Beta

×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Dispelling BSD License Misconceptions

ScuttleMonkey posted more than 7 years ago | from the deflating-the-demons dept.

The Courts 202

AlanS2002 writes "Groklaw is hosting an article by Brendan Scott which looks at the misconceptions surrounding the BSD license. From the article: 'We observe that there exists a broad misconception that the BSD permits the licensing of BSD code and modifications of BSD code under closed source licenses. In this paper we put forward an argument to the effect that the terms of the BSD require BSD code and modifications to BSD code to be licensed under the terms of the BSD license. We look at some possible consequences and observe that this licensing requirement could have serious impacts on the unwary.'"

cancel ×

202 comments

Sorry! There are no comments related to the filter you selected.

What about the MIT license? (3, Insightful)

Watson Ladd (955755) | more than 7 years ago | (#17617040)

The MIT license looks like it explictly permits relicensing. Would someone more qualified in the legal art then me care to comment?

Relicensing doesn't matter! (1, Interesting)

Anonymous Coward | more than 7 years ago | (#17617812)

Is re-licensing really that important?

The BSD license means that you can do what you want with the code, as long as you display the license. Plain and simple. So if you modify BSD licensed code (and compile it), the code you didn't write but used is under the BSD license (meaning you can do what you want) and you can license your own code however you want.

What the article is trying to argue is that you must include the license with the distribution, ignoring that it can be specified that different licenses can be specified for subsections of code. IANAL, but the key word in the sentence that I think they abuse is "redistributions" which I interpret to be "redistributions of the code that this license applies to" rather than "distributions of this code and other code". And I think that the word "redistributions" rather than "distributions" argues my point.

Even if their definition was the correct one, it still wouldn't matter to most (such as MS), because redistributing the source code is a choice left to the programmer.

This seems like FUD to me, not something I would expect from Groklaw.

Re:Relicensing doesn't matter! (1)

FST777 (913657) | more than 7 years ago | (#17618856)

Consider the following scenario:
You write a piece of software, using portions of a BSD-licensed work. You intent to distribute only in binary form, with your own license, to refrain other parties of distributing your work. Since you used the BSD license, you MUST put the conditions mentioned in the original license in, our accompanying with, your license.

The question is: do you have to put the following sentence in there too?
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met
If so, any of your customers is free to redistribute all the work you provided.

What if a third party had access to your source code, for instance for stress-testing or to prevent Zero-Day exploits. They signed and NDA, but then they receive the final product, with the BSD-license. The new question is: are they permitted to make deriative works or binary distributions of your work, since with the BSD-license they are "free" to do what they want with the accompanying source code?

All in all, it stands or falls IMHO with the question if you have to put the mentioned sentence in or not. If yes: your customers are as free to do what they want with their copy of your work as you are free to do what you want with the original software. If no: you have total control over your software, provided that you provide the list of conditions.

Re:Relicensing doesn't matter! (1, Interesting)

Anonymous Coward | more than 7 years ago | (#17619098)

As I stated earlier, it says "redistributions of source code..." not "distributions of source code..." implying that it only applies to something given to you. Thus you have control over your software, but not the software given to you. (Thus not permit redistribution of the whole binary, which sounds like the argument in the Merchant of Venice when Shylock was permitted the pound of flesh, but not a drop of blood and I leave as an open question for people knowledgeable in law and CS as to how it would be resolved)

As for the NDA, that would depend on how it is written. But if you legally agree not to use product x, even though you're allowed to, you still can't use product x.

Re:Relicensing doesn't matter! (1)

jrumney (197329) | more than 7 years ago | (#17619390)

The question is: do you have to put the following sentence in there too?

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met

If so, any of your customers is free to redistribute all the work you provided.

They are free to distribute the work that is released under the BSD license. They are not free to distribute any modifications to that work that you choose to release under a more restrictive license. If they can't figure out which is which, then they'd best not distribute.

Re:Relicensing doesn't matter! (1)

FST777 (913657) | more than 7 years ago | (#17619504)

That leaves the supplier with the responsibility to exactly specify to which portions that sentence apply. If not, it would apply to the whole package.

It's the "with or without modification" that still leaves me puzzled. It could be argued that the whole license applies to the whole work, with modifications.

Re:What about the MIT license? (1)

albalbo (33890) | more than 7 years ago | (#17618358)

I don't think so. It says you can sub-license it; that doesn't let you change the conditions of the licence. That's basically how people receive the license to use the software.

In general, no-one except the copyright holder has the ability to set the licensing terms of something. I disagree with the article, though, in that the practical consequences aren't particularly disastrous - complying with the terms of new BSD / MIT / etc. in parallel with those of another license doesn't look too hard to me.

If it's on a torrent it's free, free I tells you (0, Troll)

Anonymous Coward | more than 7 years ago | (#17617064)

Yeah right like anyone gives a shit about any sort of "copyright" around here - get a clue loser .

Re: You're damn right. (-1, Troll)

Anonymous Coward | more than 7 years ago | (#17617436)

IBM stole OpenServer code and put in Linux for christsakes and Groklaw spend their lives defending that.

"Open source" and P2P have a noble facade, but they are all about STEALING.

Fascinating (2, Informative)

JoshJ (1009085) | more than 7 years ago | (#17617082)

If this turns out to be true, it could have some pretty far-reaching effects, potentially damaging Microsoft, Apple, and even certain F/OSS projects. There seems to be quite the firestorm of controversy over the BSD license lately- perhaps it'd just be better to use a license that isn't so controversial- MIT if you want something to be available for use in closed-source products, or GPL if you don't.

Re:Fascinating (2, Informative)

AlanS2002 (580378) | more than 7 years ago | (#17617160)

The BSD license doesn't mean that you have to release the source code to any modification/redistributions. It just says that modifications/redistributions (weather that is source, binary or both) have to be released under the BSD license. In that regard I don't think it would have much impact on Apple (I presume you are referring to Mac OS X) or Microsoft (I presume you are talking about their FTP app.

Re:Fascinating (2, Interesting)

QuantumG (50515) | more than 7 years ago | (#17617588)

Does mean we can legally reverse engineer that binary to see what proprietary extensions they have made to the original BSD licensed code however.. and, if we were doing decompilation, the resulting "source code" would be BSD licensed too.

Re:Fascinating (3, Interesting)

kcbrown (7426) | more than 7 years ago | (#17617654)

The BSD license doesn't mean that you have to release the source code to any modification/redistributions. It just says that modifications/redistributions (weather that is source, binary or both) have to be released under the BSD license. In that regard I don't think it would have much impact on Apple (I presume you are referring to Mac OS X) or Microsoft (I presume you are talking about their FTP app.

I think your understanding of the license is basically correct, but you don't seem to be seeing the implications.

There is one point that has a huge effect on both Microsoft and Apple: redistribution in any form not explicitly allowed by the licence is prohibited. That's standard copyright law. Additionally, you can't add licensing terms that conflict with the original terms. It's unclear to me whether or not you can add any terms at all (e.g., by adding the GPL to it).

