MS-Funded Study Attacks GPL3 Draft Process

QCMBR writes "A new Microsoft-funded study by a Harvard Business School professor concludes that developers don't want extensive patent licensing requirements in the GPL3. There are significant problems with the study, however, especially given the very small sample size. 'Although 332 emails were sent to various developers, only 34 agreed to participate in the survey — an 11 percent response rate. Of the 34 developers who responded, many of them are associated with projects like Apache and PostgreSQL that don't even use the GPL.' Ars points out that the GPL3 draft editing and review process is highly transparent and inclusive 'to an extent that makes MacCormack's claims of under-representation seem difficult to accept given the small sample size of the study and the number of respondents who contribute to non-GPL projects.'"

Naturally

Does anyone really expect that Microsoft would fund a completely selfless and accurate poll no matter what the subject?

Re:Naturally

Of course not, now where is the Linux funded study by a Harvard Business School Professor about Microsoft's standard EULA?

Re:

Linux tends to have very little profit margin (compared with Microsoft) so its not surprising they chose not to waste it on this sort of pissing match.

Re:The arguments are pretty sound.

Right, the GPL restricts your freedom to restrict freedom.

Re:The arguments are pretty sound.

Except the restriction of freedom and the intolerance of intolerance are necessary to achieve the actual goals of freedom and tolerance. If everyone has unfettered freedom, you're liable to get shoved, stabbed or shot. If everyone has unfettered tolerance, then you increase the sum total of intolerance in comparison to a point a little further back where intolerance isn't tolerated.

Re:The arguments are pretty sound.

Exactly. And thus the degree of freedom for the community as a whole has been decreased by the act of limiting the freedom to limit freedom.

Right, and by outlawing slavery we're restricting people freedom to own slaves.

Re:The arguments are pretty sound.

Exactly. And thus the degree of freedom for the community as a whole has been decreased by the act of limiting the freedom to limit freedom.


Right, well *obviously* we need a new licensing scheme which will limit the freedom to limit the limits on limiting freedom. Duh.

Much like Ronald Reagans Starwars-programme engineering advisors who, when asked what the US would do if the Russians build anti-anti-missile missiles responded "Then we'll build anti-anti-anti-missile-missile missiles".

Honestly, its a no-brainer for anyone who has read Lewis Carroll..

Re:

So taking this same line of reasoning, the degree of freedom for society as a whole has been decreased by eliminating the freedom to own slaves.

Re:

Exactly. And thus the degree of freedom for the community as a whole has been decreased by the act of limiting the freedom to limit freedom.

No freedom is maximized for the community by having the absolute minimum number of restirctions necessary to ensure

Re:The arguments are pretty sound.

The GPL and the BSD license both aim to "maximize" freedon - however, the difference is not about communities, or developers vs users.

The GPL is designed to maximize freedom for all recipients - the first user to get the source must offer the same abilities to anyone he chooses to distribute to.
The BSD license is designed to maximize freedom of those who get the software from the original author - almost carte blanche. On the other hand, users of derivative works only have as much freedom as the developers along the chain decide to allow them to have.

Re:The arguments are pretty sound.

That isn't how I see it. I liken the software I write to my children. A BSD license is like me saying you can do anything you want to with my child including enslaving him and making him work for your own personal profit. Or perhaps more like using my child to help you create your own child that you will then enslave for your own profit. I do not want my children or their children or their children's children to be enslaved. I am saying that if you want to enslave a child, go create one from scratch, and leave my child alone. You are free to do that. However, if you feel that my child is the best at performing a certain task, and you want to have my child help you perform that task, then you have to promise not to enslave him, or use him to create slaves. To me, the GPL is more free than BSD as is forces freedom to exist from generation to generation. To me, the whole "BSD is more free because it allows anybody to do anything with your code" is akin to "Country Xyzpdq is more free because it allows anybody to do anything with anybody". That argument falls short pretty quickly when people start going around taking your stuff or killing your friends or family. And no, I'm not comparing BSD enthusiast to thieves or killers, only pointing out what I consider to be the silliness of the argument.

I'm also not getting what you mean by "[The GPL] actually takes away a lot of freedom..." How so? If I license my code under the GPL, you and anybody else are free to do whatever the GPL states you are allowed to do with the limitations of what the GPL states you are not allowed to do. Without the GPL, you aren't allowed to do anything with my code at all. In other words, just because I choose to license my code to you under terms other than the GPL doesn't make that license automatically BSD. And if I don't license it to you at all, then you can even look at it.

