Robert L Mathews wrote:

(I generally don't put privately run NNTP servers in the "reliable" category; Tucows may be different, but I doubt it).


Agreed -- All servers have outages of some sort. When my mail server is down, I'm busy fixing it, so I don't mind that I can't read my OpenSRS discussion list -- I've never tried to read discuss list when my server is having problems.

More importantly, it's my problem, I fix it.

If the list server goes down, no problem, it just means I won't see new posts until the list comes up, I can still access my local archives, read and reply to messages, etc. This can't happen with a pull technology unless I'm constantly pulling.

I not only need to constantly poll (anybody think OpenSRS would mind if I poll the RSS feeds once a second? -- Email is pushed to my server, my server pushes a notice to me, so polling every second would actually result in a delay from when messages are posted until when I am notified about them), but if the pull server is down, I know about it and I can complain about it.

If the list server goes down while I'm composing a message, I won't lose it either, my mail server will resend it for me. This doesn't happen if a web forum goes down -- More often then not the message will be lost, unless I happen to notice, go back, save the message myself and try posting it later. That's a lot of effort to help out a competitor using my supplier's service.

Slightly off topic, anybody else read any of the various "RSS is crushing my server" articles recently? There is a reason why infrequent or unpredictable updates should be pushed rather then pulled.

--
They call it "PMS" because "Mad Cow Disease" was already taken

Reply via email to