That means one of two possible things: either Apple and Microsoft are in violation of copyright because they are redistributing code that incorporates BSD code under a more restrictive non-BSD license, or the recipient of said code has the rights to the binaries under the BSD license (and can thus freely redistribute the binary!).

Those are surprising and far-reaching conclusions.

With respect to Microsoft, it could have a big impact, because they incorporate BSD code into their TCP/IP stack if I'm not mistaken. In the case of OSX, a very large amount of the code is BSD code or incorporates BSD code. In both cases, they relicense the binary code under a far more restrictive (as regards redistribution) license than the BSD license.

Re:Fascinating (2, Informative)

init100 (915886) | more than 7 years ago | (#17618828)

With respect to Microsoft, it could have a big impact, because they incorporate BSD code into their TCP/IP stack if I'm not mistaken.

According to Microsoft, this is no longer true. BSD code were only used earlier to get TCP/IP functionality into Windows quickly when it became obvious that Internet (and not Microsoft Network, a.k.a. MSN) would be the next "big thing"

But of course the source isn't available , so we can't verify this claim.

Re:Fascinating (3, Informative)

sik0fewl (561285) | more than 7 years ago | (#17618474)

The BSD licence does not say that any modifications must also be released under the BSD licence. It does say that that a copy of the original licence, copyright notices, disclaimer, etc, must be including in any redistribution of the source or binary.

In fact, the BSD licence does not say anything about licensing any code—at all.

Re:Fascinating (1, Interesting)

Anonymous Coward | more than 7 years ago | (#17617194)

It's interesting. It wouldn't turn out to be the first "too clever by a half license" the Artistic license, for example also turned out to not be clear enough. The worst about this isn't that the license may be wrong; that can be fixed. It's that it shows that they weren't thinking clearly enough about the legal aspects. How many other skeletons may they have? I guess that's what caused the AT&T lawsuit and it's another reason to try to be as careful as the Free Software Foundation.

Re:Fascinating (1)

smitty_one_each (243267) | more than 7 years ago | (#17618040)

Before the law, there stands a guard.
A man comes from the country, begging admittance to the law.
But the guard cannot admit him.
May he hope to enter at a later time?
That is possible, said the guard.
The man tries to peer through the entrance.
He'd been taught that the law was to be accessible to every man.
"Do not attempt to enter without my permission", says the guard. "I am very powerful. Yet I am the least of all the guards. From hall to hall, door after door, each guard is more powerful than the last."
By the guard's permission, the man sits by the side of the door, and there he waits.
For years, he waits.
Everything he has, he gives away in the hope of bribing the guard, who never fails to say to him "I take what you give me only so that you will not feel that you left something undone."
Keeping his watch during the long years, the man has come to know even the fleas on the guard's fur collar.
Growing childish in old age, he begs the fleas to persuade the guard to change his mind and allow him to enter.
His sight has dimmed, but in the darkness he perceives a radiance streaming immortally from the door of the law.
And now, before he dies, all he's experienced condenses into one question, a question he's never asked.
He beckons the guard.
Says the guard, "You are insatiable! What is it now?"
Says the man, "Every man strives to attain the law. How is it then that in all these years, no one else has ever come here, seeking admittance?"
His hearing has failed, so the guard yells into his ear. "Nobody else but you could ever have obtained admittance. No one else could enter this door! This door was intended only for you! And now, I'm going to close it."
This tale is told during the story called "The Trial" [imdb.com] .
It's been said that the logic of this story is the logic of a dream... a nightmare.

Re:Fascinating (3, Insightful)

Timesprout (579035) | more than 7 years ago | (#17617260)

Its not facinating, its more license genital tugging that OSS seems obsessed with. The argument itself is untested, weak and clearly contrary to the spirit of the BSD license which ultimately is quite liberal so even if the assumptions made in the argument were true little would actually change. Reads more like a GPL is the best license in the world scare tactic.

Re:Fascinating (2, Interesting)

Aim Here (765712) | more than 7 years ago | (#17619136)

The argument might be weak and contrary to what most people think about the BSD license, but it's not necessarily untested; there was one instance where the holders of a BSD-licensed copyright argued something similar.

The email client pine used to be nominally BSD-licensed, until the FSF tried to make a GPLed version. Then Washington University got all sniffy about relicensing, claimed that the BSD license didn't say what everybody thought it said (something about distribution being allowed, and modification being allowed, but distributing modified versions being not allowed)and threatened legal action against the FSF. The FSF backed down before any lawsuit took place (it's probably good politics not to piss off a University), and Washington changed the license so that everybody understood what they meant.

So the one test case where someone got lawyers involved over the wording of the BSD license went in favour of a nonstandard reading of the license.

Re:Fascinating (2, Informative)

timeOday (582209) | more than 7 years ago | (#17617270)

Yes, I am baffled by this. It makes BSD sound like the GPL. I've always thought BSD was pretty much the wild west except you have to pass along their copyright statement with your software.

Re:Fascinating (1)

Anne Thwacks (531696) | more than 7 years ago | (#17617346)

You also have to pass on their total lack of warranty - hence MS are debarred by law from warranting that Windows is fit for any purpose whatever. Its not logical - its the law!

IANAL etc

Re:Fascinating (2, Informative)

LurkerXXX (667952) | more than 7 years ago | (#17617850)

Incorrect. MS is totally free to warranty that Windows is fit for any purpose they care to warranty it for. What they can't do is pass on that responsibility to the authors of any BSD code in it for the BSD parts. That's why the warranty part of the BSD license stays with the author notice. So 'The buck stops at MS' if they want to warranty it for any purposes, but they are absolutely free to warranty it if they like.

Re:Fascinating (1)

zotz (3951) | more than 7 years ago | (#17617750)

I haven't read the article, but from skimming the comments so far, if correct it would be more like a copyleft without source requirements. Would you agree? (If not, do I really need to read the article?)

all the best,

drew

Re:Fascinating (1)

NSIM (953498) | more than 7 years ago | (#17617716)

For Windows, probably not as far reaching as you might think. My understanding is that only piece of BSD code was the IP stack, and this has been rewritten from scratch in VISTA (coincidence?)

Re:Fascinating (1)

GigsVT (208848) | more than 7 years ago | (#17618268)

Microsoft uses BSD (style) licensed code all over the place. It would be stupid for them not to. An example, the broken PNG support in IE was because they used an ancient version of libpng for the longest time, assumedly forked so they didn't really want to resync with the upstream code.

These pieces of software are freely licensed for a reason, they want people like MS to use them. It would be silly for MS to write zlib and libpng and other common things from scratch.

Re:Fascinating (1)

Intron (870560) | more than 7 years ago | (#17618094)

What the article entirely misses is that a license is only as restrictive as the licensors wish to pursue it. If someone decides to release closed-source modified BSD code, who is going to write a check to a lawyer to stop them?

Good news (0, Redundant)

Anne Thwacks (531696) | more than 7 years ago | (#17617132)

Now MS will have to release Windows under the BSD License!

IANAL ;-}

Re:Good news (2, Insightful)

Undertaker43017 (586306) | more than 7 years ago | (#17617252)

Would that signal the start of the Armageddon? Since surely releasing Windows source code upon humanity would constitute a plague of some sort.

