http://www.silicon.com/research/spec...9155969,00.htm

I think we knew it was coming but now it's official.

Torvalda?

And I can't blame him. I wouldn't use GPLv3 either, even for new projects. I oppose the concept of trying to force everyone to opensource everything. But enough preaching.

Yes, I think that Mr. Stallman doesn't speak enough about, or maybe doesn't quite want to acknowledge, where his paychecks come from. If you're a public speaker and you collect money from people you speak to ... if you're a "trust fund baby," or you "made it big back then and you're still cashing the checks" or whatever ... you have to acknowledge that this, also, is part of your "big picture," and it may not be so for others.

For all of the references to pure-science (biology, physics), the computer is the driving force of commerce in this world .. and that means it has to accommodate commerce's rules. (Even physics and biology have a commercial side.) Things are built and bought and sold, people get paid from these activities and they, in turn, buy things. For every thousand products in slashdot or freshmeat or sourceforge that are strong and viable, there are ten thousand pieces of "abandonware." In the world of commercial software, too, we have a host of "May flowers," who flash into the world in a blaze of glory, but have shallow roots, and one day they are simply gone. All of the things have to come together, in a commercially viable way. Even a "laborer of love" must buy groceries. Software is very expensive and very hard to sustain. I realize that the main piece of little-niche software that I shepherd is quite unusual for having a (so far) ten-year run... it's older than the entire "dot-com bubble!"

What we have now isn't ideal and always needs improvement -- and rewards it -- but at the very least, it is "viable," and I think we have to be sure that we somehow remain there at all times. Over the last how-many years, Linux has managed to do just that. So, Linus' "pragmatic" approach must be working. As far as I can see, Stallman's only work product since gcc .. admittedly quite an accomplishment .. has been a vast quantity of writing telling us how good things are supposed to be in the world that he lives in. I respect him and what he says, but I fail to see much practicality in it. You can't simply take what doesn't cleanly fit your world view, exclude it, and say that those awkwardly-fitting parts just shouldn't exist at all. It just means that your world-view, however eloquently you speak of it, isn't sufficient to manage the total requirement. And I don't think it's wise, nor is it necessary, to "force a showdown" on such things ... unless ... unless an element of notoriety and counter-culture helps to keep your name in the papers and thus to pay your bills?

That is exactly why I have NO respect for RMS. He made his money, so blast anyone else who wants to make a living with computers straight to Hell, right?

Not all software SHOULD be free. However, free (as in beer/as in speech) alternatives SHOULD and MUST be available. I like Photoshop a heck of a lot better than The Gimp, however I quit using it because Photoshop under wine is slow, and I refuse to run Microsoft Windows for anything other than games and to sync my phone. I have a big problem with Microsoft's anti-customer tactics as of late, as well as their absolute refusal to make ALL administration fully scriptable, and for their "Get the Lies" campaign. When Microsoft compares uptime/downtime, they redefine "downtime" to NOT include maintenance. Now, nearly every *nix app I've encountered (be it on Linux, DEC Unix, HP/UX, Slowlaris, IRIX, or BSD) can be maintained LIVE and can be backed up LIVE. Not so with SQL Server and Microsoft Exchange - two Microsoft applications which are otherwise excellent.

The only *nix app I've worked with which cannot be maintained live is Meeting Maker (now People Cube) - their business model revolves around vendor lock, billable support, and forced upgrades. "OOps, your database won't mount? Send us your database, and oh by the way, that will be $xxx.xx plus you NEED to upgrade if you want this fixed because that release is 19 months old and we drop support for releases after 18 months. Thank you for choosing Meeting Maker."

As the Silicon.com article states, "Torvalds represents a pragmatic approach that accommodates computer industry prevailing practices." Without a pragmatic approach, things generally don't work for long.

However, proprietary standards tend to preserve the status quo in terms of wealth. Wealthy people can have access to data and poor people cannot. Wealthy people can manage their lives more efficiently, and poor people cannot.

Here's a trivial but obvious example... The US Treasury has bond management software on their site that requires Windows. It's good to have the software. It would be better if everyone could use it without shelling out the bucks for a proprietary system, updates, and security software.

Proprietary software has made it possible for many of us to run our lives more efficiently. Free software would make it possible for the rest of us to do the same.

However, without the profit incentive, further advances in efficiency would slow down.

Where is the right compromise? Don't know, but if I were to err, it would be on the side that allows equal access to data.

According to the article, DRM is a big part of the divide. I fear that some form of Trusted Computing is in our future, and I won't like it.

Now, that's hitting the nail square on the head!

Stallman already accepts that his views are not necessarily pragmatic, but idealistic. I believe I had even read his words that free software isn't necessarily going to be *better* than proprietary software. I don't remember which article it was in, however.

I am not a Stallman fan on every issue he speaks on (not even close), however on free software I'm inclined to agree with him in the long run more than anyone else.

I think Stallman's long term goal of free software could easily be destroyed with the next generation of DRM. This is why the GPL 3 is designed to fight against that. On the one hand this could make future versions of linux impossible, because DRM may actually be a necessary inclusion to be "legal" under some of the new laws proposed. Pathetic but true.

