Slashdot Mirror
MIT Media Lab Defaults To Free and Open Source Software (networkworld.com)
Reader alphadogg cites a report on NetworkWorld: MIT Media Lab, that 30-year-old tech innovation factory that has had a huge hand in churning out everything from LEGO MindStorms to the Guitar Hero video game, has now wowed the open source and free software crowd. Lab Director Joi Ito over the weekend revealed that MIT Media Lab has changed its approach to software releases to FLOSS (free/libre/open-source software) by default.
Frankly, that's what those guys are good at. I don't really care if they release their Actionscript, etc code or not.
I just bought some new floss last week. It's just string with wax on it. Not complicated at all... so why do they need a conceptual map of it anyway? Does Colgate have a giant CAD blueprint of said floss?
The default does not have to be enforced. I am pretty sure that the really good and useful stuff does not get treated as FLOSS. Nothing more than a publicity stunt here, folks. Research should anyways be public so what the hell.
The GPL removes no freedom at all. It only doesn't grant it. That means, if you really want to distribute closed source software based on GPL, you only need the permission of the copyright holders. The GPL does grant freedoms, the only freedoms it doesn't grant is the freedom to remove freedom. This isn't granted by states either (you may not imprison people, but the state can).
I don't say the GPL is perfect, but for huge monolithic software it is a good model.
What the fuck is that "Conceptual Map of the FLOSS" image at the top of the announcement?!
I don't see why the GPL is in the "Free Licences" block alongside the BSD license. The BSD license deserves to be there, but the GPL does not. The GPL goes out of its way to remove freedom and choice (like the freedom to modify the code and to distribute binaries without revealing the changes, or the freedom to use a snippet of GPLed code without virally tainting the rest of the code base), while the BSD license tries to maximize it.
And why isn't the MIT license listed?!
What a fucking weird diagram.
Every time someone says this sort of thing, I have a question. This question has never received a good answer, mostly hand-waving. The question is:
If the BSD license is superior to the GPL, why didn't Open Source really truly take off as a movement with mainstream recognition until after the Linux system and its GPL license? Linux is something lots of average non-techie users have heard of and some of them are using it.
I believe the answer is that the GPL works better in the real world. It achieves a better trade-off of objectives. Isn't this normally how things work? The blind idealism usually is outperformed by pragmatic objectives. So if the BSD license is superior why doesn't it achieve superior results?
MIT Media Lab Defaults To Free and Open Source Software
That's not what the article says at all. They've removed the extra approval required to open source projects. They are open sourcing everything by default. It's up to the project whether they release it as proprietary or open source.
And also the freedom to nuke gay baby whales with weaponized peanut butter.
ALL THESE FREEDOMS ARE IMPORTANT, WAKE UP SHEEPLE!
Freedom can only be taken away, and that's exactly what the GPL does
If you don't like it, don't use it.
It denies freedoms that are otherwise perfectly reasonable, and would exist were the GPL not denying them.
Really? Cool!
Say, could you do me a solid and use all that freedom you have in the absence of the GPL to make a quick change to the Win10 source code to me? Just the teensiest tweak to a GUI default setting...
I'd do it myself, but, well... Virally tainted and all.
Thanks!
LOL, yup, that's freedom ... adhere to this long list of restrictions, or don't use it.
Whereas, something like a BSD or Apache license places very few limitations on your 'freedom".
GPL means you are "free" to be enslaved by a certain ideological view of "free", by telling you the ways in which you're not really free.
It has its place, but it's by no means perfect, and it certainly has a restricted definition of "free".
Lost at C:>. Found at C.
Partly because GPL had a louder salesman during a time when open source was taking off. Partly because it was applied in ignorance by people wanting to be a part of the open source community, through apathy or misguided intent.
First is not better. It's sad that RMS Toejam was among the top advocates for open source, but hey, at least he started the discussion.
Except the GPL doesn't exist in a vacuum - in the absence of the GPL (or some other license), copyright law automatically removes virtually all freedoms except to observe and discuss. No sharing, no modification, not even necessarily the right to use it (running software requires copying it into RAM, a point which has been raised in a few lawsuits, though I don't recall if there's ever been an official ruling on the subject).