Re:Good news (1)

rucs_hack (784150) | more than 7 years ago | (#17617338)

Not quite. I don't have the exact facts to hand, but in the case of the TCP/IP stack, the BSD Bods were asked to develop it on their own so as to ensure internal consistancy, then everyone else was to use what they produced. The tcp/ip stack in windows came from that effort, as did the stack in many other OSs.

Much therefore of Unix stuff in windows has a perfect right to be there. I was told by someone who enjoyed poking around deep in windows that some original BSD text can still be seen in the windows stuff, although I have not a clue how to find it myself.

If anyone else does know more about this, then feel free to make my poor amount of information seem even more insignificant...

Re:Good news (1)

nxsty (942984) | more than 7 years ago | (#17617560)

Much therefore of Unix stuff in windows has a perfect right to be there. I was told by someone who enjoyed poking around deep in windows that some original BSD text can still be seen in the windows stuff, although I have not a clue how to find it myself.

Its probably this:

c-5ac4e155:/Volumes/Untitled/WINDOWS/system32 Simon$ strings ftp.exe | grep -1 Copyright
OemToCharBuffA
@(#) Copyright (c) 1983 The Regents of the University of California.
  All rights reserved.

Re:Good news (1)

rucs_hack (784150) | more than 7 years ago | (#17617982)

yup, seems pretty conclusive that was what he was talking about.

Only in Australia! (see article for details) (4, Informative)

Anonymous Coward | more than 7 years ago | (#17617138)

One tidbit seems to be ignored--this would only apply under AUSTRALIAN law, per the article.

But, if true, it might mean that the BSD is indeed "viral" in Australia!

Wonder what Microsoft might have to do about all that old BSD networking code they use if this is true?

Re:Only in Australia! (see article for details) (1)

JoshJ (1009085) | more than 7 years ago | (#17617180)

Oh, good catch. I missed that. If this is "true" then I'd just expect Microsoft to just not release Windows in Australia anymore. Or just get the Aussie court system to look the other way.

Re:Only in Australia! (see article for details) (2, Insightful)

CDarklock (869868) | more than 7 years ago | (#17618866)

I see at least one glaring flaw in the reasoning. The author states that redistribution of modified source code is a distribution of source code and hence it must be distributed under those restrictions.

But redistribution of unrelated source code is also a distribution of source code. Why stop at applying the license to source code written explicitly to extend the licensed code? You could extend it to source code written by anyone using the licensed code, whether their new code interacts with the licensed code or not. In fact, if you interpret the license literally, you could extend the license to any and all source code everywhere - even if the author never agreed to the license. It doesn't say "any distribution of THIS source code", it says "any distribution of source code". That could mean any source code by anyone, anywhere in the world! This license actually out-virals the GPL, by requiring everyone everywhere to release all of their code under the BSD license... even if they don't know it!

Of course, it would be patently absurd to suggest that the license applies in this way, yet this is precisely the sort of argument the author uses. The ambiguity of the phrase "any distribution of source code" is the focal point of his premise, but it's clear to anyone with half a brain that the license applies only to redistribution of the code that already contains it. The argument is nothing more than an exercise in mental masturbation.

Re:Only in Australia! (see article for details) (1)

Beryllium Sphere(tm) (193358) | more than 7 years ago | (#17618932)

Australia's legal structure inherits from the UK, as does the US system. The guy did mention that his speculations might well apply to the US and the UK.

It's Funny. Laugh. (4, Insightful)

nuzak (959558) | more than 7 years ago | (#17617206)

This novel interpretation of the BSD license requires various syntax games with the license text that simply aren't supported by common sense interpretation. And yes, while common sense may not be the output of much of any area of law, contract law included, it's still got a lot of weight as input.

It's not a viral license, no matter how much anyone wants to twist their personal interpretation of it. All in all, it's pretty funny, telling the licensors how they actually intended a different outcome than what they, well, intended.

It's not funny. (0)

Anonymous Coward | more than 7 years ago | (#17617356)

This novel interpretation of the BSD license requires various syntax games with the license text that simply aren't supported by common sense interpretation. And yes, while common sense may not be the output of much of any area of law, contract law included, it's still got a lot of weight as input.,?

That's the problem with the BSD and the GPL v.x and the LGPL, it's not easily interpreted and it's not intuitive. My god, any paragraph can be interpreted in different ways, as a matter of fact, when I post my interpretation of any paragraph, there will be a bunch of other posts each with their own differing interpretation. There are also flames about how stupid I am. If I were to start any business that uses F/OSS in any development way, I will have to hire an attourney that specializes in IP law. Period. I am not a lawyer, so I am not qualified to read a contract or license. If you are not a lawyer and you think you're qualified to interpret the GPL, LGPL, or BSD, you're kidding yourself.

Re:It's not funny. (1)

Lord Bitman (95493) | more than 7 years ago | (#17617994)

same goes with starting any business that deals with products covered by IP law. YANAL, and most people in any sort of business will admit that.

Re:It's not funny. (0)

Anonymous Coward | more than 7 years ago | (#17618156)

If you are starting a software company, shouldn't you hire an attorney that specializes in IP law anyway?

Re:It's Funny. Laugh. (5, Insightful)

LurkerXXX (667952) | more than 7 years ago | (#17617386)

Call a spade a spade. It's not a 'novel interpretation', it's FUD.

Someone who want all software to be 'open' only in the GPL notion is trying to spread FUD that the truely free BSD license also has the same viral restrictions. Don't buy it, don't spread it. It's FUD.

To the author of this crap 'interpretation': If you want your software under GPL, then write it that way, but don't try to spread crap about the BSD license.

Re:It's Funny. Laugh. (1)

nagora (177841) | more than 7 years ago | (#17618072)

Call a spade a spade. It's not a 'novel interpretation', it's FUD.

It's neither. The issue is that the BSD licence must remain in modified BSD code even if not released - no one argues about that, do they?. If someone then "steals" the code and puts it on Usenet, is there any legal recourse to preventing that code from being redistributed? Simple answer is: no. The BSD licence that is in the code allows anyone to distribute it regardless of how they got it, so long as the BSD conditions are met, which is easy.

The BSD license is viral and made so by the requirement that its conditions are copied into modified versions. Which was hardly a secret, now, was it? What's the fud/novel interpretation?

TWW

Re:It's Funny. Laugh. (1)

LurkerXXX (667952) | more than 7 years ago | (#17618384)

If you would bother to read the FUD article, the author puts for that modieied BSD code may not be relicensed under other licenses. That's the FUD interpretation.

Including an attribution/disclaimer is NOT viral under the generally held meaning of viral. All changes don't have to be given back to the community. That's the viral bit that some folks complain about. Please point me to a reference, other than this FUD author, which complains that an attribution/disclaimer is 'viral' in any sense.

Re:It's Funny. Laugh. (1)

Chandon Seldon (43083) | more than 7 years ago | (#17618634)

Since the code is still marked with the BSD license, and the BSD license is a copyright license, you can treat any code with the BSD license text on it as if it were licensed *solely* under the BSD license by simply ignoring any other license - you have permission to treat it as BSD licensed code, so the other license doesn't matter - it's not letting you do anything new.

