At 04:36 PM 2002-08-16 -0700, Russ Allbery wrote:
>I don't believe that it's possible to change this until someone develops
>archives that's going to be acceptable so long as every time you click on
>anything you have to do a round-trip page retrieval from the server.
>The ideal web archive looks and functions exactly like the user's mail
>client, but with enhanced search features. This is exactly why Gmane is
>so cool, because that's close to what it is.
Now, let's think about this. It is impossible to do a good archive on the
web because of round trip time, but the fact that gmane is at the other end
of a Round Trip is not a problem?
I don' get it, Lucy. Something is at work here other than round trip time.
>(Which indicates to me that the real future of archiving of mailing lists
>is in read-only IMAP folders.)
Which are also on the other end of a round trip interaction with a server.
I'm willing to agree that it will be difficult or impossible to do a
reasonable web archive, but I think that it is because the web model sort
of sucks because it does not have a rich enough set of search
functions...then again, why is that not part of the world of
CGI? Essentially, the assertion is that an imap client or a nntp client
can search a bunch of files and return the right ones in a short enough
time (including the milliseconds for the server RTT) and that is
acceptable, but you can't plug that same code into a web server and have it
return the files? I really do not think that server RTT has anything to
do with it.
"Forgive him, for he believes that the customs of his tribe are the laws of
-- George Bernard Shaw (1856-1950)
Nick Simicich - email@example.com