Forum Moderators: open
I've just been surfing around on lycos and found a possible hidden use for their 'Trendometer' Scripts.
Visit this page:
[uk.lycos.co.uk...]
You can type in two search 'words' (not terms - if you type in a term on the first word will be used.)
eg. type in 'Szlampy' then 'Jamsy' and you'll notice that I get typically more votes than Jamsy (perhaps because I talk better sh*t than he does?)
This voting system is based on a ranking system that Lycos uses to evaluate keywords. eg. If someone does a wild search on the web for 'Lingerie' then that keyword may get 1 vote, whilst if someone is searching a shopping area of lycos and refines a search using the keyword 'Lingerie' then the keyword may gain 3 votes, etc.
This tool isn't as useful as the 'inventory' but can be quite useful in determining the popularity of a keyword, so I thought I'd tell you all about it.
Regards, Szlampy
[edited by: heini at 3:42 pm (utc) on July 4, 2003]
[edit reason] removed url per TOS / thank you [/edit]
Thanks for your very interesting first post. However, I don't think the tool is going to be useful to too many people until we know a few more things. Any light you can shed on these subjects would be most helpful for WebmasterWorld members to assess whether there's any merit to the Lycos UK Trendometer.
(1)
This voting system is based on a ranking system that Lycos uses to evaluate keywords. eg. If someone does a wild search on the web for 'Lingerie' then that keyword may get 1 vote, whilst if someone is searching a shopping area of lycos and refines a search using the keyword 'Lingerie' then the keyword may gain 3 votes, etc.
How do you know this? I've scoured all the major search engines for any information re "Lycos" and "Trendometer" and have found...only this post. (Cheers to Google and Inktomi.) Are you making an assumption about the source of the Trendometer results data, or did you read it somewhere?
(2) The Trendometer appears to be a Googlefight-style tool to pit two terms against each other with the "winner" being the one with the higher number. However, such a tool usually measures keyword competition, the number of results delivered for each term. You're claiming that this tool measures keyword popularity, the number of searches, modulated by some special scoring system. How do you know this?
(3) Even if we assume that there is some relationship between the Trendometer results and Lycos UK keyword popularity data, the results are too dodgy for serious use. For example, I typed "love" into the first field and "sex" into the second. The results:
1. love (61% - 118569298 votes)
2. sex (39% - 77186309 votes)
Ah, love (or the search for it) triumphs over sex. For the heck of it, I reversed the terms, i.e., put "sex" in the first field and "love" in the second. The results:
1. sex (61% - 118569298 votes)
2. love (39% - 77186309 votes)
Not good. Lycos UK just switched the terms, leaving the stats the same. I tried this in Opera with JavaScript off, and even tried closing and re-opening the browser after each search to prevent any caching issues. It didn't matter which of the two terms was in the first search field: it would consistently "win" the contest with 118569298 votes.
Mind you, not all terms get these same stats. For example, "red" vs. "blue" (and "blue" vs. "red") yields:
1. red (56% - 92120936 votes)
2. blue (44% - 72489692 votes)
So there is something going on under the hood...but for all we know its a made-up formula delivering results based on the ASCII values of the letters.
Does anybody have any additional info re the Lycos UK Trendometer?