For that reason I believe that Linus may be right. However, on ideology I have to agree with Stallman. I don't agree with users not having the ability to know about what is running on their system (and that means access to the source). I also don't agree with DRM at all, because it's designed to "roll back the clock" and I don't think that helps anyone.

What needs to be learned as society is how to better support free/open software. All of us could do a better job of supporting the free alternatives (myself included).

One person here rights that free alternatives should and must be available. Why? If you agree that proprietary software should be in existence why should free alternatives exist? This honestly doesn't make sense to me. I can understand not believing proprietary software is immoral (as Stallman states). What I do not understand is why these alternatives need to exist if you believe proprietary software is desirable? I think proprietary software is something we have to "deal with" but in the long run I'd like to see it gone, which is where I agree with Stallman. I don't think this is going to happen immediately, which is where I agree with Torvalds.

Anyway, the goals should be the same I believe. Proprietary software should eventually fail under the weight of better and more developed open alternatives. But this can only happen with better support on the part of users, as well as more acceptance on the part of the general computing populace. And we can only get more acceptance by being more useful and user-friendly (and no, by that I do not mean throwing away complexity, but more in packaging it so that the complexity isn't thrown at someone all at once).

Stallman's worldview has never existed. It's difficult to say that he's excluding anything, because it's more about the way things should be, rather than the way things are. That's the entire point. If he was trying to be practical it would be the way things are. The entire point of idealism is to change the way things are. If you accept the way things are and work within that, sure it may get work done, but it doesn't change anything.

Practical people, for all their great work, don't change the world. It's impractical people who change things, by doing what they can to conform the world to their views.

You may believe that what Stallman wants is impossible, and perhaps it is, but frankly, we don't KNOW that yet. There is still a long way to go, and personally, I am with him on this goal.

I'm not saying that, for example, Adobe should make Photoshop freely available, nor should Microsoft make Windows freely available if they do not wish to do so.

What I mean is that free alternatives should and must be available. If you want to run something other than Windows, then DRM (trusted computing) should allow you to choose an available alternative. Nor should DRM prevent you from developing and installing your own drivers, giving away free/free drivers. Also, DRM should not prevent you from installing free/free, commercial, or other software on your chosen platform.

If an alternative to say, Exchange is desired,one should be free to choose an alternative, or to develop one if so is inclined.

I personally think that, if I am trying to edit an image and the standards (i.e. of my customer) are very high, "free" is never an issue. Quality, reliability, features, robustness .. these all are.

I fully expect to pay for this, and in return for my purchase of a license I expect to receive all of the protections and rights that are afforded to me under, say, the U.S. Uniform Commercial Code.

You may feel differently. But I feel that, if products like Microsoft Word or Adobe Photoshop were available for this platform, it would sell quite briskly.

As for "DRM," I think that none of us have anything to fear and that the proponents of such schemes are just wishing in their pants as they always have done. If you want to sell something, you won't get very far by asking the customer to accept a pair of handcuffs along with it. And you certainly won't "win friends and influence people" by insisting that what they have done since, oh, the invention of the wire-recorder is "illegal." These guys opposed cassette tapes, too. And they didn't want radios installed in cars. The video-cassette recorder was, to them, a piracy-machine.

Real customers have an annoying tendency to do what they want, and to insist that those businessmen who intend to remain in business must deliver on their requirements. They also have a tendency to reward those businessmen with showers of cash.

If the very things that the DRM people are belly-aching about were all that bad, then Apple would never have sold five million copies of songs on their first three days of business. There would not be literally hundreds of "internet radio stations" out there, legally broadcasting every "out-of-print, nobody out there cares about this old stuff" kind of music imaginable, 24/7. A Luddite fear of change is exactly what the DRM folks do not need. But it is, historically, exactly what they have consistently sought since, oh, the 1940's.

Hmm... cost to buy, < $1,000. Cost to develop a replacement ... let's just say, many times that.

And, "if I am a trucking-company, then my core business is moving freight, not writing computer software." And furthermore, any alternative, even your so-called "free" ones, are going to cost me money, one way or the other. So, I'm looking at buying something. And I want to know about, for instance, your warranty ...

"This the world where my business lives: the world of commerce. Those are the rules I play by, and those are the rules I require all of my suppliers to play by. And if you make the sale, I happen to have the money."

I don't quite understand how Linus can be so sure. The license is just started to be discussed and I am sure the GPLv2 looked amazingly unrealistic when it first came out.

Just that linux was just a 'project' at the time. Now a lot of 'interests' depend on it.

When the license is out of the cycle of discussion and one knows what it says exactly, then we'll know what how the developpers feel about it and a real position can be argued. Right now, I think it's the license that needs to be discussed not whether or not it is likely to be adopted and by whom.

The real issue here is not whether or not Torvalds likes GPL3. The fact of the matter is that there are several hundred copyright holders in the kernel, not just Linus. Therefore, you would need the consent of ALL authors (including some who are dead, so you would need to track down the legal power of their estate) to switch the license for the whole.

The kernel is not released with the "or any later version" provision that the GPL may optionally have, so is bound to only GPL2. Some components may be dual-licensed, and those components could go GPL3/GPL2 dual, but the whole would still be GPL2.