MUD-Dev
mailing list archive
[ Other Periods
| Other mailing lists
| Search
]
Date:
[ Previous
| Next
]
Thread:
[ Previous
| Next
]
Index:
[ Author
| Date
| Thread
]
[MUD-Dev] Re: about MOO
Quoting Dan Root (dar#thekeep,org):
> >There was once a disk based version of LambdaMOO, IIRC, though it may have
> >never really gotten off the ground.
>
> Lots of talk, but I never saw any code, nor anyone who claimed to have it,
> other than, as you mention, LPMOO, which is amusing, but not really the
> right way to run a MOO. :)
I hacked one up in a couple of afternoons a couple of years ago, then
promptly threw the code away again because the MOO I was working on at
the time didn't really need it -- it was just a bit of fun. I never
bothered coming up with a good on-disk representation, so it was quite
slow. (As I recall I just used the normal db object write/read code,
which is text-based.)
So I claim to have had *something*. Of course, I don't now.
The internal db abstraction in 1.8.0ish LambdaMOO is just good enough
to make it a lot easier than you might expect, particularly as most
aquisition of an (internal) object pointer goes through a single
function. Scatter a few unfind()s to match the find()s and you're,
well, at least some of the way there (although not heading for
stunning efficiency.)
--k
_______________________________________________
MUD-Dev maillist - MUD-Dev#kanga,nu
http://www.kanga.nu/lists/listinfo/mud-dev
- Thread context:
- Re: [MUD-Dev] about MOO, (continued)
[ Other Periods
| Other mailing lists
| Search
]