The only question is: If the BSD license text is distributed with a program, does it apply to the whole program?

Re:It's Funny. Laugh. (1, Informative)

Anonymous Coward | more than 7 years ago | (#17618894)

Um, no, it's not quite that cut and dried.
*If* the author of any new code using parts of BSD code in it decides to release code rather than just a binary, it is their responsibility to indicate that *parts* of the code fall under the BSD license and include the attribution/disclaimer accordingly. It is also their responsibility to state any more restrictive license they care to use for the code. You must abide by their license and can't ignore it simply because BSD code is included. I don't think you will find a court that agrees with you that you can ignore parts you don't like. You are free to try of course, it's your lawyer bill.

The only question is: If the BSD license text is distributed with a program, does it apply to the whole program?

There will likely be a license the program is presented to you as licensed under. There will also be a file indicating some code is contained inside with the BSD attribution/disclaimer. The software author should make it clear that this is for only part of the code. Once again, I don't think there is any legitimate question of the BSD license applying to a program licensed under another license, but you are free to waste the courts time and pay a lawyer a lot of money for nothing if you want.

Re:It's Funny. Laugh. (2, Insightful)

nagora (177841) | more than 7 years ago | (#17619772)

Including an attribution/disclaimer is NOT viral under the generally held meaning of viral. All changes don't have to be given back to the community.

True, but the BSD license goes further than the disclaimer and attribution: "Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.". That's the viral bit. Viral has nothing to do with giving back to the community, it is simply a question of whether the license can be removed by someone that modifies the code. BSD can't be. End of story. The interesting question here is what the implications are IF the code does find its way into "the community". What effect does that have on any copyright claims?

It's not FUD if it's true. Law != Common Sense (0)

Anonymous Coward | more than 7 years ago | (#17618500)

> Call a spade a spade. It's not a 'novel interpretation', it's FUD.

And it's something a lawyer could present to the court in Australia. And that court might believe them.

Now, you can piss and moan all you want about how it's wrong according to common sense, but can you please tell me exactly *when* courts have operated according to common sense in modern times? Perhaps when they thought that posting a URL was the same as copying a work? Or perhaps when tomatoes became vegetables, even though botanists know they're actually fruit?

Perhaps you don't think that good lawyers can get very far on nothing but a crazy, distorted legal argument, like SCO vs. IBM that's only on, what, the third year now? Even though any normal person would've dismissed it as stuff and nonsense (but mostly the latter) after about 5 minutes.

You can cry "FUD!" until you're blue in the face, but it's moot unless the courts agree with you. Perhaps the BSD license could use some fixing?

Re:It's Funny. Laugh. (0)

Anonymous Coward | more than 7 years ago | (#17619034)

Actually, making BSD license be more like GPL in this sense wouldn't be "viral" like the GPL is. The nice thing about BSD is that using a few of their routines does not make the reset of the source code BSD. 10 lines in 100MB of source code can result in either 10 lines of BSD licensed code, or 100MB of GPL licensed code. It's contrary to the principles of open source to require everyone who isn't a friend to do their own implementations from scratch.

The idea of "sharing source code" is exemplified when I can take the BSD version of the internationalization library and use it in my embedded system. That library was "reinvented" from the GPL version instead of being a port. If there had only been the GPL version, I would have been required to reinvent the library since the FSF will not "share" this code with me without irreconcilable restrictions. I also get to use a BSD licensed C run time library to go with GCC, rather than have to write my own from scratch.

I am not a "software horder". I have no problems at all showing the source code that I write to others. But other software linked into the system involves trade secrets, and there are laws and regulations that forbid the end user from modifying the software (for health and safety reasons, etc).

Furthermore... (0)

Anonymous Coward | more than 7 years ago | (#17618522)

I agree 100% with you, and I would like to add that even if it were "viral" it may not be considered permanently "viral", the offending code could probably be removed, followed by a "re-licensing" of your own code.

Then you are battling copyright law vs. a twisted and weakly worded contract. I'll bet on copyright law.

And to even get to my argument, the law would have to recognize the article's point. Which isn't very likely [slashdot.org] !

IANAL of course.

Programmers, sick of legal bickering? (3, Interesting)

baadger (764884) | more than 7 years ago | (#17617216)

...personally I'd rather declare any code I produce to belong to the Public Domain or just keep it entirely closed and private. The way law is makes anything else a headache. Seriously, if even lawyers can't agree on anything why does anyone in the Open Source community even bother? Personally I care more about other people being able to benefit from my code than preventing corporations from using it for profit.

SQLite [sqlite.org] is released to the public domain and it's some damn fine code.

Re:Programmers, sick of legal bickering? (1)

99BottlesOfBeerInMyF (813746) | more than 7 years ago | (#17617402)

Personally I care more about other people being able to benefit from my code than preventing corporations from using it for profit.

Most OpenSource licenses aren't about stopping corporations from profiting and many even encourage that. They are also not about giving away code. Most are about making a trade. You can have my code if I can get credit, or use any code you write to improve it. If I'm writing some code either personally or for my company my motivation for licensing it with an OS license is to get free improvements to it.

Yes, Sick of this Shit. (4, Insightful)

twitter (104583) | more than 7 years ago | (#17617604)

Personally I care more about other people being able to benefit from my code than preventing corporations from using it for profit.

The problem comes when a company claims "ownership" of your code and then determines who benefits and under what contitions. That's what happens when you don't worry enough to make things right.

A great example of such a theft is Macsyma [wikipedia.org] (tediously detailed article that's nice but misses the point), the grand-daddy of Maple, MathCAD, Mathematica and many other symbolic algebra systems. It was developed, largely at public expense by people who expected the public to be able to have it. Instead, the results were "commercialized" in the 80's. A single copy of the original code [sourceforge.net] (much better history, as you would expect from a free software project) survived thanks to the efforts of Bill Schelter [wikipedia.org] , a GNU Common Lisp author and one of the first to port GCC to i386. Schelter managed to convince the DOE to let him legally distribute that code ... 20 years after it had been stolen from the public. Since then, development has been speedy and it will not be long before the quality matches or exceeds current commercial packages. The next time you spend a hundred bucks on one of it's commercial derivatives, remember that you might have had a free version a decade ago.

So, before you freely give your life's effort to others, you might consider what they will really do to other people with it and chose an explicit license that suits your real tastes. The GPL is the most common choice made and there's a reason for that. The same old assholes are up to new tricks, like "trusted computing" that are designed to lock everyone but themselves out of the market. In the future, if they have their way, you will not be able to run your code on commercial hardware. Is that the kind of thing you want to support in any way?

Re:Yes, Sick of this Shit. (0)

Anonymous Coward | more than 7 years ago | (#17617930)

That's a rather singular argument. In the US, Macsyma would have come before the current law on the books, which only offers true copyright protection to those who register their work.

Register your work from time to time is a smart thing to do anyways--this requires a fee, but I bleieve gives proof of date and the work to the US Copyright office.

