Re: New virtual packages: lv2-plugin and lv2-host

2012-11-21 Thread Ian Jackson
Alessio Treglia writes (Re: New virtual packages: lv2-plugin and lv2-host): On Tue, Nov 20, 2012 at 2:27 PM, Ian Jackson ijack...@chiark.greenend.org.uk wrote: And an lv2-host could include something which can use only plugins with certain features. Altough dillo doesn't support

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-21 Thread Jonas Smedegaard
Quoting Ian Jackson (2012-11-21 13:40:35) Alessio Treglia writes (Re: New virtual packages: lv2-plugin and lv2-host): On Tue, Nov 20, 2012 at 2:27 PM, Ian Jackson ijack...@chiark.greenend.org.uk wrote: And an lv2-host could include something which can use only plugins with certain

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-21 Thread Ian Jackson
Jonas Smedegaard writes (Re: New virtual packages: lv2-plugin and lv2-host): It seems to me that you are applying more strict rules now than has been applied to other names on the list in the past. This is not some kind of hazing ritual where people have to persuade a reluctant audience. I'm

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-21 Thread Alessio Treglia
On Wed, Nov 21, 2012 at 12:40 PM, Ian Jackson ijack...@chiark.greenend.org.uk wrote: I guess you are thinking that LV2 plugin packages would Recommend or Depend on lv2-host ? Yes, exactly. But I don't think that's really helpful. Perhaps lv2-host should specify something more definite, like

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-21 Thread Ian Jackson
Why tightening up rules? Policy =C2=A73.6 does not pretend packages to meet= any specs nor comply with common interfaces, it just says Sometimes there are severa= l packages which offer more-or-less the same functionality. In this case, it'= s useful to define a virtual package whose name

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-21 Thread Daniel Hartwig
On 21 November 2012 22:51, Alessio Treglia ales...@debian.org wrote: Actually I receive lots of mails from users asking me questions like How could I find an exhaustive list of LV2 toys currently provided by Debian?, Does the X sequencer support LV2 plugins?. So, I think we'd do a good

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-21 Thread Alessio Treglia
On Wed, Nov 21, 2012 at 3:03 PM, Ian Jackson ijack...@chiark.greenend.org.uk wrote: So I'm afraid I still don't understand how this virtual package would help improve the dependency resolution. Although I'd agree that defining a new lv2-plugin would not be needed, making LV2 plugins packages

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-21 Thread Ian Jackson
Alessio Treglia writes (Re: New virtual packages: lv2-plugin and lv2-host): On Wed, Nov 21, 2012 at 3:03 PM, Ian Jackson ijack...@chiark.greenend.org.uk wrote: So I'm afraid I still don't understand how this virtual package would help improve the dependency resolution. Although I'd agree

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-21 Thread Russ Allbery
Jonas Smedegaard d...@jones.dk writes: It seems to me that you are applying more strict rules now than has been applied to other names on the list in the past. I don't know if Ian is, but I certainly would. We have a bunch of existing virtual packages that aren't really useful because they

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-21 Thread Russ Allbery
Alessio Treglia ales...@debian.org writes: Why tightening up rules? Policy §3.6 does not pretend packages to meet any specs nor comply with common interfaces, it just says Sometimes there are several packages which offer more-or-less the same functionality. In this case, it's useful to define

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-21 Thread Simon McVittie
On 21/11/12 17:27, Russ Allbery wrote: Without that, it's questionable whether the virtual package serves any purpose, and indeed you'll find that the CD ripping packages in Debian don't reference mp3-encoder ... and perhaps more tellingly, only one package Provides it, and that package isn't

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-21 Thread Jon Dowland
On 21 Nov 2012, at 17:27, Russ Allbery r...@debian.org wrote: don't know if Ian is, but I certainly would. We have a bunch of existing virtual packages that aren't really useful because they don't offer any sort of guaranteed interface, and therefore cannot be meaningfully used in package

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-21 Thread Vincent Danjean
Le 21/11/2012 17:48, Ian Jackson a écrit : Although I'd agree that defining a new lv2-plugin would not be needed, making LV2 plugins packages Depend/Recommend a generic lv2-host package would seriously help as it allows maintainers to avoid to fill up Depends: fields with long and incomplete

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-20 Thread Ian Jackson
Alessio Treglia writes (New virtual packages: lv2-plugin and lv2-host): virtual-package-list.txt.gz [1] says: 1. Post to debian-devel saying what names you intend to use or what other changes you wish to make, and file a wish list bug against the package debian-policy. So,

Re: New virtual packages: lv2-plugin and lv2-host

2012-11-20 Thread Alessio Treglia
Hi Ian, thanks for the quick reply! On Tue, Nov 20, 2012 at 2:27 PM, Ian Jackson ijack...@chiark.greenend.org.uk wrote: An lv2-plugin could be almost anything AFAICT, so depending on some LV2 plugin is not very useful. And an lv2-host could include something which can use only plugins with