The GPL then grants you many freedoms which you would not have had - granting the right to do almost anything with the software *except* put further restrictions on derivative works.
--- Most topics have many sides worth arguing, allow me to take one opposite you.
Whereas, something like a BSD or Apache license places very few limitations on your 'freedom".
Well, then it sounds like you should use a BSD or Apache license for your open source project. Nobody's forcing you to use GPL if you don't like it.
What I find particularly ironic is that they specifically call out GPL and BSD licenses in the article. Admittedly the article's author is using a free graphic.
I would expect that MIT would use the MIT license (https://opensource.org/licenses/MIT) over the BSD license (https://opensource.org/licenses/BSD-3-Clause). Maybe it is just me...
The GPL goes out of its way to remove freedom and choice (like the freedom to modify the code and to distribute binaries without revealing the changes, or the freedom to use a snippet of GPLed code without virally tainting the rest of the code base), while the BSD license tries to maximize it.
Let's start from the default:
- I write code, and you get NOTHING. ZIP, ZERO, ZILCH, NADA.
You distribute my code, I can sue you for copyright infringement.
You distribute binaries compiled from my code, I can sue you for copyright infringement.
You distribute binaries that I compiled from my code, I can sue you for copyright infringement.
You redistribute binaries that I compiled from my code and distributed to you, I can sue you for copyright infringement.
Now I get to make a choice.
1. BSD
- You can take my code, and profit from it. Fuck me, fuck the people who receive your binaries, fuck the world. You win.
2. GPL
- You can take my code, and profit from it. You can't fuck me, you can't fuck the people who receive your binaries, you can't fuck the world. We all win.
Well, that's an easy choice. I choose the GPL. ... your alternative is to WRITE IT YOUR GOD-DAMNED SELF, and then, feel free to do whatever you want with what YOU wrote.
If you take my code, you play nice, with me and everybody else.
If you don't like that
Sure, but if you're going to go around braying about just how "free" it is, you should also understand that to some people, GPL is somewhat less than "free". Because it's a restrictive kind of "free".
It's A model of "free", it's not THE model of "free". Even Linux is still GPLv2, precisely because it's a much more restrictive license. You know, kinda less "free".
As you have just inadvertently summed it up, the "freedom" the GPL offers is to use it under a very specific terms, or not use it at all. Which is kinda mostly free, except in the ways in which it isn't.
You're right, people are free to not use it. So, don't be surprised when people don't because they don't want someone else's ideology crammed up their ass.
Having benefited from BSD and Apache licenses, I'd be quite inclined to use those licenses, or at least things like the LPGL so you can link to stuff ... precisely because they don't wish to assert control over how others use it. You know, they're more "free".
Oh, by the way, in quoting my posting you created an unlicensed derivative work, and you owe me twelve trillion dollars. ;-)
Lost at C:>. Found at C.
You're absolutely right. Linux and the GPL were almost purely hype-driven movements. MySQL and PHP are other early examples. More recent examples of the same phenomenon are Ruby on Rails and Node.js.
These things became popular not because they were good in any meaningful way. Compared to the lesser-known alternatives, they were actually worse. FreeBSD was and is superior to Linux. The BSD and MIT licenses were and are better than the GPL. PostgreSQL was an is better than MySQL. Python was and is better than PHP. Django was and is better than Ruby on Rails. Twisted was and is better than Node.js.
Those who were smart enough to choose the better options never felt the need to blab about the software they were using. They used these better options, got their work done, and moved on.
On the other hand, those using the inferior options often had to beg for help on mailing lists and in bug trackers and in IRC and in other discussion forums, raising awareness of these worse technologies. While smart people would take this as a sign to not use these technologies or licenses, less intelligent people treated them as advertisements.
One of the key elements of hype is that it snowballs. Hype attracts more hype and the type of people who like to promote hype. A small squeak of hype can quickly become a deafening roar of hype.
