Why webmasters don't like Lycos Retriever
Yesterday I was looking at a new beta service from Lycos, called Lycos Retriever, which they claim is "the Web's first information fusion engine". It builds directory pages on topics selected after analysing real user search queries, and populates those pages with encyclopeadic content from websites. Lycos Retriever has not proved popular with the webmaster community, and I can understand why.
If Joe Bloggs made a directory web site consisting of flat HTML pages, each page of which was optimised around one specific keyword or phrase, and where all of the content consisted of 200+ word excerpts scraped from third-party sites without their express permission, then loaded it with Google and other types of contextual advertising, they'd be breaching their AdSense terms & conditions and penalised by most search engines (if caught) as a spammy made-for-AdSense site.
It is difficult to see what Lycos have done that is any different to that - and make no mistake these pages are heavily optimised to feature in the search results of other search engines for the relevant keyword or phrase. The <TITLE> tag, for example, only mentions the keyword phrase, with no Lycos branding whatsoever. The only difference between Lycos and Joe Bloggs seems to be the fact that Lycos began with the head start of having a high PageRank for their root domain name, a big web brand name to exploit, and the ability to slap a 'beta' label on it and have people treat it differently.
The site also serves sponsored ad listings against blank search results pages, which is pretty much against the spirit if not the letter of the terms and conditions of most contextual advertising for smaller web publishers.
So for me the question remains open - is this an interesting attempt to use computing power to solve scalability and fill the directory/encyclopaedia information space that hand-crafted directories of topics like Yahoo!'s or DMOZ failed to do, or is it just content theft to generate advertising revenue?