Then release it into the public domain.

Public domain work, under current case law, does not seem in danger of being then made copyrightable. There was a /. article around the time the movie League of Extradinary Gentlemen was released, discussing this; some group (I think Newscorp) tried to claim whole ownership over fictional characters that had made it into the public domain but which they then modified or adapted into creative works; the court shot the corporation down.

Re:Yes, Sick of this Shit. (2, Insightful)

Ded Bob (67043) | more than 7 years ago | (#17617966)

The GPL is the most common choice made and there's a reason for that.

Maybe I misunderstood the history of Macsyma, but it sounds like the GPL would never have helped since the original code was not public in the first place.

I believe all government source should be public domain or MIT-licensed for all to use. No particular party should have control of it within the bounds of the government. If a company wants to commercialize it or others want to GPL it, that is fine. Since everyone from companies to individuals pay taxes for its development, everyone should have full access to develop and use it in any way.

How unfortunate (-1, Troll)

Anonymous Coward | more than 7 years ago | (#17618448)

I think it's unfortunate (and detrimental to Slashdot as a whole) that people can troll [slashdot.org] and flamebait [slashdot.org] in one story and then get modded up in another. Moderators should look at the person, and not necessarily alway the post.

Re:Yes, Sick of this Shit. (0)

Anonymous Coward | more than 7 years ago | (#17618490)

Macsyma wasn't "stolen". No one ever took the code and then told all the other owners of other copies that they weren't allowed to use it. All that happened is that some people took a good idea and ran with it, while the older versions languished. It's also clear that lots and lots of additional source code was added to the kernel of Macsyma's ideas to improve it. If the commercial organizations had not done this, it's possible the software would still have languished and we'd be missing a lot of vital tools today. If Macsyma had been on the GPL, there certainly would not have been something like Mathematica for a very long time.

If someone wants to put software in the public domain, it's their own damned business. It's silly for some GPL advocate to tell them that they're misguided just because some people might make money off the code. Why this irrational fear over people making money or applying different licenses? The goal of open source software should be to make software source code widely available to anyone, not to restrict it only to a group of friends and people with the same political bent, and not to try and get brownie points with the coder community.

Re:Programmers, sick of legal bickering? (0)

Anonymous Coward | more than 7 years ago | (#17617626)

My AC thoughts on the SQLite license. [slashdot.org]

In a perfect world it would be the perfect license; alas...

Re:Programmers, sick of legal bickering? (1)

mmurphy000 (556983) | more than 7 years ago | (#17618172)

personally I'd rather declare any code I produce to belong to the Public Domain or just keep it entirely closed and private

IANAL, but I worked with a number of attorneys on OSS licensing between 2000 and 2003. Of course, that was a few years ago, so my recollections may be fuzzy.

A typical public domain declaration (e.g., "dedicate to the public domain any and all copyright interest...") does not disclaim warranties, meaning you may be "on the hook" if your code is used and problems arise. Implicit warranties don't stem from copyright law, but rather commercial law (e.g., the Uniform Commercial Code in the US), so public-domain provisions won't automatically disclaim any implicit warranties (e.g., fitness for use, merchantability).

I seem to recall there was also some debate over whether in the US it was even possible for ordinary folk to create public domain works, given the language in the Copyright Act of 1976 that automatically applies copyrights to every work.

If you are going to try the public domain route, please try to find some language for making that grant that's been reviewed by an attorney, versus just rolling your own.

Arcane (5, Interesting)

radtea (464814) | more than 7 years ago | (#17617254)


They seem to be saying this:

1) The BSD license clause 3 says the FOLLOWING conditions must apply.

2) They wonder if "apply" means "apply" or something else, like "apple" or "penguin".

3) They note that one of the FOLLOWING conditions is the warranty.

4) They wonder if one of the PRECEDING conditions (clause 2) ought to be handled the same way as the the warranty.

In a narrowly construed legal sense they may have a point.

In a human being sense, if anyone has ever wondered why we all hate lawyers and think they are wankers, this is pretty much it.

It is, of course, impossible to create an unambigous document, and yet lawyers pretend to be able to do this, and then make a fortune out of their failures.

No one, ever, anywhere, has ever had any question as to what the BSD license means. So clearly there is a valid and correct reading that means what everyone knows it to mean. So clearly any reading that completely reverses that meaning must be making a mistake somewhere.

This post, by the way, can be interpreted as a love sonnet addressed to a musk ox, if you look at it closely enough and make up the meaning of a sufficiently large number of words, and wonder when I say, "it is, of course, impossible to create an unambigous document" if I really mean, "misey were the borogoves, and the momrath outrabe."

Re:Arcane (1)

Subliminalbits (998434) | more than 7 years ago | (#17617420)

Its a shame I don't have mod points, because that truly was hilarious.

Re:Arcane (2, Informative)

75th Trombone (581309) | more than 7 years ago | (#17617594)

Close, but no cigar:

'Twas brillig, and the slithy toves
Did gyre and gimble in the wabe:
All mimsy were the borogoves,
And the mome raths outgrabe.

Re:Arcane (2, Informative)

nuzak (959558) | more than 7 years ago | (#17617718)

From this typing mistake, we can conclude that the grandparent in fact engages in improper relations with rutabaga.

No your honor, I have not been drinking.

Re:Arcane (1)

gardyloo (512791) | more than 7 years ago | (#17618962)

From this typing mistake, we can conclude that the grandparent in fact engages in improper relations with rutabaga.

      Ah, if by this you mean that the forglewhoops have been dabbling in string theory again, and that Kansas will outlaw such frippery if paid enough by Gill Bates, I totally agree!

Re:Arcane (1)

radtea (464814) | more than 7 years ago | (#17619420)

Gah! I'm a terrible speller and try to proof my posts. But not, apparently, my subject lines.

Re:Arcane (1)

NoOneInParticular (221808) | more than 7 years ago | (#17617752)

Quite a coincidence. Today, I was reading the subversion license (acknowledged as BSD style) to see if we could incorporate it in some add-on of our products. We need version control, you see. I read the text, and what was immediately obvious was the provision that next to the copyright 'this list of conditions' needed to be included. My common sense told me that if we distributed that software, tightly integrated with out product, what would 'this list of conditions' apply to? The subversion part of the code, the subversion + integration part, or our entire binary? How should we make clear that only the subversion part of the code is covered by the condition that the binary can be freely distributed? I am not a lawyer, but I saw an issue here.

After looking around a bit I found some web-pages that told me that Subversion has a BSD/Apache style license, and I was momentarily satisfied, thinking by myself: "I read so much about the BSD license, this should be ok. Just make sure that the corporate lawyers take a look before we go that route." I also read the BSD license itself and noted that indeed, subversion is a BSD (not Apache) style license.

As I was satisfied by the 'common wisdom' around the BSD-license I didn't really pursue this, but my first instinct was to be very wary of the license. Thus, to quote you:

No one, ever, anywhere, has ever had any question as to what the BSD license means. So clearly there is a valid and correct reading that means what everyone knows it to mean. So clearly any reading that completely reverses that meaning must be making a mistake somewhere.