We often see the same people riding the hype wave. Rust is a good example of this. It's quickly becoming one of the most-hyped programming languages, despite being inferior to C++11 and C++14 in almost every way. Something that's interesting to note is how a number of the Rust core team members were formerly big names within the Ruby, Ruby on Rails and JavaScript communities, but jumped ship to Rust when the Ruby and JavaScript hype started to die off.
Linux, the GPL, MySQL, PHP, Ruby on Rails and Node.js are accidents of marketing gone wrong, rather than examples of things that are technologically beneficial in the real world. Their success is due to hype and ignorance, not merit and understanding.
Mod this AC up. It's needed.
General Relativity: Space-time tells matter where to go; Matter tells space-time what shape to be.
Sure, but if you're going to go around braying about just how "free" it is, you should also understand that to some people, GPL is somewhat less than "free". Because it's a restrictive kind of "free".
Of course I understand. That was the whole point of the GPL. It gives you certain rights, but not the freedom to take the same rights away from others. Sounds like a fair trade to me, but there are plenty of other licenses to choose from if you don't like it.
LOL, yup, that's freedom ... adhere to this long list of restrictions, or don't use it.
Whereas, something like a BSD or Apache license places very few limitations on your 'freedom".
GPL means you are "free" to be enslaved by a certain ideological view of "free", by telling you the ways in which you're not really free.
It has its place, but it's by no means perfect, and it certainly has a restricted definition of "free".
Every time I see this philosophical argument, I shake my head, because the proponents of both licenses tend to completely ignore the viewpoint of the other and get bogged down in the definition of "freedom." BSD is free in the sense of anarchy. Here's our code, do with it what you will. "Fork it? Clone it and close it? Go for it, man, be free." There is a lot of good in this, and some bad. GPL is free in the sense of the picnic pavilion down at the local park. "We have some good rules in place here to make sure that everyone can freely enjoy this today, and in the future." There is a lot of good in this, and some bad.
I tend to come down on the GPL (v2, that is--v3 is a mess) side for "community projects" and the BSD side for pure utility projects. If you have a shared vision for something you really want to build than GPL's "you have to give back" philosophy is wonderful. You need look no further than the current state of Linux to see why. On the other hand, if you really just need a standard for everyone to adhere to so we can all go and get some other work done then BSD type licenses are almost always superior, because there is no problem incorporating them into closed source projects (which is great, because you want all those products to interop in some way).
tl;dr: both are good for some things, and bad for other things. Quit your bitching and pick the right one for your project, and pick the right project for your need.
What part of "shall not be infringed" is so hard to understand?
Does your workstation run Smalltalk on Multics over X25-based Infiniband, by any chance? ;-)
Mod this AC up. It's needed.
Yeah but my invincible sense of entitlement is impervious to reason. Other people owe me their work on my terms, for free! If they ask that I not reduce the freedoms of somebody else in return, well they're just a bunch of big meanie-heads!!
You can "use" (as in run) GPL licensed software for any purpose whatsoever. You just can't create proprietary closed source software with extracts of the source code provided to you. That's why the GPL promotes freedom, and the BSD license promotes proprietary software.
GPL does remove freedoms, though.
Take a piece of code under (modified 2 clause - assumed since 3-clause is incompatible) BSD. It gets incorporated under GPL (fine, that's why the authors picked BSD). Now, however, a patch is made to that code, which is GPL'd. That patch cannot be incorporated back into the upstream project because it's GPL'd which conflicts with the original project's BSD license.
It's not a hypothetical - it has happened before. GPL advocates say GPL protects against "code theft" by proprietary licenses "stealing" BSD code, when the GPL does exactly the same thing. Except instead of locking it away in a closed box, the GPL parades it around, proud it has taken BSD code and locked it up as if it was a proprietary license.
That's probably the most insulting part of GPL - the claim BSD allows locking up "with proprietary licenses" when in reality, the GPL does the same. SO yes, the GPL has removed freedoms from the code that was formerly BSD licensed.
If you have a legitimate need to use GPL, then use it. If you want to distribute software for the purpose of free use by others, don't use GPL.
