[arch-proaudio] Dropping dead project: lash
Hi all,
I've just dropped lash to the AUR. It has essentially been dead for nine years, which is why I don't want to spend any time fixing it, or including it into any softare in the main repositories.
AFAIK, all projects depending on it, can be built without it and non-session-manager is a more modern approach (as the application only needs to implement an OSC interface) anyways. If this is not the case and your AUR package depends on lash, you will have to fix lash and its issues with current glibc (removed Sun RPC), as it introduces a new dependency on libtirpc.
Best, David
parabola has the following 3 dependents of 'lash':
azr3-jack fluidsynth zynaddsubfx
'fluidsynth' and 'zynaddsubfx' are coming from arch [community] - have these have been modified to no longer depend on 'lash' ?
is there any guide or advice for rescuing these from the trash heap of deprecation?
On June 15, 2018 9:44:25 AM GMT+02:00, bill-auger bill-auger@peers.community wrote:
parabola has the following 3 dependents of 'lash':
azr3-jack fluidsynth zynaddsubfx
'fluidsynth' and 'zynaddsubfx' are coming from arch [community] - have these have been modified to no longer depend on 'lash' ?
is there any guide or advice for rescuing these from the trash heap of deprecation?
Hi,
yes, fluidsynth and zynaddsubfx have been rebuilt without lash support (hydrogen was yet another).
azr3-jack I don't know about. If you need lash, you'll need to fix it in the AUR. I got tired of doing this for long-undead stuff. I really hope I can deprecate a2jmidid in the not too distant future as well, as it's yet another one...
Best, David
Hey, I'm still using a2jmidid! :) (Last time I checked, Jack2 didn't have that built into it yet.) But I don't really mind having to install it from the AUR.
Albert
On Fri, Jun 15, 2018 at 10:06 AM, David Runge dave@sleepmap.de wrote:
On June 15, 2018 9:44:25 AM GMT+02:00, bill-auger bill-auger@peers.community wrote:
parabola has the following 3 dependents of 'lash':
azr3-jack fluidsynth zynaddsubfx
'fluidsynth' and 'zynaddsubfx' are coming from arch [community] - have these have been modified to no longer depend on 'lash' ?
is there any guide or advice for rescuing these from the trash heap of deprecation?
Hi,
yes, fluidsynth and zynaddsubfx have been rebuilt without lash support (hydrogen was yet another).
azr3-jack I don't know about. If you need lash, you'll need to fix it in the AUR. I got tired of doing this for long-undead stuff. I really hope I can deprecate a2jmidid in the not too distant future as well, as it's yet another one...
Best, David -- https://sleepmap.de
Yeah, what Albert said. a2jmidid is actually still pretty vital as most MIDI programs still *only* support either ALSA MIDI or Jack MIDI, and I use some of both. But if all I'd have to do was install it from the AUR instead of the repo (rather than deal with some sort of dependency nightmare), that's fine.
When you say those other packages have been rebuilt without lash support, do you mean that by the time lash is gone, versions of them that don't depend on lash will have left testing and will be ready for me to update to? If so, great!
On Fri, Jun 15, 2018 at 1:41 AM Albert Graef aggraef@gmail.com wrote:
Hey, I'm still using a2jmidid! :) (Last time I checked, Jack2 didn't have that built into it yet.) But I don't really mind having to install it from the AUR.
Albert
On Fri, Jun 15, 2018 at 10:06 AM, David Runge dave@sleepmap.de wrote:
On June 15, 2018 9:44:25 AM GMT+02:00, bill-auger bill-auger@peers.community wrote:
parabola has the following 3 dependents of 'lash':
azr3-jack fluidsynth zynaddsubfx
'fluidsynth' and 'zynaddsubfx' are coming from arch [community] - have these have been modified to no longer depend on 'lash' ?
is there any guide or advice for rescuing these from the trash heap of deprecation?
Hi,
yes, fluidsynth and zynaddsubfx have been rebuilt without lash support (hydrogen was yet another).
azr3-jack I don't know about. If you need lash, you'll need to fix it in the AUR. I got tired of doing this for long-undead stuff. I really hope I can deprecate a2jmidid in the not too distant future as well, as it's yet another one...
