Reply
Thread Tools
Posts: 3,319 | Thanked: 5,610 times | Joined on Aug 2008 @ Finland
#171
Originally Posted by johnkzin View Post
Actually, I very solidly disagree with this one.

The worst thing about most open source projects is that their release engineering and release cycles are abysmally immature.

"release early and often" == "no one with more than 3 functioning brain cells should depend on this software for anything useful".

Release when stable. Release when there's a point to the changes being distributed. Release when the release has matured and been vetted.
No, no and no. Again, you're killing yourself with terminology. Release means something COMPLETELY different in proprietary and Open Source contexts. I know what you meant, but we're talking about two different things here.
 

The Following 3 Users Say Thank You to attila77 For This Useful Post:
allnameswereout's Avatar
Posts: 3,397 | Thanked: 1,212 times | Joined on Jul 2008 @ Netherlands
#172
Originally Posted by Alex Atkin UK View Post
Also talking about hiring a programmer is great if you are a big business, but as an end-user its cheaper to buy a copy of Windows where it "just works".
True, but if you're using N900 in corporate environment the freedom is quite attractive. It is also possible to vote with your money as consumer, just a bit harder. That is why I endorse a donation-based system where the users like you and me, being part of community, can vote with their wallet. Because sometimes a feature or a bugfix is worth some money.

Funny you should bring up PulseAudio too as ever since Mandriva switched to PulseAudio I cannot get sound to work properly. PulseAudio does not support digital passthru over SPDIF which highlights my point exactly of what for me is the problem with Linux. They happily remove functionality I have been using for years in favour of something new and better, which is not feature complete. Just like how mucked up KDE 4 is compared to KDE 3.
Yes, OSS also supported sound cards which ALSA did not support. X.Org also still does not fully support some graphics cards. It is all a matter of 1) developer itch or financial incent 2) rarity of hardware or feature. If PA or KDE4 or ALSA doesn't cut it for you, you use the predecessor (legacy software) instead. There is nothing wrong with that, and there is nothing wrong with running an old, stable version of an OS. I also gave PA as example because it supports audio over network seemingly. There is no such thing for OSX or Windows, while PA is ported to OSX and Windows.
__________________
Goosfraba! All text written by allnameswereout is public domain unless stated otherwise. Thank you for sharing your output!
 
ysss's Avatar
Posts: 4,384 | Thanked: 5,524 times | Joined on Jul 2007 @ ˙ǝɹǝɥʍou
#173
@tag writer:

I notice the tag "try to correct an error" on this thread. Would the writer care to explain more?
 
Posts: 716 | Thanked: 303 times | Joined on Sep 2009 @ Sheffield, UK
#174
Originally Posted by allnameswereout View Post
True, but if you're using N900 in corporate environment the freedom is quite attractive. It is also possible to vote with your money as consumer, just a bit harder. That is why I endorse a donation-based system where the users like you and me, being part of community, can vote with their wallet. Because sometimes a feature or a bugfix is worth some money.
I totally agree in that case.

I would happily donate to any project that is going to deliver a feature I want or even that already has. The trouble is, if you donate to an open source project there is a chance the developer will get bored or real work will get in the way.

Commercial on the other hand, it IS their real work so you pay them to implement a feature, they better do it. They also generally have more resources to throw at it and will not just do "the fun/interesting" bits, which any developer doing it as a hobby is likely to do.

Originally Posted by allnameswereout View Post
Yes, OSS also supported sound cards which ALSA did not support. X.Org also still does not fully support some graphics cards. It is all a matter of 1) developer itch or financial incent 2) rarity of hardware or feature. If PA or KDE4 or ALSA doesn't cut it for you, you use the predecessor (legacy software) instead. There is nothing wrong with that, and there is nothing wrong with running an old, stable version of an OS. I also gave PA as example because it supports audio over network seemingly. There is no such thing for OSX or Windows, while PA is ported to OSX and Windows.
The thing is, I can't get it to work how it used to as they expect you to do it their way or be an ALSA GURU who knows how to hack it to work how you want. Granted Windows you might not even be able to change its behaviour at all, but then they are unlikely to release with the functionality missing like PA has.
 
R-R's Avatar
Posts: 739 | Thanked: 242 times | Joined on Sep 2007 @ Montreal
#175
Originally Posted by qgil View Post
If the hardware vendors can't make business then it's really game over. Their current business model is mostly based on licensing. It's a tough market of big investments that require big operations to be profitable. The competition is fierce. And still free software is doing some progress in that area as well, so someone is putting the brain to work with business models considering OSS.
Do you imply that hiding the hardware "API" (registers) will somehow prevent others from reverse engineering the chips to clone them?

I mean, with millions at stake, I'm sure they can afford a few IDA Pro hackers... Binary is source code after all, no matter how obfuscated you make it...
If a few kid can crack games, I'm sure dedicated people can reverse engineer anything!

So, what is the real advantage in locking down your hardware through some proprietary gate-keeping software when you benefit from it being used as widely as possible and you don't sell any software at all ?