We often see the same people riding the hype wave. Rust is a good example of this. It's quickly becoming one of the most-hyped programming languages, despite being inferior to C++11 and C++14 in almost every way.
If you can't write rust, the compilation will fail. If you can't write C++, the program will segfault, and perhaps only on some specific condition. C++ doesn't have unified unit testing, it still has the C preprocessor, and still has very easy access to many unsafe C routines. Rust has a borrowship concept that may seem complicated but knowing every single aspect of C++ well its much much harder, C++ is much more complex.
Rust has deserved every bit of hype in my opinion, it can be seen as a full and suiting replacement for C++.
Yes, BSD can be converted to GPL, but GPL can't be converted to BSD. That's because the GPL does use copyright law to protect the freedom of the software.
It does use the same legal tools as the proprietary software, but it promotes user freedom with those tools, and not restrictions towards the user.
GPL does remove freedoms, though.
No, it doesn't. Under copyright law, you literally have no freedom to do anything with the code. The GPL does nothing but grant freedoms.
Copyright law removes freedoms, the GPL grants most of them back.
Take a piece of code under (modified 2 clause - assumed since 3-clause is incompatible) BSD. It gets incorporated under GPL (fine, that's why the authors picked BSD).
ok
Now, however, a patch is made to that code, which is GPL'd.
That's just an example of freedom being removed by copyright law, not the GPL. Under copyright law, upstream has no rights to that authors patch at all.
And actually it's subtler than you've made out. You can't relicense BSD code as GPL without the permisson of the author. You can however aggregate them. If the patch is only against the BSD portion of the code, it's not a derived work of the GPL stuff, and so the patch author could license the patch however they wish.
Copyright law allows them to do so.
That's probably the most insulting part of GPL - the claim BSD allows locking up "with proprietary licenses" when in reality, the GPL does the same.
Except it doesn't because the GPL is not a proprietary license for any normal definition of the word.
SJW n. One who posts facts.
Having benefited from BSD and Apache licenses, I'd be quite inclined to use those licenses, or at least things like the LPGL so you can link to stuff ... precisely because they don't wish to assert control over how others use it.
The GPL places no restrictions on use of the software. It places restrictions only on distribution of it.
SJW n. One who posts facts.
You missed a bit about what the GPL also allows:
2. GPL
- You can take my code, and profit from it. You can't fuck me, you can't fuck the people who receive your binaries, you can't fuck the world. We all win. You can also whine horrendously all over slashdot about how the free stuff you've been given isn't eaxctly to your taste.
SJW n. One who posts facts.
SO yes, the GPL has removed freedoms from the code that was formerly BSD licensed.
Yes, that's the consequence of the freedom given by the BSD license. It allows others to take its code, modify it, and parade it around.
You can't relicense BSD code as GPL without the permisson of the author.
Why not ? The 2-clause BSD license says:
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.
Nothing in those two clauses prevents licensing under GPL.
Of course, you could implement a C/C++ interpreter in Rust, including the preprocessor, and all its unsafe practices. So, Rust is still unsafe on a higher level.
An important thing to note about "GPL v3 or any future version" type language. Its not free, its a blank check. A developer has no idea what that future GPL version could say. So Linux and its simple GPL v2 and no mention of future versions is a known, unlike the FSF's current "best practices" which is an unknown, a blank check.
Windows 10 is BSD licensed, wow, I missed that news story.
SO yes, the GPL has removed freedoms from the code that was formerly BSD licensed.
Yes, that's the consequence of the freedom given by the BSD license. It allows others to take its code, modify it, and parade it around.
You do realize the author was simply pointing out hypocrisy? Pointing out that "freedom" is receiving an Orwellian redefinition? The fact is that the GPL is restrictive, it has benevolent motivations, perhaps socially good motivations, but its still restrictive. Saying these restrictions are "good for you" doesn't change this simple fact and redefine restrictive as "free".
The question is: If the BSD license is superior to the GPL, why didn't Open Source really truly take off as a movement with mainstream recognition until after the Linux system and its GPL license?
