[plug] New PLUG news server

[Bernd Felsche]

James Devenish <devenish at guild.uwa.edu.au> writes:

>In article <g3lp22xi3a.ln2 at innovative.iinet.net.au>, Bernd Felsche wrote:
>> The method described above provides for downloading of most news
>> articles during absence. And a means of providing comfortable
>> access.

>Put it this way: If I am absent,
> - I expect to be able to put an autoresponder on a server and have the
>   autoresponder send out messages promptly as appropriate.

Nice for spammers.

> - I do *not* expect to have to configure a workstation to be powered on
>   to dial up periodically to the Internet in my absence, download my
>   mail, then send belated autoresponder messages. 

>Likewise, I would expect that the news service proceeds merily without
>me and that I can `pick up where I left off' at arbitrary times in the
>future.

It will; it can just take a lot of time to download all the backlog
of articles. That's a tradeoff. You don't have to set up your machine
that way. It's a convenient tool to reduce the delays when you
return.

You can of course simply `pick up where I left off'; it'll just take
longer to do so.

>> There are headaches/issues with injecting 5-year-old articles into a
>> newsgroup. The server will simply reject the articles if they exceed
>> the maximum permissible age for the newsgroups.

>That sounds like a particularly artificial barrier.

Practical, historically speaking. Back in the dark, dim days, when
it used to take 3 days for articles to reach .oz servers from .com,
.mil, .edu and .gov, some servers mis-behaved; or more precisely,
some end-nodes misbehaved and connected to multiple points causing a
slow circular propagation. Space was limited on most systems so
article expiry was usually "rapid" and the servers seldom kept track
of articles that they'd seen 2 weeks ago. The problem of circulating
articles persisted; certainly into the late 1990's.

leafnode2 has a configurable 'maxage' as a default and for newsgroup
patterns.

>> Also, for people who have not previously connected, the backlog of
>> many thousands of headers will cause significant initial connection
>> delays.

>What are you talking about? If I connect to news0.optus.net.au,
>with hundreds of gigabytes of mail in its spools and tens of
>thousands of messages in newsgroups and I wish to read a newsgroup
>containing tens of thousands of messages that I have never read
>before, there is no problem. My newsreader will simply find out how
>many messages are in the group and prompt me for how many I wish to
>read, or alternatively I can just choose to ignore all past
>messages -- it's up to me. 

But to find out what those messages are about, you have to at least
parse the headers.

Hwo do you knwo that the thread you really "need" to see isn't the
one that's one past the fence-post in the number you nominate?

>No one forces me to sit in front of a computer reading every single
>backlogged message if I don't want to. But I *can* if I want to.

I didn't state that you had to.
-- 
/"\ Bernd Felsche - Innovative Reckoning, Perth, Western Australia
\ /  ASCII ribbon campaign | I'm a .signature virus!
 X   against HTML mail     | Copy me into your ~/.signature
/ \  and postings          | to help me spread!