I remember the old search engine days of Magellan, WebCrawler, Excite, AltaVista, Lycos and Infoseek. I tried all of them and they all had some strengths and some weaknesses. Then a guy at work told me about Google, and how "relevant" the search results were. Well, I gave it a shot and sure enough, Google gave me a nice selection of websites on the 1st page that were either exactly or close to what I was looking for. I was stubborn, and still used some of the oldies for awhile, but eventually gave them up because of Google's great results. Over the last few years it seems that Google has strayed somewhat from giving me results like they used to. Some likely causes are: 1- Webmasters getting smart to Google search technology and taking advantage of it, 2-Google being involved in far more enterprises other than search engines thereby diluting their core competency, 3- The web is getting much more complex and loaded with a lot more data/information, 4- The advent of Adwords and Google's interest in gaining more momentum in that business venture and perhaps (only a theory) catering more to those customers versus the everyday Google "Search" customer by keeping some very relevant sites lower in SERP. Let me emphasize, #4 is only a possibility and by no means fact. So what is Google doing about going back to their expertise and getting more relevant search results. One claim by Google was to hit paid link sites hard so that they eliminate irrelevant results, i.e. Google called this gaming the system,but was this their real motive or was Google really trying to boost it's Adwords revenue? I really don't believe that TLA's, directories or other paid link sites had all irrelevant links, in fact they probably had many very relevant links to sites. Of course there will always be some links that are irrelevant in those sites. I don't believe that hitting these sites with severe PR downgrades is the answer to getting relevant search engine results! I am not an owner of this type of site, nor plan on becoming one, I am only attesting that there has to be much better ways for Google to go back to their roots of serving the world the most "relevant" search results. Has Google ever surveyed the general population of a country or the world? I think the survey could be very complex, but I like to keep things simple. What if Google did a random survey (not webmasters) of the general population with this one question: Dear customer, when you use Google for search, how many of the sites on the 1st page have met your search expectations? 1,2, 3...10, don't know. By doing this, they could set a baseline. That way when they launch another algorithm change, after a period of time they could resurvey the population. If the statistics show better search relevancy, they know they are on the right track! If Google is to continue to dominate the search engines, they need to focus on the right methods to return the most "relevant" search results. By using a statistical method to validate their search engine changes such as surveys, they will continue to make their customers, the search engine users, very happy. Just my 2 cents!