Mon 26 Jan 2009
Ah the recent spate of technology bashing reminds me of when D-Bus, Hal and udev were young and inexperienced. Many people yearned for the days of automounter, cried bloody murder when Wine couldn’t find their cdrom and their favourite command line tool just couldn’t grok the new device syntax.
What’s that? The technology matured and now people would rise up in arms if we ever reverted to the old days? To be fair we do try not to break things but some things are broken already and are being barely held together by dead kittens and fairy dust.
As I said when we completely broke D-Bus to move to the new wire protocol (0.2x-0.3x) – sure there are growing pains involved but in time people will just get on with their lives and it will become another layer of the system that nobody cares about because it just works. It pays to take the time to fix things right even if there will be pain in between.
[read this post in: ar de es fr it ja ko pt ru zh-CN ]
January 27th, 2009 at 3:48 am
Regarding my opinion on pulseaudio.
What I feel wrong is to drop a working solution for one which despite a lot of work still makes so many people have no wound, or bad quality sound, or a machine too slow too read dvd, or…
Enforcing people to use it when it works for 99% of people might be good as the 1% may be fixed then. But currently I feel that much more than 1% people get issues with pulseaudio, and there are not enough people to fix them… (And some can’t be fixed, it will always add overhead so not powerful machines, like netbooks, will not be able to do things they can do when it is disabled).
I was in favor of pulseaudio being enabled by default on distros, because there was an easy way to disable it for people who had troubles with it and a few people got new functionalities.
Despite all the mails by lennart and others, I still see no real benefit in using pulseaudio for me (I have a laptop with 2 internal speakers, most of the time my sound is mute except when I have only one source of sound: when I listen some music or watch a movie, my brain can’t handle both at the same time), and have cases when it’s better to disable it.
For a long time I have been hearing that the issues are not in pulseaudio but in alsa or the apps. I believe it, but actually I don’t care. alsa has a lot of bugs but works fine, showing alsa bugs that will need years to get fixed, meaning that people won’t have sound during the next few years doesn’t make sense to me.
The desktop is supposed to be used, I don’t see the point of telling people “we broke it and you should not use our desktop during the next few years until we stabilize this if you want a decent experience”
Mails like http://mail.gnome.org/archives/desktop-devel-list/2009-January/msg00205.html does not help too.
« Before we can properly fix this we need some more groundwork in ALSA done. Due to lack of manpower I
unfortunately don’t see this coming any time soon. »
« Also note, that you can always drop to “alsamixer -c0″ to fiddle with all the exotic mixer controls directly. It is very unfortunate that this might be necessary for now, »
Yes this is very unfortunate that you have to tell people to use alsamixer and that won’t change anytime soon.
January 27th, 2009 at 4:10 am
Is it really fair to ask your userbase to be constantly testing alpha-level software in the name of progress? Where and when does it end?
Don’t get me wrong, I totally appreciate the scope of the problems involved and the difficulty in getting these things debugged without actually putting software in the field. But surely there must have been some things that could have been done in hindsight to make the transition to pulseaudio smoother. What can be taken away from this experience for the next time a subsystem in the linux desktop needs to be reworked?
January 27th, 2009 at 6:04 am
But this is really majorly critical stuff here. Pulse broke their proprietary flash crap (it works fine for me, though) ! So no web games for them
January 27th, 2009 at 7:04 am
I don’t think that this is just about bugs or whether Pulse hit distros too early. The question is it’s gremlin-friendly design. The problem is no-one seems clear that there isn’t a nicer solution to these apparently imortant technical sound issues.
January 27th, 2009 at 8:52 am
Hey J5,
DBus, udev, and Hal are all great technologies now and hopefully in the future we’ll be able to count PulseAudio among them.
For now, though, PA is causing a lot of frustration for a lot of users. I was getting emails and comments on my blog every week from people who stumbled upon my PA rant that I posted 8 months ago saying that they too were having problems with hangs, underruns, etc. In fact, I’m /still/ getting these emails and comments on my blog from users using nearly every distro from Fedora to Ubuntu.
I never blamed Lennart, I blamed the distros for shipping it when it wasn’t ready (I believe that even Lennart himself said it wasn’t ready on one of the gnome mailing lists around the time I ranted about it 8 months ago).
Remm: Too bad that it’s not only Flash that breaks PA. I was getting hangs just using common FLOSS apps that nearly everyone in the GNOME community uses on a daily basis (e.g. pidgen and totem).