Linus answered this himself. The BSD on PC projects got tied up in court by AT&T. If it were not for this fluke of history he would not have written Linux. A PC based Unix was such a compelling "want" that whatever project had the first credible implementation would have dominated. Because of AT&T that was Linux. The GPL had nothing to do with it. Linux did not benefit from the GPL, the GPL benefited from Linux.
Two severe flaws in your analysis.
1. There are people and organizations who profit from BSD code and *do* contribute back. Apple does for example.
2. There are people and organizations who profit from GPL code and do *not* contribute back. Google for example. You only have to contribute back if you redistribute. Use it all you want internally, modify it, and you are free to contribute nothing back at all.
No, nothing prevents distribution with GPL software. You cannot relicense software that you do not own the rights to (i.e. that you did not write).
I do not know why this needs to be explained, on Slashdot of all places...
Those who advocate genocide deserve every protection afforded by law, and none afforded by common human decency.
"Take a piece of code under (modified 2 clause - assumed since 3-clause is incompatible) BSD. It gets incorporated under GPL (fine, that's why the authors picked BSD). Now, however, a patch is made to that code, which is GPL'd. That patch cannot be incorporated back into the upstream project because it's GPL'd which conflicts with the original project's BSD license."
Which is exactly why the patch creator chose the GPL instead of the BSD license -and that's also fine, that's why the authors picked BSD.
"It's not a hypothetical - it has happened before."
Why you seem surprised? That's exactly by design, not a side effect.
"That's probably the most insulting part of GPL - the claim BSD allows locking up "with proprietary licenses" when in reality, the GPL does the same"
Have you even read your own sentence? Are you aware it makes no sense? The GPL people claim that the BSD license allows locking up "with proprietary licenses". The GPL doesn't alllow locking up "with proprietary licenses" so in reality, no, the GPL does not do the same than proprietary licenses no matter how your distortion field paints it.
"LOL, yup, that's freedom ... adhere to this long list of restrictions, or don't use it."
Only, again, that's not the case. The case is that you can *either* "adhere to this so called by gstoddart long list of restrictions", *or* "just use and modify it at leisure to the full extent allowed by copyright laws".
At your entire choice.
You cannot relicense software that you do not own the rights to (i.e. that you did not write).
Technically, it wouldn't be relicensing, as you don't remove the BSD license. Rather you just add the GPL license to it. The effect is pretty much the same, though, except for the requirement to keep the BSD disclaimer attached.
The question about Win10 was to demonstrate the sillyness of the line: "It denies freedoms that are otherwise perfectly reasonable, and would exist were the GPL not denying them."
Nothing in those two clauses prevents licensing under GPL.
Nothing in that paragraph says anything about relicensing. The BSD license does NOT give youthe right to relicense the work. You can distribute it all mixed up with GPL code, but youcannot relicense.
SJW n. One who posts facts.
It gives you certain rights, but not the freedom to take the same rights away from others.
Which some of us mean makes it more free, not less. We wouldn't call a society that allowed slavery more "free" than one that forbade it. Freedom is not about the number of rules/restrictions. It's about what those rules are.
Stefan Axelsson
The BSD license does NOT give you the right to relicense the work.
As long as you keep the BSD clauses, there's no rule that says you can't redistribute it with a GPL license on it, without making any change to the code or adding GPL code to it.
If you don't like your government, don't pay it. "Don't use it" is only a valid argument when there are choices or when choices don't harm each other, like fragmentation. Lots of competition for shoes is a good thing. Not all shoe companies make shoes that work for me. Lots of competition for roads just means I get to pay more tolls.
It takes away freedom from the developer and gives freedom to the user. Kind of like saying, screw the farmers, the consumer is all that matters. In many cases the developer and the user are the same, but this is rare. Many developers don't consider this an issue, but the probably with GPL is it's viral and means developers that want to only share certain code are left out of forced to do something they don't want to. Of course they could fragment the market even more and start their own code with some other license, but this is harmful to society in most cases.
