Re: [Pan-users] Groups w/huge header count won't load?

[David Kelly]

On Jul 26, 2004, at 8:43 PM, Jim Henderson wrote:

> Another option as well (one I didn't mention before because it didn't 
> really occur to me, even though I've spent the last few days setting 
> it up) is to set up your own news server (I use INN) and then use the 
> 'suck' program to pull the messages off of the server.  Then you can 
> more easilly control the selection of messages you wish to look at.  I 
> know some use Leafnode as well to do this, same idea, really.

A problem with that and large newsgroups is that some have so many 
messages, and so many huge binary attachments, that one could never 
catch up with a 2 million bit/sec broadband connection.

Catch-22 is that one needs to see the headers to select what to 
download. It is true that running one's own private news server that 
one can filter and discard headers on the fly. With pan one has to bite 
off a chunk then filter.

There are caching news servers which act as a proxy between here and 
there. These are useful when one has multiple simultaneous users (wife 
and kids) but only 2 simultaneous connections allowed to the news 
server. The caching server may answer from its cache or decide to fetch 
from the ISP server. Meanwhile rolling all of your local news 
connections into a single connection. Much like NAT does for IP 
addresses. All that is a bit much just to partially overcome Pan's 
management of large newsgroups. I "get by" by limiting Pan to the 
newest 500k headers.

MT-Newswatcher (Macintosh) will never bite off more than the newest 
100k headers from a newsgroup at a time. However unlike Pan MTNW will 
successfully bite off the _next newest_ 100k headers one after another 
until all headers in a group have been viewed or until the user has had 
enough and simply marks the entire group as read.

"Mark as read" in MTNW is essentially the same as "delete" in Pan.

--
David Kelly N4HHE, address@hidden
========================================================================
Whom computers would destroy, they must first drive mad.