@linuxdsp: Your state of linux audio thread was destined to do nothing but collect a bunch of anecdotes. The linux world is full of this stuff spread over the entirety of the previous decade, on web forum after web forum and blog after blog after blog. Of course, none of the anecdotes are false (well, one assumes not, anyway), but they just don’t add up to anything. People don’t spend much time online saying “hey, everything just worked!”. People don’t even respond to threads about “the state of linux audio” with “hey, everything just worked” (even though sure, a few people did just that). Posting on a web forum with a serious intent to collect data rather than just a haphazard set of random anecdotes strikes me as (a) pointless (b) guaranteed to leave behind yet another breadcrumb to be discovered via a search engine that points to a single point in time, something which may be clear to you, but won’t be clear to 80% of the people who read it later.
You’ve developed some really amazing plugins that work on Linux, but you are, by your own admission, not part of the open source world. Like it or not, Ardour and most of (though at this point, far from all) of the pro-audio/music creation software that exists for Linux is developed by people for whom the open source/libre characteristic is at least as important as the fact that it runs on Linux. Ardour as a project and myself as an individual have worked quite hard to accomodate and work with people/standards that prefer proprietary software, but our attachment to open source/libre software is as much a statement about how we think about the future and the evolution of software as it is about any claims regarding Linux or other open source/libre software right now.
@t0bY: what did you find childish and sickening? Linuxdsp prioritizes user experience right now over long term ideals and goals regarding the way software is created, distributed and evolves. That’s a defensible position, and one shared by almost all proprietary software developers I’ve ever worked with. But the open source/libre world tends to have pretty much the opposite priorities: we care more about the infrastructure and ethos of software development rather than today’s user experience. This is in part because we think that getting stuff “right” (whatever that means) at lower levels and in terms of the licensing and development model ultimately leads to better software and reduces the risks of abandonware significantly. That belief could be wrong, certainly. But it is a motivating factor for many FOSS developers, and is in substantial contrast to developers who think that giving users what they think they want right now should be highest priority, regardless of what effects this might have over the long term.
we care more about the infrastructure and ethos of software development rather than today's user experience
IMO this is the complete wrong point to start. You can do this, if you are developing something for yourself and only yourself. If you write software for users, which should pay for your work, you should fit their needs. If its open source or not. At the end, the user has to use the software.
Maybe its something in between … or maybe I just didn’t get your point.
Issues like this are very hard to debug without a proper backtrace and to get a backtrace one needs code with symbols or the source. This was further complicated by the fact that it’s a one-off threading issue.
I very much respect your DSP skills and your plugins usually get a lot of things right.
Then again the question remains: Why should I spend my spare time to track down issues that some proprietary commercial software causes in Ardour? I’m mainly scratching my own itches. If something bothers you, the ardour source-code is there… just fix it.
As for professional attitude, try to contact NI about a problem (yes, some of their plugins crash regularly in various hosts). You’ll need a much thicker skin than with most free-software devs
@sonnie: there are users today, and there are users tomorrow. Open source development tends to pay more attention the things that may have long term impact on the user (and on other software developers). You might prefer that the focus was on features today … find a different group of open source developers or use proprietary software if that is really important to you. The FOSS world wants to avoid creating large collections of garbage code as part of some frenzied, market-driven rush towards features. This is extremely common in proprietary software development environments. It might work for today, but eventually it forces nightmares, code obsolescence (requiring total rewrites) and worse. The features are nice for users today; the mess is not very nice for the future.
Certain interpersonal actions and attitudes, which are badly hidden behind mock debates.
Using people as scapegoats for ones personal issues with proprietary software, is a weak move I find.
Yet another problem lies in the middle of your sentence. " [. . .] we think that getting stuff ‘right’ (whatever that means) [. . .] leads to better software [. . .] " .
Thus far it proves right that there is a 50:50 chance of either being taken seriously here, or being labeled a ranter against free software. - btw. this website advertises ardours compatibility features with proprietary soft and hardware… chough chough.
I’d actually prefer the argument that implementing stuff like this a time and resource issue.
I have been biting my tongue in the background, but I will interject one thing right now before this continues…
Before you respond to anyone, ask yourself this… “Is the ‘debate’ actually helping anyone or any issue at this point?”
I am not convinced the past several posts the answer to this would really be yes Not pointing at anyone in particular, just trying to head off pointless arguing when the conversation between the programmers here would be much better served over a beer than over a forum where visual context for the reply is lost.
Seablade
/me takes off moderator’s hat… that he really doesn’t have…
