I use Swish-E with the HTTP method because I am indexing Web sites all over
the place, *not* files on my local machine or network.
I don't quite understand why the HTTP method is regarding as a last resort.
If one is spidering an unknown Web site, surely it is the only resort? So
far, I am very pleased with the indexes I have built using this method
(although Swish-E does fall over on some files for no obvious reason).
I don't find the indexing too slow - I know how slow the Internet can be.
From: Roy Tennant [SMTP:email@example.com]
Sent: Thursday, February 17, 2000 6:15 PM
To: Multiple recipients of list
Subject: [SWISH-E] http method v. file system method
I'm rather taken aback at all the people who appear to be using the http
method to index with SWISH-E when it seems, from what I can gather, that
they should not. The http method should *only* be used when the file
system method simply cannot (that is, when the system being indexed is
remote). To expect the http method to be as fast as, and work as well, as
the file system method is unrealistic. To recap, the http method is the
one to use *when all else fails*, not as the first resort.
Received on Thu Feb 17 13:44:16 2000