![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
Hello, I would like to interject something into your discussion. We are quite aware of the frustration that many people face when trying to 'get' relevant information from a search engine. In all actuality, there is really nothing wrong with the method in which keywords are _supposed_ to function. The problem is that most companies _SPAM_ web indexes to try to get hits to their client site(s). This practice of spamming an index is the first part of the problem. The second is the search engine companies themselves. InfoSeek for example, in the beginning of the summer change the way they 'produce' results. In effect, sites that were on the top are now in the middle or the bottom. Furthermore, there are no companies in the industry that make their livelihood by making sure their clients are on the top of the index. Books have been written on the subject. One of our companies functions is web hosting, another is search engine submissions. I hear about this problem ALL the time. Creating a _NEW_ thing is not the solution. Forcing the world into compliance with the existing would be much better. Most search engine companies are now attempting to base searches on "Key Phrases" not just a word. Others are adding relevancy. Does the keyword (phrase also appear on the text of the page) is it in the title as well? All these things will determine search results. Web companies that do not comply with the rules, jam thousands of word into graphic comments on the page. Create page titles with hundreds of words in them. I hope you get the idea. Companies like AltaVista check for this cyber fraud and once caught will not allow future submissions to their index from the submitter. If you invent something new, people will look for a way around it. Unfortunately we are not talking about people working for the betterment of the web as a community project, just greedy people causing search engine ruin because they want HITS and they don't care the cost..... -Al -----Original Message----- From: Fisher Mark <fisherm at tce.com> To: Valdis.Kletnieks at vt.edu <Valdis.Kletnieks at vt.edu>; scott at heelspurs.com <scott at heelspurs.com>; Christopher D. Maddux <cmaddux at landorf.org>; Richard Czerwien <tippleru at landorf.org>; george at karavitis.com <george at karavitis.com>; 'Timothy Glenn Stockstill' <timothy at STOCKSTILL.COM> Cc: ietf at ietf.org <ietf at ietf.org> Date: Monday, August 24, 1998 4:43 PM Subject: RE: Dewey Decimal Web Searching >A different approach to the search problem (AKA "drinking from the >firehose") is to use a search engine that uses the # of citations of a >particular URL (links to that URL) as part of the relevancy ranking process. >Google (<http://google.stanford.edu/>) is one such engine. Google has >become my main search engine because 90-95% of the time, if the information >is on the Web, it will show up in the top 10 results (and usually in the top >3). It isn't a panacea, but I expect technology like Google's to be adopted >by the other search engines as the word of Google's relevancy ranking >accuracy spreads. >========================================================== >Mark Leighton Fisher Thomson Consumer Electronics >fisherm at indy.tce.com Indianapolis, IN >"Browser Torture Specialist, First Class" >
Note Well: Messages sent to this mailing list are the opinions of the senders and do not imply endorsement by the IETF.