I had questions, today, before I read this article, right after reading a BSD-style license, before I knew it was BSD-style. I was stupid enough to be satisfied when I saw the label 'BSD' applied to the license, assuming that that should be okay. Okay not because the text was okay, but because everybody said it was okay. And then this article. They see the same thing, and brutally kill the point by drowning it in legalese. The point is there though.

Now my question for you is: consider a situation where you create closed-source software, you want to incorporate some BSD code, and you read the BSD-license. How would you formulate the license of your code in such a way that (a) contains the BSD-license and (b) it only applies to the BSD part of the code and not to the code you wrote yourself, which EULA style. To make it interesting: you're distributing it as a single binary.

For reference, here are the three clauses of a current BSD (source wikipedia). Read carefully:

* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
* * Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of the <organization> nor the
* names of its contributors may be used to endorse or promote products
* derived from this software without specific prior written permission.

Re:Arcane (1)

zotz (3951) | more than 7 years ago | (#17617788)

[This post, by the way, can be interpreted as a love sonnet addressed to a musk ox, if you look at it closely enough and make up the meaning of a sufficiently large number of words, and wonder when I say, "it is, of course, impossible to create an unambigous document" if I really mean, "misey were the borogoves, and the momrath outrabe."]

Gee, and until I got here I was under the impression that the subject had to be 'Ode to a musk ox!' A sonnet you say? Are you certain?

By the way, you must be brillig!

all the best,

drew

Re:Arcane (1)

hotdiggitydawg (881316) | more than 7 years ago | (#17618242)

This post, by the way, can be interpreted as a love sonnet addressed to a musk ox
By the way, you must be brillig!
If by "brillig" you mean "Canadian", I could well believe it...

Re:Arcane (1)

Sancho (17056) | more than 7 years ago | (#17617816)

The common interpretation of the BSD license is just that--common. Of course, because legal mumbo-jumbo increases at a constant rate, it has failed to adapt. In other words, the original authors of the license probably never anticipated this amount of scrutiny.

In truth, they're largely right. The clause explicitly says that source-distributions must effectively contain a reproduction of the entire license. It does it by separating the license into three parts, then demanding that each part be included. It's rather inelegant, and is akin to typing:

newLicense = concat(oldLicense[0],oldLicense[1],oldLicense[2]).

Except for the original authors explaining their intent, I can't figure out how anyone would ever come to the conclusion that the 'misconceived' interpretation is correct.

Re:Arcane (0)

Anonymous Coward | more than 7 years ago | (#17619148)

We obviously need a BSD License Version 3! Death to TiVo revisionists!

Re:Arcane (1)

QuantumG (50515) | more than 7 years ago | (#17617892)

Uhhh, you're on crack. "Following" is only in the BSD license in one place:

Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:

[then the conditions]

[then the warranty disclaimer]

Re:Arcane (1)

McDutchie (151611) | more than 7 years ago | (#17618182)

In a human being sense, if anyone has ever wondered why we all hate lawyers and think they are wankers, this is pretty much it.

You'd think geeks, who have this affinity with coding, would like attempts to define language precisely, since this is what programming languages do.

Richard Stallman understood that writing legalese is like coding: he compared writing the GPL with writing a program. It's a damn good hack, too.

No, we don't (0)

Anonymous Coward | more than 7 years ago | (#17618556)

The whole reason programming languages exist is because we understand our everyday languages are not defined well enough. By creating a language that is fixed and closed to interpretation we save ourselves a lot of headaches. If we are confused what code Does, we just run it. argument over. If you want a comparison of nerds who like language, I would have thought you'd have mentioned Larry wall inventor of perl and a linguist. Considering the author's skill set, I think the legibility of perl code is no coincidence.

Re:Arcane (0)

Anonymous Coward | more than 7 years ago | (#17618330)

They seem to be saying this:
1) The BSD license clause 3 says the FOLLOWING conditions must apply.
Nope: It says "Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer."
2) They wonder if "apply" means "apply" or something else, like "apple" or "penguin".
and come to the conclusion that it means "apply" and not something else as everyone else seems to read into.
3) They note that one of the FOLLOWING conditions is the warranty.
Which therefore ought to apply.
4) They wonder if one of the PRECEDING conditions (clause 2) ought to be handled the same way as the the warranty.
Which obviously is included in clause 3 because it says "this list of conditions".
What was your point again?

Re:Arcane (1)

RAMMS+EIN (578166) | more than 7 years ago | (#17619282)

``1) The BSD license clause 3 says the FOLLOWING conditions must apply.''

Not in my copy of it:


Copyright (c) The Regents of the University of California.
All rights reserved.

Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
are met:
1. Redistributions of source code must retain the above copyright
      notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright
      notice, this list of conditions and the following disclaimer in the
      documentation and/or other materials provided with the distribution.
3. Neither the name of the University nor the names of its contributors
      may be used to endorse or promote products derived from this software
      without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
SUCH DAMAGE.


The part that states "provided the following conditions are met" comes _before_ any of the clauses. The conditions properly state that (1) the copyright notice, (2) the list of conditions, and (3) the disclaimer. I see no unclarity here.

Disclaimer: IANAL.

Uhhh.. a common misconception by idiots maybe.. (4, Informative)

QuantumG (50515) | more than 7 years ago | (#17617362)

When Microsoft or some other proprietary software company that wants to use BSD licensed code, and actually has lawyers on payroll, decide on the wording for their license, it always reads like this:

Copyright (c) 2003-2007 Microsoft Corporation.
All rights reserved.

[Copy of the EULA goes here]

This software contains components from XXX which are available under this license:

[Copy of the BSD license goes here]


So they are not relicensing the BSD licensed components. They are providing those parts of the software under the license of which they were required and they are doing all they are required to use that code by providing the license in the documentation. The power of this is that the BSD license doesn't require the source code to be released to the user (and Brendan Scott, the author of the paper, recognises this in section 7.3) so the company can keep their modifications secret.

Re:Uhhh.. a common misconception by idiots maybe.. (1)

Chandon Seldon (43083) | more than 7 years ago | (#17618706)

That looks like it's clear and valid, so that solves it. On the other hand, if you just put the BSD license into the documentation without the "his software contains components from XXX which are available under this license" line, the whole thing is probably BSD.

We should all LOL at this conclusion (2, Insightful)

Secret Rabbit (914973) | more than 7 years ago | (#17617450)

Personally, I think that this is just an attempt to make the BSD license somewhat equivalent to the GPL just so that people will use the GPL more. After all, we all know that the GNU Foundation has gotten really active with regards to activism:

http://linux.slashdot.org/linux/06/12/31/010221.sh tml [slashdot.org]

And the BSD/MIT licenses are the GPL's nearest competitor according to a poll here at /. (yes, I know that polls here aren't exactly accurate, but it does provide a indicator).

http://slashdot.org/pollBooth.pl?qid=1364&aid=-1 [slashdot.org]

All this is is twisted lawyers, writing twisted conclusions based on twisted sophistry. It is nonsense.