In the end, GPL vs BSD is nit picking. GPL is still great, but I hate how they say it promotes "freedom". It promotes sharing, but not freedom.
Just because something is popular doesn't mean it's better. Take Windows or PHP.
The question about Win10 was to demonstrate the sillyness of the line: "It denies freedoms that are otherwise perfectly reasonable, and would exist were the GPL not denying them."
It demonstrates no such thing since the discussion is about FLOSS licensing, Microsoft is a straw man. Sorry your not getting that.
Technically, you can compile CDDL code, then re-license the binary as GPL. This will make a lawyer's head explode. GPL on a binary states you need to license the code as GPL, but at the same time, you can't license the code as GPL, but you can make the binary GPL. Talks about a paradox.
It's tempting to grab a CDDL ZFS kernel module, compile it, then license the binary as GPL and see what happens. Include the binary with a Linux distro. Why can't you? It's GPL.
No. You're an idiot. Stop saying this. You don't get to change the license for any software you did not write. You don't get to add, remove, or modify the license terms. The rule preventing this is USC Title 17. Ask the FSF, or any attorney. Alternately feel free to violate the license and get sued for it -- the terms "statutory damages" will become the stuff of your nightmares.
This is not legal. If you add code to a package (that is not derivative of the original), and license that as GPL, then you can feel free to attach both licenses. In no circumstances can you relicense someone else's copyrighted work without their explicit permission.
For fuck's sake
Reading up on this, apparently the issue isn't with the binary, but the source. CDDL does not permit you to relicense the source, and GPL insists that for any combined work, the source code must be able to be licensed as GPL (only). Theoretically a FUSE module would be acceptable, but probably not worth it. However, there is nothing preventing end-users from installing ZFS as long as it's not distributed.
Also, in my considered opinion, CDDL is a seriously fucked-up license.
Those who advocate genocide deserve every protection afforded by law, and none afforded by common human decency.
You're probably thinking of MAI SYSTEMS CORP. v. PEAK COMPUTER, INC., in which the 9th Circuit held that copying into RAM can be infringing.
The 9th revisited the issue with MDY Industries, LLC v. Blizzard Entertainment, Inc. More or less the same result. It is, strictly speaking, utter madness.
A couple months ago I was emailing with RMS about how sad it was that MIT still required copyright assignment by all students, faculty, and staff. So I'm glad to see some more progress. Related:
http://pdfernhout.net/pledge-t...
"The FSF could start a new campaign to get foundations and non-profits to pledge that all content and software they fund or develop for the public using charitable or public dollars will be released under free licenses."
Ultimately research funders are going to need to change their policies drive this, as I suggested about fifteen years ago:
http://pdfernhout.net/open-let...
"Foundations, other grantmaking agencies handling public tax-exempt dollars, and charitable donors need to consider the implications for their grantmaking or donation policies if they use a now obsolete charitable model of subsidizing proprietary publishing and proprietary research. In order to improve the effectiveness and collaborativeness of the non-profit sector overall, it is suggested these grantmaking organizations and donors move to requiring grantees to make any resulting copyrighted digital materials freely available on the internet, including free licenses granting the right for others to make and redistribute new derivative works without further permission. It is also suggested patents resulting from charitably subsidized research research also be made freely available for general use. The alternative of allowing charitable dollars to result in proprietary copyrights and proprietary patents is corrupting the non-profit sector as it results in a conflict of interest between a non-profit's primary mission of helping humanity through freely sharing knowledge (made possible at little cost by the internet) and a desire to maximize short term revenues through charging licensing fees for access to patents and copyrights. In essence, with the change of publishing and communication economics made possible by the wide spread use of the internet, tax-exempt non-profits have become, perhaps unwittingly, caught up in a new form of "self-dealing", and it is up to donors and grantmakers (and eventually lawmakers) to prevent this by requiring free licensing of results as a condition of their grants and donations."
Longer version of the above originally prepared for the Markle Foundation:
http://pdfernhout.net/on-fundi...
A 21st century issue: the irony of technologies of abundance in the hands of those still thinking in terms of scarcity.