[fdo] Re: Scheduling subsystems (crontab, at) and the desktop

[Patrick Patterson]

On Thursday 22 July 2004 10:59, C. Gatzemeier wrote:
> Am Thursday 22 July 2004 14:55 schrieb Maciej Katafiasz:
> > so stay tuned :). Right now there's preliminary document available at
> > http://zeus.polsl.gliwice.pl/~mathrick/spec.txt
>
> Hey, do I read this right? Are you rather planning a unifying framework
> than an actual crond replacement?
> Because that was what I also meant by pointing to CFG in my first respose.
> It seems there is quite a lot of overlap with the discussion on general
> configuration API/framework we had here a while ago. "A crontab is a file
> with a format" thus can be configured.
>
> A particular thing you seem to address that CFG does not yet have is a
> notion for is handling of notifications. Another thing is d-bus, which was
> not yet there when CFG was first planned, but would fit in there very good
> now. Maybe you would like to have a look at the architecture and see if
> your ideas could be applied.
>

Hi All:

I just joined the list a few days ago, during the DDC in Ottawa, so I don't 
have the whole historic on this, but it sounds like what you are talking 
about already exists - for those of you who attended the DDC, you should 
recognise it - UniConf - the Universal Configuration Framework... it handles 
replication, cacheing, notifications, as well as the standard "get/set" type 
configuration issues.

(It also talks and plays nicely with gconf, the Windows registry (did I 
mention it's cross platform :)), KConfig, etc.

http://open.nit.ca is the current home for it, so please take a long look at 
it before designing something that this might just solve :)

</end soapbox>


-- 
Patrick Patterson
Technical Ambassador
Net Integration Technologies R&D