[MUD-Dev] Persistance/stability

[Matt Chatterley]

On Thu, 31 Jul 1997, Miroslav Silovic wrote:

> > have to explicitly reset it). When I first joined the first version of
> > this list, there was a lot of discussion (like now) with Chris L about
> > databases, commits, and lock/lockless stuff, and that just sort of
> > re-inforced my notion that other MUDs were persistant languages too.
> > It's clear that Diku's aren't, but couldn't we consider MOO's to be?
> > How about Mushes? I believe that Mushes don't save over a reboot (?),
> 
> MUSHes and MOOs save everything across the reboot. MOOs even save the
> active processes (something Cold doesn't do, and I don't know about
> MUSH).

MUSH servers (Penn/Tiny) do not save what might be called 'active
processes' between full restarts - but mind you, the beta 1.7.0 version of
Penn can reboot without actually kicking players off and killing itself
completely.
 
> > but get the impression that MOO's do. LP's don't, unless explicit
> > saving to files is done by LPC code. Am I vaguely correct here? What
> 
> I was rather apalled when I first discovered this. Mind you, it's much
> easier to implement than total persistance, and sometimes necessary
> (if you're doing a server in, say, LISP, you have a nonpersistant
> systems with pretty damn good compilers out there that are unable to
> save the methods unless you keep track of the source).

Yup. Its not hard at all to get LPC to save most of the stuff you're
handling, and just a little bit hard to do the rest (although the design
requires a fair ton of thought). Think about why it doesnt - given the
original intentions of both bases, MOO had to save everything, and LP had
to save nothing. There was always the potential for a high level of
persistancy with LPC, but noone really took up the offer. :)
 
> > Adam also mentioned that most MUDs don't stay up for more than a week
> > or two. That surprises me, too. I would have thought that LambdaMOO
> > would have been up for at least months at a time. Given that most don't
> > save everything over reboots, that sort of corrupts the impression of
> > a real "world", doesn't it?
> 
> In fact, even if a MUD is up for only a week at the time, 'everything
> gets reset in 5 minutes' breaks a spell for me.

Several muds reboot daily - but usually at low player times (and with them
it tends to be necessary 'maintenance' which is unavoidable). That they
don't save over reboots does spoil the image, if you're trying to explore
the world. If you just want to run around and hack things (as was the
target on the games I refer to), it was cool, because everything
respawned, although you had to go and re-equip yourself. :P Noticeable
resets of anything leave big scars in the environment.
 
> Unfortunately the only mainstream programming langauge with full persistance
> of both objects and methods /and/ a good compiler is Rscheme (one I could
> find, at least. I'd like to hear if there are others). Unfortunately
> the bloody thing keeps the complete transaction log and if there is 
> a db compactification function, it certainly isn't documented. :(

Ick. Well, I've not actually heard of one to date.

Regards,
	-Matt Chatterley
	http://user.itl.net/~neddy/index.html
"Doublethink means the power of holding two contradictory beliefs in one's
	mind simultaneously, and accepting both of them." -George Orwell