Unless something has changed, Darwin is *not* Open Source in the OSI sense, ie it is not Free Software (free as in freedom).
Isn't it a BSD-style license? I'm really undecided as to which I feel is 'better', as I find one more ethically acceptable, and the other more economically viable. But then I'm neither an expert on ethics or economics!
I agree with that wholeheartedly, at least it has been so in the past.
I have to say Rob's demo of OS X really made my jaw drop. In terms of functionality,ease of use and integration, it is exactly what people want. I think Linux has a lot to learn from OS X, certainly if anyone wants it to work as a successful desktop OS. I don't think that it's any coincidence that most of the people I have met who use linux are 1)in education/research, 2) hackers (I mean programmers), 3) people who work in technical jobs, network admins etc. 4) people who have *time*.
And I would admit this; if my girlfriend needed to buy a computer, I would recommend an iMac. She is artistic, not technical, she doesn't enjoy hacking computers, and she wants to take it out of the box, connect it up and start using it. Besides which, I can go exploring the BSD underneath when she's not looking ;-)
Of course I wouldn't recommend M$ X-pee-pee.
on Fri, Apr 26, 2002 at 08:31:46AM +0000, Ricardo Campos wrote:
Apparently, it's the APSL. http://www.gnu.org/philosophy/apsl.html Something was quite amusing: * It is not a true copyleft, because it allows linking with other files which may be entirely proprietary. (copyleft != freedom)
The fact x86 performs so well says a lot about intel's engineers. It is a crap architecture, full of hacks and twists and such to make it backward compatible to 20+ years ago. (Even itanium does, but the actual IA64 doesn't afaik, and sucks less.)
xs@kittenz.org xs@kittenz.org wrote:
Given that it is known (indeed, said elsewhere on that site) that there are non- and weak-copyleft free licences, what is amusing about that statement of fact?
on Fri, Apr 26, 2002 at 10:45:02AM +0000, MJ Ray wrote:
Perhaps the logic didn't translate so well.
GPL =~ copyleft, so GPL software isn't free (freedom) software because, as the above implies, linking with proprietary code is not allowed. Coming from the "Free Software" Foundation...
Similar things are like (from the GPL):
This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.
And then they complain about central control, when fsf centrally controls the license on your software and can change it at any point, at the end-user's option, of course, which may possibly be to their advantage.. (Loop hole in GPL v6.0 allows user to sue you, but when you released software using GPL v2.0, no such loophole existed.)
xs@kittenz.org xs@kittenz.org wrote:
GPL =~ copyleft, so GPL software isn't free (freedom) software because, as the above implies, linking with proprietary code is not allowed.
Now you're trying what is known as "Yet Another Definition Of Free Software". When the FSF and GNU talk about Free Software, it is quite clear what they mean. They are referring to the freedoms to use, study, modify and redistribute the software. Linking the software against proprietary code would remove some or all of those freedoms from later users, so the freedom to do that has to be given up in order to preserve the other ones.
It is not possible to have a licence which has all the freedoms in a consistent way. It's a trade-off and they're quite open about that. See, for example, http://www.fsf.org/licenses/why-not-lgpl.html and others under http://www.fsf.org/philosophy/philosophy.html#LicensingFreeSoftware
I think I prefer losing the ability to link against restricted software to losing the ability to use the software. What about you?
Nothing requires a developer to use the "or any later version" clause in their declaration, but many do. The reason for that is so that if a loophole is discovered in the current one, FSF can "upgrade" the licences of all software using the clause in one move, with no work required from the author. Worst case scenario: you don't like the upgrade, you relicense.
I think you miss the point again, anyway. The complaint is not with any particular "central control" but with proprietary software being "out of the user's control". It is the terms of the software licence which are really important, not the licence of the licence. Even there, FSF is better than many, with author choices and strict controls over how changes to that licence are applied, while you don't get that with some others.
on Fri, Apr 26, 2002 at 11:53:19AM +0000, MJ Ray wrote:
Assuming there are any later users, and assuming those later users don't have access to that same proprietary code.
If I want to link a program with a closed source, patent-ridden, maggot-eaten library, isn't that up to me? After all, I don't have to distribute that linked program, nor the source modifications, if there are any.
I think I prefer losing the ability to link against restricted software to losing the ability to use the software. What about you?
Linking against restricted software doesn't cause the loss of the ability to use the software. See Motif/Lesstif. There's more than one way to do the same thing.
Nothing requires a developer to use the "or any later version" clause in their declaration, but many do.
And GNU doesn't explicitly say you can leave it out. Many people writing software for free, can't afford lawyers.
Yes, but you're assuming that the fsf will only improve and update the gpl to fix loopholes, and that those updates are correct. In the worst case scenario, once the upgrade is out, you've already lost if your source code or program used by anyone but you.
(For a license to be gpl compatible, doesn't it have to be >= the freedom given by the gpl? anyhow.)
xs@kittenz.org xs@kittenz.org wrote:
If I want to link a program with a closed source, patent-ridden, maggot-eaten library, isn't that up to me? [...]
I don't think anything in the GPL prevents that, but it does stop you distributing it as a derived version. It's defintely got a few questions in the GPL FAQ about this situation.
Linking against restricted software doesn't cause the loss of the ability to use the software.
Yes it does, as it obliges people to use the restricted software, possibly including discriminatory licence terms ("free for education" anyone?).
And GNU doesn't explicitly say you can leave it out. Many people writing software for free, can't afford lawyers.
That's why licensing@gnu.org and other similar services are provided. Anyway, why are you assuming that free software is written for free?
Yes, but you're assuming that the fsf will only improve and update the gpl to fix loopholes, and that those updates are correct. [...]
Why do you think the safeguards the FSF has in place will fail to do this in the future?
on Fri, Apr 26, 2002 at 04:17:35PM +0000, MJ Ray wrote:
motif/lesstif, office/<insert your favourite here>, etc..
If it's GPL'd and you want to use it, you have the source code from which you can derive the API of any disagreeable libraries. If you can't be bothered to write a free version, look for alternatives/lobby author/etc.
Anyway, why are you assuming that free software is written for free?
I wasn't.
Because if they need to use them, it will mean the safeguards failed previously.
</friday-moan>
Ricardo Campos corez23@linuxmail.org wrote:
BSD is a free licence. APSL is not. So, I'd say not. I think you have a different objective for economics to me. I want a free and fair market, while it sounds like you want a way for companies to grow fat from cheap/free labour of programmers. That's what BSD licences say to me, anyway: "here, please take all our hard work, make money from it and give us nothing back, not even the ability to fix your copy of our original work".
I have to say Rob's demo of OS X really made my jaw drop. In terms of functionality,ease of use and integration, it is exactly what people want.
http://www.asktog.com/columns/044top10docksucks.html amongst many others.
I think Linux has a lot to learn from OS X, certainly if anyone wants it to work as a successful desktop OS.
We already have a lot of OS X's ground work at http://www.gnustep.org/ but it actually has some flaws that are more noticeable when you can see inside it. It's still quite cool, though.
Maybe it's just that we become one of the above, empowered by the ability to do so?
Anyway, anyone who wants to hack on the foundations of a desktop can email me. We're still at the fixing window management level, sadly, but it's getting better. It's C++, so knowing either that, Xlib, C and generally how to find info would be good.