[1436] in Coldmud discussion meeting

root meeting help first first in chain previous in chain previous next next in chain last in chain last

Re: ColdCore Roadmap

daemon@ATHENA.MIT.EDU (Tue Nov 10 23:08:51 1998 )

From: cshepher@brick.net (Chuck Shepherd)
To: coldstuff@cold.org
Date: Tue, 10 Nov 1998 23:15:01 -0600 (CST)

Some of my experiences with development with ColdCore so far (I use
ColdCore 3.0a9.02 as the base, with some fixes from newer versions).

With the old version of the Core, a non-writable Core didn't allow
me to make the changes I needed.  For example, one of the things I
did was change the behavior of @finger to include roleplay oriented
settings.  The only way to do this was to set core writeable.

I then made other changes-- to the channel code,
plain/text method, and the @send command, to give examples-- as
well as added local user-objects where I put commands (i.e.,
$local_user and $staff).  This is how people (mostly) make changes
with LambdaCore on MOO-- they start with a base Core, and modify
it from that point.

Another approach (that ColdCore uses) is that the Core should
be left alone, and not modified.  This makes upgrades easier,
but also tends to point us in the direction of a "stock"-type
Core.  Most Cold developers I know like tinkering,
so few of us probably have an original-style core.

One idea might be to have Core "packages".  ie, one base package
with the networking code, one for each daemon,  one for the parser,
etc.  These could even have dependency requirements.  I don't see
any reason for Dreamwheels, for example, to have separate networking
code than the main ColdCore.  Right now I can do some of this by using
@dump on the objects, and transferring them over by hand.  This is
time consuming, though, and doesn't always work.

To give an example of how this could be useful:  One of the additions
I want to make is to write puppets.  This is dependent upon the parsing
code, which was rewritten after 3.0a9.  It would be nice to upgrade to
the new command parser, so I could use what someone else wrote (okay,
I can dream :) or can make it easier for others to use any puppet
code developed on Dreamwheels.  The command cache upgrade is
something I've put off for some time already... it doesn't sound
simple.  And I have enough modified code to make it easy
to port everything over to a newer version of ColdCore.

I would be interested in hearing from what problems others face--
especially if we can increase the code sharing in the Cold community.
It's hard to know what issues others in my  situation face (writing a
roleplay oriented, mush/moo-type world).

-- Chuck

Chuck Shepherd cshepher@brick.net