It’s easy to redirect blame to Flash so you can feel better about yourself, but it doesn’t actually solve the problem for end users.
January 27th, 2009 at 8:57 am
I still don’t use HAL.
I’ve encountered too many Nautilus mounting and X input bugs on my distribution of choice. In the end I found it easier to simply mount drives myself and setup X for my input devices manually.
Not everybody stops whining because the technology matures, some people just seek an alternative (Which is why it is important to have them, apparently those bashing GNOME now Qt is under the LGPL don’t see that.)
January 27th, 2009 at 11:34 am
A couple of points – sound is broken for all but the simplest cases in Linux. A lot of the issues surrounding it is specifically because applications like flash go in and try to grab the device for itself. If you don’t have a sound server or your hardware doesn’t support mixing, nothing else will play. Not to mention the need for low latency operation so that video’s audio and video are in sync. Sound is complicated, operating systems are complicated and Desktops are a multiplier to that complication. We either fix those pieces of the subsystem now or we continue to bear the burden of bandaging a dying and incomplete system.
Some people would like to say don’t release until its ready but how does it get ready if you don’t release and release often? Do you ask the Linux companies to bear that cost of Q&A? They do do that – go purchase a support contract with one of the many longer lived commercial distributions. They can’t bear the cost if someone isn’t paying them – that kind of fiscal management is what banks used to do. Eventually someone pays them and they contribute to the fixes but the process there is often slower.
So that leaves the community to do the Q&A if you want to use the latest and greatest and get things working in the shortest amount of time – perhaps instead of complaining, bugs should be filed.
If you aren’t a person who likes to contribute back via bug reports, code or money stick to one of the more conservative distros like CentOS, Debian or Slackware. If you are someone who likes to install from source and configure your own systems – you break it, you get to keep both pieces. There is a solution for most people somewhere and like I said in a couple of years or sooner sound will hopefully be something that just works and we will be looking back at this with nostalgia.
I trust Lennart because unlike all the people I have seen complain, he actually knows a thing or two about sounds and how to make it just work in a maintainable way. When I hear things like “alsa works for my limited sound use” I cringe because that is what holds us back, that is what keeps app writers writing to a broken system, and what causes sound to break in weird ways that have to be constantly maintained. Do you think it is little gnomes fixing all that stuff in your system? No, someone has to maintain it and they are not going to maintain a broken system just for you.
January 27th, 2009 at 12:00 pm
How about getting a QA team to test PA before you dump it on your users?
End users shouldn’t be forced to be your unpaid QA team.
January 27th, 2009 at 1:44 pm
@Joe:
“End users shouldn’t be forced to be your unpaid QA team.”
We aren’t. I’m glad to be some unpaid QA. That’s the only thing I can do to contribute in fact, as I have no skill to actually fix the issues.
If you don’t want to play that role, don’t. J5 said it, use a more conservative distro.
January 27th, 2009 at 2:30 pm
I agree with you. But caring about upgrade paths is more and more important each day, because even though I’ll be able to handle a system being broken by a dbus upgrade, my mom won’t, and I really want her to be able to dist-upgrade unnattended one day =).
January 27th, 2009 at 2:44 pm
Keep going Lennart! I am one of many, I’m sure, who are thrilled to have this. I have 3 computers in my office.. It’s annoying to think I may have been reduced to 3 sets of speakers. But, I found the TCP option and low and behold, IT JUST WORKED!
I have no idea what PA holds for my desktop future.. I can tell you, I am eagerly waiting to see(hear) it!
Keep up the good work.. I for one will happilly QA PA for you.
January 27th, 2009 at 2:46 pm
@Joe,
How you can think asking users to file bugs is somehow an unfair request when they are getting thousands of man hours of work for free is beyond me.
The FOSS community is an ecosystem supported by people who unselfishly contribute what they can. For some that is code, for others that is bug reports, still others help people get setup with Linux and other FOSS software and other who don’t have the time contribute money so that those who can do the work can eat and live a life of their choosing.
We also give you all the bits so if you don’t want to contribute you can freely do what you want and support yourself. Or you can complain, but I don’t care to support those who are not willing to support the ecosystem.
I would really like to hear your moral argument to match your moral outrage at me asking users to file bugs and help get things working. I bet I can prove it is a pretty immoral and selfish argument.
P.S. A lot of distros already have formal QA teams but they can’t test every hardware out there. Be glad you use an OS that allows and encourages users to directly interface with the developers as opposed to close sourced operating systems where you have to pay and still have no way for fixes to come from outside the company (i.e. if they don’t care about it there is no other channel to get it fixed).
