09-13-2010, 12:32 PM
Of course it's unfair but it's also a sign of something having gone pretty seriously wrong throughout the years that we ended up with this situation in the first place. I find it hard to believe Martin as the project manager - of a said to be single-member project - is completely innocent of the current situation.
Originally Posted by mugginz
09-13-2010, 12:41 PM
Tis the nature of the open development beast that you can have a situation in which there is a requirement for code to be written and someone volunteers to write that code with only an eye to being helpful, and not a long term project management eye. If when he volunteered it was a condition that he not only provide code but also must provide a team of others to back him when his speed of output didn't suit those who would use the fruits of his labours for free then he might have passed on the opportunity.
Originally Posted by nanonyme
He's publicly asked for others to help with testing and bug reports and it would seem fair for him to do so. If more users provided their time and labour then KDE would be better for it.
I don't think it's his fault that there aren't more people willing to help.
09-13-2010, 12:42 PM
what the fuck ? am i suddenly woke up in different world, world where stupid windows-like update and what-majority-supposedly-have problems exist for FOSS users, world of ultimate "stability" masturbation and jerking off on two in-rapid-development projects in a same time ?
Originally Posted by pingufunkybeat
when there is an issue, the first thing you do is update.
if there is a bug in your kernel and you submit a report, the first thing you going to be asked is to debug this shit as deeply as you can and/or update to the latest git snapshot and reckeck. that is normal.
if you have issues with libGL, the first thing you are asked is to fucking update or GTFO.
and guess what will happen when you will stumble on bug in KDE ?
his card supported alright and mesa devs are easy to reach and there is no empty extension stubs for misleading app in libGL in believing of their false support. buggy maybe but there is a "disable/enable" check box for user to temporarly disable particular glitchy effect until update.
this is not what desktop environment should auto decide for user and even if it is - he has all he needs to do his work at reach of his hand.
the way i see it - the whole damn graphic stack is being completely rewritten - ~10 years of work being redone in few by few. there is no "stable" for that any time soon and you have to deal with it. there is some hotheads who even want drivers merged in server, damn it; forget about stable or use binary shit for now.
what a bullshit drama.
i don't like people who lie and accuse others in their own indolence. should i write a fucking patch to his mother for her to fix it ?
what the fuck do you even talking about ?
anyway, when there are words like "market share", "most widespread" and "stability" anywhere near, the discussion is busted by asshats and become useless.
09-13-2010, 12:45 PM
The KDE user base is highlighting a need for more man power.
Some are more demanding of it.
Some are demanding one of the KDE volunteers provide more time than he has.
When some of those do it via trolling on forums I ask myself, "self, I wonder how much time those trolls are themselves donating to the KDE cause."
I suspect that in some cases it's not very much at all.
09-13-2010, 12:46 PM
What I'm getting at is that if there's no volunteers to help him and he doesn't have time for the project, there's only one option really which makes sense and puts a stop at complaints on the project quality: end it. Allow someone else to pick up the pieces if they want to and help them if they turn up.
09-13-2010, 12:58 PM
So what happens if he pulls the plug and no one else comes along?
Originally Posted by nanonyme
But more to it, if KDE 4.5.1 breaks on your hardware, stick with 4.4 until 4.5.something works there. Or help make 4.5.2 work on your hardware or switch to Gnome, or complain to the KDE higher ups, but don't hate on Martin.
There are shortcomings in mesa, that we know. Martin's trying to code around them. Sometimes when drivers advertise capabilities they can't reliably support in all situations there can be a bit of to and fro in order to get some software happy with the drivers the code has to speak to.
There can be moments in time when there's breakage. In the long run his plan seems to cover the contingencies required to provide a happy outcome for all. Just at the moment people will need to be patient or help or something. What ever they do I just would hope they don't bitch at Martin for being a volunteer who is providing good work that just at the moment isn't able to deal with some of mesa's brokenness.
09-13-2010, 01:00 PM
Okay, I'm gonna stop. This is just getting inflammatory.
09-13-2010, 01:02 PM
09-13-2010, 01:10 PM
Can't Mesa get a version string (or whatever; I'm not too knowledgeable on this front) switcher? Like for example:
09-13-2010, 01:17 PM
Well were a driver say's "Hey yeh I can do that" when asked but actually can't reliably then I guess that's why he has the blacklist.
Originally Posted by V!NCENT
But in order to provide a "if this exact version then don't use that" which works 100% of the time you'd need a farm of machines spitting out test results from not only the different distro's, but the distro's updates, and every git update for every card/cpu/mobo combo and so on.
Tags for this Thread