Well... life seems to be sorting itself out for the moment.

[wolfwings]

I'll be staying at manawolf and sardonicus's place for the immediate future, providing driver services for manawolf while her ankle heals.

For today... I'm going to go to the Fur-B-Q obviously. And so are manawolf and sardonicus and Janus. :-)

As for the immediate future... I'm going to start working on the 'next generation' fork of FuzzBall MUCK that has sat lingering on my disk-drive for ages.

I still want to remove everything from the server that is single-purpose or better handled in server-side MUF code. Like 'pennies', the 'kill' flag, the 'kill' and 'rob' commands, since the 'kill' flag doesn't actually have any meaning for anything but player-type objects in the database, and pennies are just a crude form of quota, better handled with MUF implementing type-specific quota's.

Comments

[kiala.livejournal.com]

make FB talk to a real DB backend, please? ^^

That, my friend, is a long-term goal of this 'NG' fork.

[wolfwings.livejournal.com]

Specifically, I'm removing dozens of functions first. Like compression support and 'diskbasing' of individual properties are already removed from the fork. I've started renaming files from things like (misc.c) to (md5passwords.c) which is what they are actually used for, etc.

Once enough 'feeps' are (re)moved from the C-level source code, and some put into MUF where they belong in the first place, THEN we'll be pushing the database I/O interface to a distinctly-seperate API, then simply writing a seperate module to read/write from a real database and then add support for real, honest-to-goodness 8-bit compression (Vitter Huffman to start, likely adding BWT as an option later). And yes, strings will be stored including a length at that point, instead of 0-terminated, so the text-handling will become fully 8-bit clean internally. :-)

Unfortuantely, this side-project has lingered for quite a while, so I'm going to update the fork from a fresh copy of the latest FB6 CVS before going too much further, and see about propegating some changes (renamed files to match function) upstream now, instead of later.