As far as technical excellence goes, what license one uses has nothing to do with ones proficiency at programming. And if you are truly interested in finding the most technically excellent (man this is starting to remind me of Bill and Ted) way to write your piece of software, I would think you would want to know how it is improved in the future by Company X, something the GPL forces them to let you know if they plan to redistribute it. Therefore, it could be argued that those who use licenses like the GPL are really the ones that are truly interested in technical excellence as they want to see a better way to do what they set out to do if anybody ever figures one out.

The GPL is a kitestring

The GPL allows people to use excellent software, without cost, with the freedom to use, modify, and redistribute that software. However, there are some strings attached. Is there and element of greed to that? Yes, there is. I agree that the GPL is as much a constitution as a software silence, and that's how I like it. Even those under the banner of freedom need laws and regulations. The existence of laws may seem contrary to the concept of freedom, but that is not true. Just as a kite cannot truly fly without a string, FOSS cannot flourish unless there is a code of conduct. I am not donating to free software projects so they can become one-way code farms for proprietary software companies.

Re:The arguments are pretty sound.

Take a piece of code with no special license, just default, plain-jane copyright. If you're not the creator, what freedoms do you have to redistribute it?

None.

Now I think it's pretty clear that you can do what you like with the code up to the point of distribution, though not everyone agrees. Regardless, you have absolutely zero freedoms with regard to redistribution of modified or unmodified code.

Now take a piece of code available under the GPL. If you're not the creator, what freedoms do you have to redistribute it?

You have the freedom to redistribute it as far and as wide as you like, provided that you allow everyone who receives it from you the same freedom. You have the freedom to distribute it modified or unmodified. Furthermore, I've only met a few people who believe that the GPL makes any attempt to restrict what you can do with the code apart from redistribution, and every one of those people seemed very confused about copyright and the GPL.

I take from this all two points.

First, under the current Berne Convention regarding copyright law, recipients of copyrighted code have, by default, no rights to redistribute such code.

Second, under the GPL, recipients of copyrighted code have the right to redistribute such code.

I do agree that the BSD and MIT licenses grant more freedoms, but the argument that the GPL reduces the net freedoms in the world where there is no right to redistribute in modified or unmodified form by default is, pardon the phrase, a patently ridiculous semantic game.

Re:The arguments are pretty sound.

If I allow you to make a closed program using my code, and then restrict use of that new program, how am I helping give anyone freedom, other than you the right to restrict other's? Those are the users I am talking about. If I do that I am just providing you with free code so that you can turn around and go against the reason I made it (for it to be free for all to use). BSD works for header files and some libs, but I'd never put anything truly interesting under it. Look what happened with wine. We'd probably have a very capable windows emulator right now if people's attention had not been directed to the anything-but-free Cedega. Ever dollar they make is one that could have gone to making wine better.

The reason legit rules exist is so that people cannot restrict other's rights or infringe on their property (physical, tangible property, that is. It's not property if I can receive it and you can keep it). The GPL does the first.

The fact that Microsoft can't go into the Linux kernel, change some things and call it Windows 2.0 is not a bug, it's a feature. Without the restrictions GPL, open projects could NEVER become substantially better than their closed counterparts. Have a new interface that leaves Vista's in the dust? Microsoft can just copy it. BSD/MIT licenses are an unending and unbeatable game of catchup.

All this, mind you, would be unneeded if there was no ability to control code in the first place. Everyone would be on equal footing, even with closed code (decompiling and reverse engineering are much easier than you may think). Is that ever going to happen? Maybe. Any time soon? No. So, the GPL is the realist's way to "software utopia", the BSD/MIT the idealist's.

Re:The arguments are pretty sound.

"GPL has some pretty harsh restrictions on what the users of the code can do."

No. The GPL has no restrictions on what users of the code can do. The GPL isnt an EULA. The GPL is a copyright license, and as such only becomes relevant once you want to do something you would otherwise be forbidden to do by copyright law, ie, copy, modify and distribute.

"care about the people who only intend to use the compiled software."

Care about the people as in ensuring that they too have access to the code, should the software not perform the task they wish? Care about the people as in care about their right to share the software with friends if they enjoy it?

Caring about people takes many forms; sometimes it means denying others the ability to gain power and control over them.

Typo!

Oh, no! Not an "atack"!

Re:

Only one tack? I'd have thought Microsoft would be able to afford far more than that. Tacks aren't even effective weapons anyway; sure, with its resources, Microsoft could probably get some really big ones, but then it'd be cheaper and easier to just trade

Whatever

I mean seriously, whatever.

In other news...

A FSF-funded survey concludes that MS sucks!

