Re: [DNS] thread.119

Re: [DNS] thread.119

From: Kim Davies <kim§cynosure.com.au>
Date: Thu, 17 Jan 2002 08:07:46 +0800
Quoting David Keegel on Thursday January 17, 2002:
| 
| Anybody else think they know a search engine where sites get a
| better ranking purely because they have a domain name which
| matches the search query (and not because they paid money)?
| Bonus points if its a search engine people actually use.

I don't think you are going to get a definitive answer to this. I would
think it reasonable that a search engine could use the URL as some
criteria in evaluating the page.

Of course, if everyone is being literal in saying that domain names
are not used by search engines, you'll be pleased to know if you go
to Google and type in "cars.com" the first hit is www.cars.com,
similarly for "home.com", "science.com", "television.com" etc.

I guess to really prove that search engines consider domain names in
queries, if I type in "cs.curtin.edu" into Altavista the top matches are
pages that only match in the URL(s). That is, there is no reference to
cs.curtin.edu in the body text. (Note this won't work in Google because
it doesn't support stemming. I chose this example because it is only
substring of the URL, not a complete valid domain name)

In general, the formulas search engines use are proprietary and
generally held close to their chests, so they aren't going to tell you
with any level of detail how their rankings are calculated. Indeed to do
so would only encourage others to learn the rules to up their ranking
which is arguably the last thing they want to happen.

The only way to absolutely settle this would be to create two identical
sites on a topic, one under a generic domain and the other under a more
obfuscated one. Register both with the search engine, and evaluate from
there. This is left as an exercise to whoever really thinks this point
is worth further investigation.

As for the the more general question of generics having no success
compared to other domains, as I said a month ago: There are a number
of generics that have succeeded, and given that generics are in such a
small minority, I think it is unreasonable to assert they are failures
with no value.

The fact is, many people think they are valuable. If anyone here wants
to convince them that they are not then I think you are probably wasting
your time. Let them find out for themselves.

kim
Received on Fri Oct 03 2003 - 00:00:00 UTC

This archive was generated by hypermail 2.3.0 : Sat Sep 09 2017 - 22:00:04 UTC