Skip to main content.
 home | support | download Search for

# Re: An idea for calculating ranking

From: Peter Karman <peter(at)not-real.peknet.com>
Date: Fri Jun 10 2005 - 14:01:48 GMT
Bill Moseley scribbled on 6/10/05 8:26 AM:

> On Thu, Jun 09, 2005 at 03:05:59PM -0700, koszalekopalek wrote:
>
>>1) Adobe Reader - Download
>>2) Macromedia Flash Player Download Center
>>3) Rhapsody & RealPlayer — the best audio and video players — Real.com
>>4) Apple - QuickTime - Download
>>5) MapQuest.Com Maps, Directions and More
>>6) Netscape Browser
>>
>>This only proves that they do something similar to what I
>>proposed. Adobe ranks first because there is a gazillion of
>>links out there that say click >>here<< to download Adobe.
>
>
> You mean google's famous "Page Rank"?  They are not adjusting the
> results by the text of the <a> tag, but by the number of links.
>

Page Rank is not so simple as counting links. Click on the Adobe Reader link and
search with your browser for the word 'here'. I get no matches.

That suggests to me that the text that is used to link \to\ a site makes a
difference. Not \all\ the difference, but some.

Try the search with common words like 'to' and 'for'. Adobe Reader keeps popping
up near the top of the results, as do the rest. But the number one hit for 'to'
is 'Welcome to the White House' ...

If Page Rank were simple, google wouldn't be so hot. If it were simple, swish-e
could be "just like google." But it's not.

Bill's point is a good one: Swish-e is a primarily an \indexer\ not a search
engine. The search is very straightforward: unlike google, you get exactly what
you ask for. If you search for 'here' you get docs with that word in them.

--
Peter Karman  .  http://peknet.com/  .  peter(at)not-real.peknet.com

Received on Fri Jun 10 07:01:48 2005