Anyone can create a biased survey that self-serves their own interests.

Re:

Anyone can create a biased survey that self-serves their own interests.

I doubt that, the studies point the other direction [imageshack.us]

Re:

You are mistaken. A study by the Jbeaupre Group shows you can't create a biased study.

Re:In other news...

And anyone can attack a survey based on the institute that produced it.

The real question is, can one attack the survey based on it's merits? Are there flaws in the research methodology or it's conclusions? I'm betting the answer is "yes". But to write off studies based purely on the messenger is nothing but an ad hominem attack, and isn't terribly useful or enlightening.

Re:

But to write off studies based purely on the messenger is nothing but an ad hominem attack

No, it means there is an incentive for the people who did the study to be biased. Even without reading the details, if I found a study by Greenpeace saying "there's n

Re:

Yes, the survey is flawed. One word: selection bias.

Now, the second question: cui bono?.

Add those up, and you get a completely worthless survey.

Mart

Interesting....

A MS-funded study says the GPL3 is a badly done job? Then Stallman must be going in the right direction after all!

Problems not just with the study...

The piece also leaves a bit to be desired. While it states "Of the 34 developers who responded, many of them are associated with projects like Apache and PostgreSQL that don't even use the GPL.", it neglects to mention how many. Of course, I can't be fucked actually reading the study (it is in PDF after all...). But other then that, it looks OK.

On to the study it self, I agree with the authors point that far more then 34 people have participated in the drafting of the GPL v3. Not only GNU folks, but major corporations.

If nothing else, the GPL drafting process doesn't even need to open. The Free Software Foundation could easily have hidden with some lawyers for a couple of months and then simply presented the new GPL. Obviously all the FSF stuff would go over, as would quite a lot of other stuff that has the V2 or later clause. Most developers aren't lawyers, and I'm sure that they would accept the new GPL, even if they didn't have a say in drafting it (compare version two), so long as it looks alright.

Conclusion, the study is stupid and a waste of time. While I don't use the GPL for my own projects (preferring something simpler), they are quite simple projects. For anything major, the GPL does the job, and will no doubt continue to do the job well into the future.

Re:Problems not just with the study...

Well said.

My only problem with GPL v3 as a developer (a hat I've long since given up, and never enjoyed wearing) is that it gives FSF license elitists more reason to feel their license is freer, opener, and in all ways better than any MPL, BSD, or Apache license. I'd rather talk to MS sales division about licensing issues than a bloody GPL zealot.

I have no problem with GPL software, or with the FSF philosophy. I just don't need it shoved down my throat every time I ask a question on a forum or a mailing list. Yes, guys, I get it. Now, how about you help me fix this bug?

Re:

The authors mailed 300+ people and only 34 replied. From the responses of the people they decided whether they were FSF believers, pragmatists (e.g. BSD people) and inferred the opinions of GPLv3 from that.

I think their conclusion that BSD/apache people wo

What MS does not like the GPL3?

Must be good, send it to print!

really?

you would think that if microsoft really did think the GPL hindered opensource they'd do well to keep quiet about it to hinder the competition it would have brought- instead they make empty threats and use a flawed study to support their assertion

Re:

So by making developers believe that Microsoft is "against" GPLv3, it is in fact promoting it...

so if they dont say anything the GPL keeps its momentum and if they say something we ignore them and go to GPL anyway? microsoft is getting worried and they

Huh?

...many of them are associated with projects like Apache and PostgreSQL that don't even use the GPL... ...given the small sample size of the study and the number of respondents who contribute to non-GPL projects.

This prevents them from having a valid opinion of the GPLv3? Maybe they have good reasons for not using the GPL that should be taken into account?

I mean honestly, if you survey 2000 GPL fan boys, what do you suppose they will say about the GPLv3?

Well if such a small percentage responded

would this not indicate a high degree of apathy which tends to bear out the main point?

Ya but...

The best part is that Microsoft has now become the single best reason *to* embrace the GPL3. And to think I ever had doubted.

So...

...exactly what does this have to do with My Rights Online? I'd put it under "Politics" and be done with it.

Bugger Me

In the past Microsoft sniffed and derided the GPL and anything vaguely open source as communist or just plain non-capitalist and generally plain ignored it. Now they're actually funding studies to tell us how about it is, and not only that, they have an agenda of what parts they don't like about it - namely patent reform.

Considering the rather silly deal Microsoft struck with Novell, and the silly deals they'd like to strike with other Linux vendors to get the message across to the corporate sector that if you use open source software you pay Microsoft for IP, this looks a touch suspicious. Maybe the FSF have touched a bit of a nerve somewhere.