Originally Posted by qgil View Post
If they were commodity you would find several options to choose at the same level or better. This happens in the traditional desktop, but it's still not the case in the mobile. The foundations and the API are all public so there is no hard obstacle for the free software community to commoditize them. As mentioned before, even Nokia is a contributor supporting directly or indirectly projects like Mozilla or Canola.
Well, there are a few browser available and a lot of media players on maemo's repository...
 

The Following User Says Thank You to R-R For This Useful Post:
R-R's Avatar
Posts: 739 | Thanked: 242 times | Joined on Sep 2007 @ Montreal
#176
 

The Following 5 Users Say Thank You to R-R For This Useful Post:
Posts: 716 | Thanked: 303 times | Joined on Sep 2009 @ Sheffield, UK
#177
You lock it down you gain a competitive advantage. You buy yourself a few months leeway while they are hacking away trying to clone your device.

As was said earlier, if you don't do that they may even release a clone product before the official product itself launches.

I do not necessarily agree with the Release Early, Release Often for all software though. For one I wish KDE 4 had not come out until it was up to standard with KDE 3. Too many distros have adopted KDE 4 and practically dropped KDE 3, when 4 is just not ready yet. Had it not been "released early" they would have stuck with 3 that bit longer and avoided the ton of teething problems people have with 4.

I get it, you need people to test it. But surely there are plenty enough people willing to risk early software rather than forcing it onto an unwilling public. But this is getting off topic.
 
R-R's Avatar
Posts: 739 | Thanked: 242 times | Joined on Sep 2007 @ Montreal
#178
Originally Posted by Alex Atkin UK View Post
You lock it down you gain a competitive advantage. You buy yourself a few months leeway while they are hacking away trying to clone your device.

As was said earlier, if you don't do that they may even release a clone product before the official product itself launches.
I do not agree on this part but I've already said enough on the subject...

Originally Posted by Alex Atkin UK View Post
I do not necessarily agree with the Release Early, Release Often for all software though. For one I wish KDE 4 had not come out until it was up to standard with KDE 3. Too many distros have adopted KDE 4 and practically dropped KDE 3, when 4 is just not ready yet. Had it not been "released early" they would have stuck with 3 that bit longer and avoided the ton of teething problems people have with 4.

I get it, you need people to test it. But surely there are plenty enough people willing to risk early software rather than forcing it onto an unwilling public. But this is getting off topic.
Well, R.E.R.O. has nothing to do with bad naming/numbering choices...

It's all about giving access to the software early so to get as many eyes as possible on the code and fix things.
It's definitely not branding alpha version as final 4.0.0 product ready to be shipped! If KDE ****ed up, it's totally unrelated! :/
 

The Following User Says Thank You to R-R For This Useful Post:
qgil's Avatar
Posts: 3,105 | Thanked: 11,088 times | Joined on Jul 2007 @ Mountain View (CA, USA)
#179
Originally Posted by R-R View Post
Do you imply that hiding the hardware "API" (registers) will somehow prevent others from reverse engineering the chips to clone them?

I mean, with millions at stake, I'm sure they can afford a few IDA Pro hackers... Binary is source code after all, no matter how obfuscated you make it...
If a few kid can crack games, I'm sure dedicated people can reverse engineer anything!
They make their money licensing and selling to companies like Nokia, who actually play by the business rules and won't create a team to reverse engineer their work. If you introduce a free software license to the mix you do need to have a look to your business plan in order to keep your profits or increase them.

In any case, feel free discussing the topic with them. Free software in the hardware adaptation layer is something that companies like Nokia would be also interested about.

Well, there are a few browser available and a lot of media players on maemo's repository... [/QUOTE]

As starters, the serious browser alternatives would imply Webkit, which is not supported in Maemo and is not a trivial task to productize. Moving away from the Mozilla family is a deeper strategic move aniway, not worth if the only "problem" is the UI layer being not open source. And still, I haven't said the browser UI will be closed, or open.

For Media Players let's not forget that Python is not officially supported and productizing it is not trivial. MPlayer is a beast on it's own and there are many reasons not to ship it pre-installed. Probably the maintainers of MPlayer and the Maemo port would agree on this.

And still I believer the Maemo 5 browser and media player would beat these options in many aspects that are essential even if not so evident: performance, footprint, stability, etc.

But we shouldn't start discussing browser, media player or other specific app here. Looking forward to your Brainstorm proposals for browser and media player, listing the serious candidates for solutions and having their own threads for discussion.
 

The Following 3 Users Say Thank You to qgil For This Useful Post:
allnameswereout's Avatar
Posts: 3,397 | Thanked: 1,212 times | Joined on Jul 2008 @ Netherlands
#180
Originally Posted by Alex Atkin UK View Post
For one I wish KDE 4 had not come out until it was up to standard with KDE 3. Too many distros have adopted KDE 4 and practically dropped KDE 3, when 4 is just not ready yet.
Stick with what works for you. Don't upgrade your OS if it works fine.
__________________
Goosfraba! All text written by allnameswereout is public domain unless stated otherwise. Thank you for sharing your output!
 

The Following 2 Users Say Thank You to allnameswereout For This Useful Post:
Reply

Tags
balance, basic rights, defective by design, get your stink on, gpl holy crusade, open source, open source advocacy, sw wants to be free, try to correct an error, why isn't the gpl law?!, zealots be here


 
Forum Jump


All times are GMT. The time now is 18:08.