configuration system notes

[Avery Pennarun]

On Thu, Apr 07, 2005 at 04:42:19PM +0100, Dave Cridland wrote:

> Now, I'm absolutely not suggesting that ACAP be "the" configuration 
> system[*] - for the vast majority of desktop applications - it'd be a 
> terrible idea to have a TCP/IP protocol in use for local 
> configuration

Hmm, anyone who proposes *not* requiring a daemon seems to get shot down
rapidly on this mailing list.  And I think all those daemons are
communicating over TCP/IP or (at minimum) Unix sockets.

> * - Thus making me almost the only person present to have both 
> written a configuration system and not suggested that it should 
> immediately be adopted as a full standard for everyone.

And I'm sure everyone appreciates it :)

That said, despite appearances, I'm not actually trying to push UniConf
here.  I'm just trying to discourage people from wasting their effort
writing yet another new version of something UniConf already does.  No
matter what system wins, UniConf will support it, so I'm not too concerned.

I think elektra, now that it has pluggable backends, is actually the current
best choice for "the" configuration system.  Perhaps if someone implemented
Havoc's "new gconf" spec, it would be nice... except that they haven't.

Have fun,

Avery