It's incredibly funny and rather unbelievably naive that Microsoft would think that anything like this would sway anyone's opinions, certainly in the same manner as one of their 'Get the Facts' studies or one of those 'Windows Server beats everyone' studies. They really haven't learned a whole lot over the years. For them to claim the open source developers, the people who they've derided and don't have much time for Microsoft either, are under represented just seems like quite an above average desperate move.

Almost enough to make me endorse GPL3

I haven't made up my mind concerning GPL3, but Microsoft's war against it is nearly enough to sway me towards GPL3. Microsoft is using cross licensing agreements, and attempting to herd Free Software into a commercial vendor only arena (Novell). Once there

you lost me at MS funded...

in this day and age, and on slashdot in particular, isn't "MS funded" synonymous with "/ignore"?

I get it... I do, I do understand

MS cannot fund any study ever without F, U, 'n' D

11%

34 out of 332? That's an *abysmal* response rate and pretty much means that the study is entirely worthless, no matter what the conclusions are or who actually answered.

what a shcoker...

The new GPL - which will undermine all of M$'s FUD claims over patents because of Novell's vouchers not having dates on them - is thought to be bad by some who was paid by... M$! I'm shocked.

I'm also more shocked, genuinely that Harvard allows people who conduct "studies" like this to be professors... It's just shocking incompetence. I'd be amazed if you could pass an MBA doing shit like this

Re:what a shcoker...

I'm also more shocked, genuinely that Harvard allows people who conduct "studies" like this to be professors... It's just shocking incompetence. I'd be amazed if you could pass an MBA doing shit like this

Come on, this is a business school, they don't know any real math. They think statistics is the art of making up numbers to prove their points.

"A Developers' Bill of Rights", proposed by MS

The funniest thing is that the paper is titled ""A Developers Bill of Rights: What Open Source Developers Want in a Software License."

Yes, Microsoft is proposing a Bill of Rights, for open source developers! Can you believe that?

Re:"A Developers' Bill of Rights", proposed by MS

The funniest thing is that the paper is titled ""A Developers Bill of Rights: What Open Source Developers Want in a Software License."

Yes, Microsoft is proposing a Bill of Rights, for open source developers! Can you believe that?

Okay, I will never - ever - again accuse them of lacking a sense of humour.

See, that's what's missing in the arena of world domination: a bit of drollery. I mean, if an power-hungry megalomaniac can't let his hair down from time to time, where's the point in it?

Where's the S.O.P.?

OK, I know that fake studies are a part of Microsoft's standard operating procedure for affecting the standards and codes proposed by governing bodies, but where's the rest? Shouldn't Microsoft be giving zero-interest "loans" to RMS, sending Eben Moglen to play golf in Scotland (a fact-finding tour), and buying a powerboat for Linus?

Seriously, though, who gives a crap what a Harvard professor, funded or unfunded, with or without a good sample size, claims the average developer wants? The GPL is not supposed to be populist, it's supposed to achieve a purpose. A purpose that most of the world - heck, even much if not most of Slashdot's readership - has never fully grasped. A purpose that is diametrically opposed to software patents.

A virgin writing about sex?

Judging by his faculty biography, Alan D. MacCormack is much like the virgin who writes about sex. He writes a lot about software development, but there is no evidence that he has actually done any.

Don't like GPLv3? Use GPLv2 or BSD.

If developers don't like the licensing changes in the GPLv3, they are always free to use GPLv2, BSD, or any other OSI-approved license. Its not like RMS is going to go around and force people to use a particular license.

If developers are upset that GNU projects will go under a license they don't agree with, well, that's just tough. Either use the BSD equivalents, fork the GPLv2 versions, or write your own. The FSF doesn't exist to please you, it exists to protect the 4 freedoms for all users of free software.

Re:

It looks like someone atacked the spell checker.

Re:Atacks?

Yes its a new form of terrorism where the terrorists puts a large map or diagram on the wall and then place a thumb tack in each spot where they plan to strike (a tack). Then then break for lunch and the afternoon session breaks down into an argument about whether they it might be better to use blu tack to identify the targets because it does not damage the wall behind. Should the maniacs proposing this strategy win out then we will face the even graver danger of Btacks (well Blutacks would just sound silly wouldn't it).

Re:

Maybe what the kernel developers have said about it?

You mean other than Linus (who, by the way, is now "pretty pleased" [com.com] about the newest draft because his concerns were addressed)? Cite sources like I just did.

Maybe MySQL's abstinence from adopting it?

T