Hola Jordi, you are right, at this point there isn't any reason to keep alive alsa-0.5, a part for getting confused unexperienced user. Unfortunately, a part for the incredibly good job ALSA developers did -I'm a ALSA fan and user since the first release....-, they often choose strategies that tend to generate confusion and incompatibilities -if you are an ALSA aficionado you might have experienced as well what I'm talking about...:-( -..and unfortunately it doesn't seems to be a lesson they have learned, just few days ago there was a quasi revolution in the alsa-devel after a new unneeded change in the API created another backward incompatibility. As keeping alive 0.5 was an error at the first place. Indeed applications that still need 0.5 are either old applications, or applications that do not require an high level of efficiency, so they can easily run under 0.9 with OSS support, but it must be a very limited number, to be honest at the moment i can't recall any. Taking out 0.5 is a good and even better, maybe even before that, it would be important to get rid of alsaconf. Alsaconf is even more misleading than any other alsa package. It configures an ALSA file for the modutils that has such an old series of incompatible parameters that is amazing the problem that can generate, especially -and here is not completely ALSA devel fault- when people think, "ok i install ALSA 0.9, i use alsaconf and then i fix it...". Sure people should read instructions...but...writing an ALSA configuration file, even if relatively easy it is anyway something the require a bit of experience so one might be tempted of trying such bad hack. With the DeMuDi project we are trying to: - make discover to work with ALSA rather then oss. (this could go into end of November 0.9 DeMuDi release) - trying to make reconfiguration after installation easier. (this might be for next release) Also you might be interested in the work we are doing, in fact we are testing kernel-images, tuned with lowlatency and preempt patches, along with alsa-modules packaged (in two different standards, with and without acpi patches applied) The test we ran are positive for now. In the long tradition of trying to not replicate already done jobs it would be nice if we join forces on this task. It goes without saying that more maintainers and developers interested in the creation of a professional audio distribution would contribute to DeMuDi, we will achieve a better result and an easier integration into Debian as subproject. Thanks, ciao, marco Il 22/09/2002 alle 21:16:55, +0200, Jordi Mallach ha scritto: > Hi, > > I'm currently trying to fix ALSA packages a bit and the second NMU of > the saga just got installed in the archive. > > I'm just taking care of alsa-utils, alsa-lib and alsa-driver for now, > but I *may* do alsa-modules at some point. > > There's still alsa-*-0.5 packages around. The question is, do we still > want them? They are unsupported versions by alsa-project now, and they > are just collecting bugs. I think 0.9 works more or less ok at this > point, and 0.5 could be just dropped from Debian. > What to do with packages that depend on libasound1, too? Does anyone > know if alsa-lib 0.5 is source compatible with 0.9? > > Thanks, > Jordi > -- > Jordi Mallach Pérez -- Debian developer http://www.debian.org/ > jordi@sindominio.net jordi@debian.org http://www.sindominio.net/ > GnuPG public key information available at http://oskuro.net/~jordi/ -- ************************************************************************ * marco trevisani * * http://trevisani.mine.nu marco@centrotemporeale.it * * http://www.agnula.org -- A GNU/Linux Audio Distribution * * Neither MS-Word nor MS-PowerPoint attachments please: * * See http://www.fsf.org/philosophy/no-word-attachments.html * * Gpg Fingerprint = 6096 84B8 046C A5C9 B538 255E 9FFF 1121 3AFB FFA6 * ************************************************************************
Attachment:
pgpdrBq1zFY7E.pgp
Description: PGP signature