January 27th, 2009 at 2:55 pm
@Gustavo
I totally agree but if you look at the other major OS’s they come out with a new OS every two years or longer. So asking for an upgrade path on a conservative distro already works, it is just that most people want the latest and greatest and run one of the 6-12 month distros – you really can’t have it both ways. Plus a lot of the people who complained are also running a development version of GNOME instead of a stable one.
January 27th, 2009 at 7:18 pm
I see by “just works” you actually mean “requires advanced knowledge for fixing, and still is a pain in the butt to figure out in case of problems, which actually happen, despite all the development and packaging effort to make things “work”"
dbus, hal, and udev may not be “young” now, but they still suck. Give me a static /dev any day… specially for servers.. why the hell you’d want all this “dynamic” stuff in a server, I wonder, but it is easily required, since *something* *somewhere* will require 1 piece, and bring the whole chain with it.
January 29th, 2009 at 12:02 pm
J5,
Why some paid FLOSS developers think their users should be forced to do unpaid QA when the developers are getting paid is beyond me.
It doesn’t matter if I’m paying your company or not, your company is paying you. You are not unpaid for your efforts. Same with Lennart. So don’t give me this “unselfishly contribute where they can” bullshit. You are PAID. You are not unselfish. Your job depends on your software working. Just like my job depends on me getting my work done. I do not have the time or desire to do your work for you on top of my own work. That is not being selfish.
If you really truly want FLOSS to be successful, then you can’t expect the masses to be happy about having to do unpaid work for you, work that they didn’t have to do on Mac or Windows.
January 29th, 2009 at 3:31 pm
@JOE
They had to pay for mac or windows. I get paid to make FOSS work for my customers, not for you but I am generous enough to also do a lot of work which I don’t get paid for (being a GNOME board member is a lot of work) and getting FOSS to work for you even if I think you are full of hot air. You’re arguments are so full of holes it all boils down to you want something for free while we just want to give you Freedom (e.g. you are concerned about the small f while I am concerned about the big F). So yes my job does depend on my software working – for my customers and I think we are doing a dam good job – at least the market and our customers are telling us that. But we also do want to make it work for the masses which is what Lennart is doing, just that it is a tall order because there are only so many resource – mainly time is limited. But here is the great catch, users can help out too which makes the time issue become less of an issue.
Now lets come to this money issue you have a problem with. You get paid right Joe? Someone pays you – your job, your mom, the government. Lets assume I consume whatever you do, I am in effect one of your users but I don’t pay or even help out. How about instead of working to get paid you just take that money but do the work I want done? I won’t pitch in or pay you, I won’t even say thank you for the work you have done but you have to do it because it is somehow your moral obligation. I don’t think you would hold onto your job very long but it doesn’t matter because you are somehow obligated to me and asking me to pay in some way would be evil. Oh I see, that only applies to FOSS developers – if they get paid to work on FOSS they aren’t allowed to ask for some reciprocity from users who are not contributing in some small way. Getting paid to work on FOSS is just like any other job, I need it to pay the rent, put food on my table, etc., it is just that with FOSS I can align my work in such a way that it benefits both my customers by working on their issues as a priority, and the community because all of that work is give back. By the way end users aren’t working for free, they get a lot more out of it then the effort it takes to file a bug. Money isn’t the only capital.
I’ve been doing this for about 15 years now both unpaid and for the last five years paid. I’m lucky that I can get paid to do what I love and do it while advancing the idea of Software Freedom. It really pisses me off when some misguided fool who comes up with some twisted moral argument that they don’t even hold themselves up to, tells me that I have an obligation to them. FOSS isn’t about being taken advantage of, it is about creating an ecosystem of communities. Even one of the most important documents in FOSS, the GPL, spells out reciprocity as its main tenant.
Believe me, that is what makes FOSS strong. The fact that it organically builds communities and even strong bonds among competitors. It is economics at its base level, a true free market that comes closest to the ideals only written about in textbooks. The only thing that can bring it down is selfishness. Those wanting to share in all the benefits but forgo on picking up even the tiniest of the related burdens.
January 29th, 2009 at 6:21 pm
Hey J5, please don’t let the trolls get to you!
Your volunteer work is good and appreciated, regardless of how much you get paid to do.
January 29th, 2009 at 6:33 pm
J5… this was truly a lesson in FOSS. Great comment. Keep up the good work and thanks for your software.