Le jeu, 24/06/2004 à 16:33 +0930, Malcolm Caldwell a écrit : > But... this 'rebuilding' would only happen when I have said 'always try > video device blah' but no device was available on gm startup. > > Specifically it would not happen on calls when the video device WAS > available at startup. > It sounds hackish, really. > > Recent versions of ohphone also work that way. > > Yes and no. Yes they do the scan but they don't do it on program > startup - they do it on call setup. Which, I admit slows things down but > does make things a bit more plug and play. > > For example if I do > ophone -blah -blah --blahblahblah --videodevice "Philips", > Without the Philips camera plugged in, plug it in and then say 'c foo' > it will just work. > > What I suggest is: > 1. scan and popluate the dictionary on startup. If the camera is > found great > 2. At call setup: > * If we know our device from 1 name great! > * If we don't know our device name, do a scan to see it it > has become available. > > Now, there may be issues if someone swaps between different camera a lot > - but I would have thought that most users would not do this. They > probably have one camera which is sometimes plugged in, sometimes not. I don't know what others here think about this, but I'm not convinced. Notice they are working on HAL for Linux, that would solve that problem in a clean way as user programs will receive an event from a daemon when a device is plugged in, plugged off. -- _ Damien Sandras (o- GnomeMeeting: http://www.gnomemeeting.org/ //\ FOSDEM : http://www.fosdem.org v_/_ H.323 phone : callto:ils.seconix.com/dsandras seconix com
Attachment:
signature.asc
Description: Ceci est une partie de message =?ISO-8859-1?Q?num=E9riquement?= =?ISO-8859-1?Q?_sign=E9e?=