Re:We should all LOL at this conclusion (2, Insightful)

GigsVT (208848) | more than 7 years ago | (#17618104)

In case you haven't notice, GPL licensed code is way more widely used than BSD licensed code.

You could equally say that by making the BSD sound GPL like, it's an attempt to show people that the BSD license is just as good as the GPL at protecting the rights of the people receiving the software.

Re:We should all LOL at this conclusion (1)

Pink Tinkletini (978889) | more than 7 years ago | (#17619344)

"In case you haven't notice, GPL licensed code is way more widely used than BSD licensed code."

Prove it. From where I stand, it seems that in terms of sheer volume, BSD-licensed code is much more prominent in daily life than GPL-licensed code. You've got the Windows networking stack and the userland tools in OS X for starters, and OS X alone is much more widely used than Linux.

The best part of this paper is this... (1)

QuantumG (50515) | more than 7 years ago | (#17617778)


8.1
    (e) what is the difference between a "modification" and a "derivative work"? If they are the same, the scope of the BSD's licensing requirement will be very similar to that of the GPL. Note also that the BSD only permits the distribution of modifications, so if there exist derivative works which are not modifications the BSD does not address whether they can be distributed or even created - and in the world of licensing that is the same as a prohibition. Compare the GPL which expressly (but conditionally) permits the creation and distribution of derivative works;


So suppose, Microsoft takes some BSD licensed code and incorporates it into Internet Explorer. They put in the documentation "this program contains software which is BSD licensed, here's the license", etc. Have they done all they are required to do? No. Either Internet Explorer is a modification of this BSD licensed code, in which case, it much be distributed under the BSD license.. or Internet Explorer is a derivative work of this BSD licensed code, and a derivative work is something different to a "modification" in which case it cannot be distributed *at all*.

Re:The best part of this paper is this... (0)

Anonymous Coward | more than 7 years ago | (#17618066)

Let's assume that IE is BSD-licensed. So what? The BSD license doesn't require source code distribution, it merely allows for it.

Thus all you would get from it would be the ability to copy it. To fix that all they'd have to do is make you enter a serial number into your browser.

Re:The best part of this paper is this... (1)

Aladrin (926209) | more than 7 years ago | (#17618120)

Nothing says the modified code has to be distributed UNDER the license. Only that the copyright notice, conditions, and disclaimer in the license are retained. Nothing prevents you from adding more copyright notices, conditions and disclaimers.

Re:The best part of this paper is this... (1)

QuantumG (50515) | more than 7 years ago | (#17618350)

hehe. Nothing permits you to add more copyright notices, conditions and disclaimers. That's the way copyright law works. Anything that is not permitted is prohibited.

Re:The best part of this paper is this... (1)

Aladrin (926209) | more than 7 years ago | (#17618544)

It's implicit. You can redistribute as long as you follow the conditions stated. Nothing in the conditions prevents you from adding other conditions.

Here's a test:

Can you redistribute BSD licensed code while sitting? Yes.

"Nothing permits you to redistribute BSD licensed code while sitting. That's the way copyright law works. Anything that is not permitted is prohibited." - Wrong.

While drunk? Yes.
While at home? Yes.
While at work? Yes.
While...

Permission was given to anyone who followed certain conditions. Anything else not stated does not matter.

This goes for any software license. It outlines the exact conditions you must follow in order to do X. If you follow those, even if you're doing a billion other things, then you can do X.

Re:The best part of this paper is this... (1)

QuantumG (50515) | more than 7 years ago | (#17618756)

Yeah no. I'm making a legal argument. You're making a nonsense argument. You are not permitted to make a derived work from a copyrighted work without permission. The BSD license doesn't give you permission. That's it. Of course, the *point* of making something BSD licensed is that you don't give a shit about what people do with the code.. you don't want to be sued.. so in 99% of cases you won't run into trouble.

Re:The best part of this paper is this... (1)

jschultz410 (583092) | more than 7 years ago | (#17619004)

Am I taking crazy pills? If you comply with the BSD license then it obviously gives you permission to create derivative works or even minor modifications.

"[2] Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:"

So long as you meet the conditions, which are to (a) reproduce the copyright notice, (b) the list of conditions and (c) the warranty disclaimer, then the license gives you permission to create modifications, derivative works, etc.

The problem is that the author has incorrectly included the original license grant into the list of conditions that must be reproduced by redistributors / modifiers / etc. A redistributor / modifier is not required to reproduce [2] in their redistribution / modification. See this post [slashdot.org] for more detail. Therefore, they are not required to allow modification, redistribution, etc. of their derivative work.

Summary (1)

nagora (177841) | more than 7 years ago | (#17617852)

So, what they seem to be arguing is that the BSD license says that the modified source is covered by the BSD license (but does not require disclosure of that source) which itself allows redistribution under simple conditions. Therefore, if previously undisclosed modified source is leaked it may not be possible to sue anyone for breach of copyright nor to gain an injunction to prevent further copying and use of that source since the source would include permission to distribute it.

In Australia, at least.

Seems reasonable to me.

TWW

Wow, that's shockingly stupid. (2, Informative)

Generic Player (1014797) | more than 7 years ago | (#17617980)

He doesn't even provide anything to try to back up his crazy claim, he just keeps repeating his conclusion that "you must distribute it under the BSD license" under all these circumstances. The closest thing he gets is claiming that the BSD license doesn't explicitly permit re-licensing. But copyright doesn't have anything to do with licensing, so its totally irrelivant. The BSD license grants you the copyright granted rights that normally are reserved for the author, if you obey the terms. Applying your own license terms later on has nothing to do with copyright, and hence the author of the software has no right to stop you. The GPL prevents you from using your own license not because of some magical "you can't relicense" part of copyright law, but because the GPL explicitly says you can't in the license.

Unofficial Word From Someone at UC Berkeley (2, Informative)

CyberLife (63954) | more than 7 years ago | (#17617988)

I just spoke to someone I know in UC Berkeley administration about this situation and they told me the following. Please note that I am paraphrasing here and none of this is to be taken as an official statement by the University of California.

The spirit of the license is exactly as people have interpreted it. It is not intended to limit or hinder people in any way. On the contrary, it is fully intended that their products be freely used, modified, and distributed. That's what academic research is all about. Berkeley has neither the time nor energy nor desire to chase people down. They just want credit for doing the work.

In addition, most of Berkeley's projects are government-funded. As such, they are not generally permitted to make any profit from the work. It has to be made public and people have to be allowed to extend it for their own purposes. The essence of public research is to benefit society as a whole, not just the corporate sector.

As for the question of third-party derivative works being used to make a profit, there is nothing stated in the license to prohibit such acts. Thus, it would seem to be legal. However, it could be argued that doing so is against the spirit of the license. Whether or not Berkeley could enforce that spirit in a court of law (assuming they even care to do so) is another matter.

If anybody wants an official statement, they should contact Berkeley's legal department.

mod do3n (-1, Offtopic)

Anonymous Coward | more than 7 years ago | (#17618064)

reap3r Nor do the for it. I don't

*sigh* (3, Insightful)

w3woody (44457) | more than 7 years ago | (#17618284)

Look, when I release software under a BSD-like license, my intent as the owner of the work is to do the following:

(1) Permit people to do whatever they want with the software--including relicensing the software, so long as
(2) if you use my software, you don't then plaster my name all over it as if I endorce whatever cause or crappy software you're creating, and
(3) you don't sue my ass if and when the software you downloaded from me breaks.