Best, David -- https://sleepmap.de
-- Dr. Albert Gr"af Computer Music Research Group, JGU Mainz, Germany Email: aggraef@gmail.com WWW: https://plus.google.com/+AlbertGraef
On 2018-06-15 08:31:41 (-0700), Jimi Bove wrote:
Yeah, what Albert said. a2jmidid is actually still pretty vital as most MIDI programs still *only* support either ALSA MIDI or Jack MIDI, and I use some of both. But if all I'd have to do was install it from the AUR instead of the repo (rather than deal with some sort of dependency nightmare), that's fine.
Please don't get me wrong here: I'm using a2jmidid myself from time to time. I wrote "in the not too distant future". I mainly have falktx' statement regarding jack1/jack2 feature parity in mind here [1]. Those are my hopes for the future of jack! Given Filipe's new employer, I'm not so sure he will have a lot of time to invest in jack2 any time soon though. So, if you're able to help, get in touch with him!
Lash, the same as a2jmidid was developed by Nedko Arnaudov, who didn't respond to my requests at all and seemingly abandoned development. The a2jmidid upstream is basically just around, because someone most likely forgot to delete it (or the server just luckily has never been shut down). It's a very unfortunate situation for a piece of software, that everyone needs, when using jack2 and MIDI! I'm just as unhappy about this as anyone else I guess.
Probably thirdparty tools such as this should be picked up by the jackaudio association on github, so they'll at least be around for a little longer.
When you say those other packages have been rebuilt without lash support, do you mean that by the time lash is gone, versions of them that don't depend on lash will have left testing and will be ready for me to update to? If so, great!
hydrogen, zynaddsubfx and fluidsynth were the only packages in Arch still "using" lash. I rebuilt them without their lash feature. Those are the current versions in [community]. Lash is in the AUR now and needs fixing to even be able be build.
[1] http://kxstudio.linuxaudio.org/News/?action=view&url=jack2-1912-release-...
To reduce midi jitter a2jmidid is required in combination with jack2's -Xalsarawmidi switch ;). It's said that jack1 nowadays provides a feature to reduce MIDI jitter, so maybe jack2's -Xalsarawmidi switch and a2jmidid aren't required for this purpose anymore. However, a while back only jack2 was able to reduce MIDI jitter, let alone that on my old computer jack1 also was unusable for audio.
In short, if somebody is interested in making music, without migrating from one tool that get dropped and replaced in favour of another unfinished alpha tool, the best bet is to use Linux, just for selected parts of an audio production.
On Fri, Jun 15, 2018 at 6:32 PM, David Runge dave@sleepmap.de wrote:
Please don't get me wrong here: I'm using a2jmidid myself from time to time. I wrote "in the not too distant future". I mainly have falktx' statement regarding jack1/jack2 feature parity in mind here [1].
That's a relief. :) I remember now that this was discussed on lad a while back. It's a real pity that Nedko seems to have disappeared from the scene and that Filipe is busy with other things now. I'm afraid that I don't have the time to give a helping hand myself, as I'm involved in quite a few projects and maintain way too many packages already. :(
FWIW, there's quite a few mirrors of Nedko's a2jmidid source on GitHub already, such as https://github.com/simonvanderveldt/a2jmidid, so at least we can be reasonably sure that it doesn't just suddenly disappear from the web altogether. Whether Simon is willing to keep maintaining it is another matter, but if not we can just fork it again if the need arises.
Albert
P.S.: David, I'm sorry that I missed your presentation about pro-audio on Arch, I was too busy in the hallway (or rather Spree-side) track around the time. I see that that the video of your talk is up already, but will you also upload your slides on http://lac.linuxaudio.org/2018 some time?
the softwareheritage project exists to preserve free software source code - the API return many replicas of a2jmidid
https://archive.softwareheritage.org/api/1/origin/search/a2jmidid
On 2018-06-15 23:27:37 (-0400), bill-auger wrote:
the softwareheritage project exists to preserve free software source code - the API return many replicas of a2jmidid
https://archive.softwareheritage.org/api/1/origin/search/a2jmidid
Thanks! That's a 'good to know'! :)
participants (5)
-
Albert Graef
-
bill-auger
-
David Runge
-
Jimi Bove
-
Ralf Mardorf