Mike McDonald mikemac@titian.engr.sgi.com
Mon, 12 May 1997 18:11:03 -0700

>Subject: Re: POS
>To: lispos@math.gatech.edu
>Date: Mon, 12 May 1997 20:41:10 -0400 (EDT)
>From: BRIAN SPILSBURY <zhivago@iglou.com>
>MM> >BS> If you mean by this that if when I have 10 threads going, and I
>MM> >BS> shutdown and and restart then I still have those ten threads
>MM> >BS> running, then I'm looking for the same thing.
>MM> >
>MM> >That's what I want alrighty.
>MM>   And the first thing 9 out of those 10 threads are going to do is:
>MM>   (error "Connection timed out")
>Your point is?
>Yank out the network connection, and the same thing will happen.
>Have your router fall over... Have someone else's router fall over...
>Software that networks should be prepared for unreliable
>connections with sockets that die.
>I don't see that this is even vaguely relevent :)

  What's the point of going thru all of the trouble to make threads
persistent if the first thing they do after restarting is croak?
(Croak == Die/"Segment Fault" for you non US english speakers) Seems
like a lot of work just so that when I boot my machine I can get a
console full of error messages.

  Mike McDonald