both internal & IBLs). Just my 2pence. Actually, IMHO it (or a close cousin) may have just arrived. Make that 3pence.
From what I understand algos such as this are already being worked into the search engines slowly but surely. Personally I think its great news for the honest folk who are just trying to run a business. It's going to change the rules just enough that some of the more dirty tactics will be muscled out and anyone trying to build real estate online for their user's to land which actually offer real value and a true user experience will have a better chance of ranking where they should.