Basically, do what you want--just leave me out of it.

In one sense the article is correct: in imposing a new license you cannot remove the old one. But as the intent of the old license is to cover my ass and keep my name around so people know what sort of a cool dude I am, so long as the new license also covers my ass and keeps my name around so people know what sort of a cool dude I am, I don't see the problem--either from a common-sense perspective or from a legal one.

Re:*sigh* (1)

nagora (177841) | more than 7 years ago | (#17619850)

Look, when I release software under a BSD-like license, my intent as the owner of the work is to do the following:

A court will not care a jot what you intended, it will look at the specific words you used. That's why it's important to read the license instead of assuming that it says what you mean.

FUD (1)

mccoma (64578) | more than 7 years ago | (#17618300)

Somehow, I have more faith in Berkley's lawyer's interpretation of US Copyright Law than a random article's flaky conclusions.

This line of argumentation is specious ... (5, Informative)

jschultz410 (583092) | more than 7 years ago | (#17618656)

It seems that this lawyer has not been trained in computer science because he is glossing over an important detail of the license to reach his incorrect conclusion. The BSD license says (using the author's numbering and my emphasis):

"2 Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

3 * Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

4 * Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

5 * Neither the name of the [organization] nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission."

The license requires, per clauses 2-4, that a user reproduce (a) the copyright notice, (b) the list of conditions and (c) the disclaimer of the original license. The author reads this as requiring that the entire BSD license be reproduced in any redistrubtion or use of the code. But this is only true if (a), (b) and (c) comprise the entirety of the BSD license. I argue that they do not!

The key question is, "What is the 'list of conditions' that must be reproduced?" The author incorrectly claims that [2] is part of the list of conditions that must be maintained by a user, which would create a viral mechanism that this paper describes.

It is obvious from [2]'s use of the phrase "the following conditions" and the fact that [3, 4, 5] are preceded by asterisks and use the phrase "this list of conditions" that [2] is not intended to be part of the list of conditions. The list of conditions only consists of [3, 4, 5]. Therefore, redistributors / users are not required to maintain the original grant of the license [2] in their use or redistribution of the code.

The flaw in the author's argument is that he is incorrectly including the original grant of the license [2] into the list of conditions [3, 4, 5]. The license truly only requires that users reproduce clauses [1, 3, 4, 5, 6] of the BSD license in their redistrubtions or use.

Fundamental error (2, Insightful)

nsayer (86181) | more than 7 years ago | (#17618718)

My analysis of the paper is that the author confuses what is permitted with what is required.

Let's take a look:

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
So long as you include the disclaimer as required, and you don't use the author's name in vain, you can do as you like.

There is no part of the license that says that you cannot distribute modified forms under more restrictive licensing, provided that you also perform the acts required by the BSD license and require the same of any sub-licensees that distribute. There is no part of the license that says that you cannot distirbute UNmodified forms under more restrictive licensing, but presumably anyone receiving a copy from you under more restrictive licensing could figure out that you obtained it from a source that merely required adherence to the BSD license, throw away the copy you provided and get their own.

In general, acts that are not specified as prohibited in a contract (in this case the license is a contractural term. You are agreeing to abide by the license in return for being provided value in the form of the code covered by the license) are permitted (modulo exceptions that aren't worthy of mention here). Since sublicensing is not mentioned, it is permitted - provided the original conditions are always met by anyone redistributing and/or using the code.

Shout it down! (1)

John Nowak (872479) | more than 7 years ago | (#17618808)

This is a very serious issue, not because there's any validity to what is being claimed here, but precisely because likely there isn't. BSD-style licenses are incredibly useful and provide a good, solid, simple alternative to the GPL. The common interpretation clearly matches the intended spirit of the license.

FUD like this, and this is the first time I'm even using that acronym, is dangerous and puts the work of thousands of people at risk. This should be shouted down immediately, even if there is any shred of validity to it, which I doubt there is.

Re:Shout it down! (2, Funny)

Random BedHead Ed (602081) | more than 7 years ago | (#17619488)

Even better than shouting down potentially valid arguments that are nevertheless dangerous ideas, let's all plug our ears and LALALALALALALALALALALALALALALALALALALALALALALALA!! !!!!!!

He's half right (1)

Brandybuck (704397) | more than 7 years ago | (#17619090)

He's half right. You cannot take my BSD licensed software and relicense it under the GPL. However, you are perfectly free to create your own derivative work and license it under the GPL. The differences between the two are slight, but they are there. You cannot take my software and file off the BSD license. But you are able to fork off my software and license that any way you want. But in order to do the latter, it has to be a derivative work and not a mere copy. IFAIK, mere translation via compilation doesn't count either.

It's funny how some GPL advocates seem obsessed on relicensing your code under the GPL. It's like telling them everything in the fridge is free for them to eat, except for the sandwich you made for tomorrow's lunch. Then you find out the next morning that the only thing they ate was the sandwich.

The only way this will be tested... (1)

SETIGuy (33768) | more than 7 years ago | (#17619260)

The only way this will be tested is if someone who has released source code under the BSD license files suit against someone who is redistributing it under other terms. I don't think that's going to happen any time soon. Until then, it's just speculation...

So Just Fix the License (0)

Anonymous Coward | more than 7 years ago | (#17619338)

Since this is contrary to the established tradition, it seems like the *BSD teams would just modify the license to pull it back in line, and retroactively re-license all the releases. Most other people who used the BSD license would probably do likewise. FWIW, I tend to keep the code separate for sanity's sake anyway, and use it like LGPL. I've never found a bug in something and not given it back to the community--that'd just be silly. In other words, it's a tempest in a teapot, nothing to see here, move along, etc...

MIT License (1)

RAMMS+EIN (578166) | more than 7 years ago | (#17619424)

I wonder if the same argument (provided it holds any water) applies to the MIT license, which I use extensively. It seems that there is a difference: where the BSD license states


Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
are met:
1. Redistributions of source code must retain the above copyright
      notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright
      notice, this list of conditions and the following disclaimer in the
      documentation and/or other materials provided with the distribution.
3. Neither the name of the University nor the names of its contributors
      may be used to endorse or promote products derived from this software
      without specific prior written permission.


the MIT license states:


Permission is hereby granted, free of charge, to any person obtaining a
copy of this software and associated documentation files (the
"Software"), to deal in the Software without restriction, including
without limitation the rights to use, copy, modify, merge, publish,
distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to
the following conditions:

The above copyright notice and this permission notice shall be included
in all copies or substantial portions of the Software.


If you read carefully, you will see that the BSD license requires the disclaimer following the conditions to be included redistributions of the software, but the MIT license imposes no such requirement! And here I was thinking the MIT license and the (revised) BSD license were equivalent!
Load More Comments
Slashdot Login

Need an Account?

Forgot your password?
or Connect with...

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>