Slashdot Mirror
"New Copyleft License" released
Stephen Williams
writes "LinuxToday reports
that Bowerbird
Computing have released a new open source license
called the New Copyleft
License. Seems to be aimed at people who want to sell
their free software, rather than charge for support."
At the rate these licenses are proliferating,
soon there will be one license for every app. Does anyone
besides me think this is getting crazy?
It says to amend the GPL (section 10), to include a proviso for the NCL if you're releasing sftwre
under the NCL. Did I miss something here? Where in the GPL does it say you can legally alter it
to include such a thing? Isn't that illegal?
...
.. "Doesn't Barbie come with Ken?" "No, she fakes it with him."
This fails DFSG #1:
This license and those who use it are missing the point. It's a step down the slippery slope of semi-free, nominally free, or we-can-make-people-think-it's-free licensing. I think such a mentality is bad for the community.
If you want to make money licensing code, fine, do it. You can always re-lease it under a free license later. But don't call code with a restricted license "free." It's an abuse of the term and an affront to the community.
No, it doesn't say to ammend the GPL. It says to add a preface to your licensing terms that grants additional permissions.
As I (briefly) read it, this probably goes against the grain of the GPL's viral nature. One of the central tennets of the GPL is that GPL'd code can't be un-GPL'd, and such a preface would explicity say that yes, the code can be un-GPL's and be NCL'd instead.
I think this will cause Stallman et al. To say that this license if evil.
Personally, I think it is the most interesting license of the recent variants that we've seen. It's got problems, but it's not just a fake GPL with get-out clauses. It's a genuine attempt at a new kind of Open Source licence, and that's a good bit of diversity.
Anf finally, no, I don't see a problem with all these licenses. Evolution suggests that winners will emerge, and the poor ones will fail. Fine by me.
-----
How about a web site which gives you a tick-box based license generator:
..."
* You'd select what you want to limit/allow
* It'd read it back to you in
a) Formal license form (text file)
b) English, with warnings (eg. "You're forcing people to release their own source code here")
c) A Geek-code style shorthand
Then, any savvy user could take one glance at the Geek code and understand the restrictions.
"This software is licensed according to the Open Source Modular License (URL here) with the following stipulations to be interpreted as described by the license:
FREE++ SRC- DIST+ COPY+++
Comments?
Even the way in which NCL software reverts to GPL after two years is identical to the way Aladdin Ghostscript reverts to GNU ghostscript after one year. That doesn't make NCL software free software.
The NCL is not a copyleft as defined by the Free Software Foundation. The Aladdin Free Public License is not a free license as defined by the Free Software Foundation. Please do not be misled by these abuses of the terms "free" and "copyleft." Recognize these licenses for what they are: non-free licenses.
Given the plethora of licenses which seem to be appearing it's becoming quetionable whether OS is helping the community. We only have to look at the furore over Apples license to see the first inclining of our worst fears: if people disagree about whether a license is OS then before too long we'll be lost.
People should definitly read/understand the licenses of the s/ware they are using. In general I think it's getting to the point where I will prefer well-known licenses like GPL over anything else. If I *need* to use the s/ware then I'll read the license etc, but to try and discourage the practice I'm now trying not to use s/ware which isn't under one of the traditional licenses.
Sound crazy?....GPL/BSD/Artistic/X licenses were fine before I see no reason to encourage people to muddy the waters further.
If the OpenSource people want to help they should make a license from the most used ones which meets the definition and then only allow the use of the mark if this license is used. Allowing people to use the mark because *they* think it meets the definition just isn't cutting it. Further ESR isn't a lawyer, what he thinks meets the definition and how the law might view the clauses could be vastly different.......
Actually, there is at least one very good model of this that I know of. SBT Accounting systems is a Foxpro application sold by a company whose license goes something like this:
> the source code comes with the purchase
> if you don't modify the source code, they provide 100% support for a limited amount of time (I think it is like a year)
> if you don't change the source code, you may purchase additional support
> as soon as you change even one character in any of the code, it is your code, you own it, and there is no support.
And I actually think this is a pretty good business model. It is the open but not free source, and they are basically saying they will back their code, because they know it works, but they won't back your code, because. . . it's yours!
"My husband invented the internet, and I censored all the naughty stuff on it. .
I like their idea; this seems to me to be close to the original idea of patents (which is what they seem to intend): the inventor gets brief exclusive rights, then anyone may use it. Any software that is released under this license will be free software (note use of the future tense).
:)
However, this license is broken. Their stated intent in the preamble conflicts with the wording of the license.
They want the author to be able to make money by distributing the software. The license grants the author exclusive for-profit distribution rights, but lets anyone distribute the software for free. From a commercial standpoint, how is this different from GPL software? If I'm selling software, I still have to compete against distributors who charge $0 for the software. Customers would have no more reason to buy from me under this license than they would under the GPL.
I think it would make more sense (given their stated purpose) to give the author exclusive distribution rights for a year or two (before GPLing it), and require that the source code be included, and allow licensees to modify the source code.
They should have thought about this a little more before releasing it. On the other hand, maybe they figured that given enough eyeballs, all bugs are shallow, including legal ones
Maybe what us free software advocates need to do is draft a shell licence for commercial software companies. If you look at their "open source" licences, they're all basically the same. We seem to have problems with the portions of thier licences that we feel funnels the benefits of the open source model to "the company" exclusively. They seem to believe that they've created an open source license that protects their IP from becoming public domain, and from the threat of cloners. Maybe we can create a license model that satisfies both requirements. ;) should be working on, not trying to get a bunch of commercial software companies to dilute the free software with a bunch of "not-so-free" software. The GPL is great for new software that has no real original owner and no need for IP protection. For commercial SW companies, it not just gives away the code (a good thing), but it gives others the right to burn it to CD and sell it as is (a bad thing). -earl
Essentially, they want to maintain the ownership of their "open" code, and we want to maintain the freedom of the improvements that result from the opening of that code. I think that's the issue that ESR, and BP and the other "champions of open source"
No, the point is clear. These people are (just like several other "open source proponents," doing what every competitive (or sometimes greedy) living thing does when it comes time to compete--they're creating lots and lots of eye-catchers. These things are intended to look "just like" the GPL "except for..." and usually, the exeception has something to do with watering down the GPL's intent. Why not be straight with people? This is not copyleft. If you want to license your software with something other than the GPL, do so, but don't pretend you're using the GPL.
A good example is GhostScript. They license under a different license, then eventually re-license under the GPL. Nothing wrong with that, but they don't pretend to have a "public license" when they really don't have one.
No doubt about it, every author has the right to license her own work any way she sees fit, but this pretentiousness about "public licenses" has got to stop.
I'll never forget a certain (nameless) "open source" guru who once wrote me: "We can't let Richard Stallman hold us back any more." Well, nameless one, this crap is the result.
Want my advice? I believe the moral thing to do is license your software as you see fit, and it you want to then switch it to free software, use a free-software license later. Don't pretend.
--------------Rev. C.C.Chips---------------- For the real truth, visit
This simply goes back to one of the fundemental problems the Linux/open source/free software movement faces...in involvement of comercial organizations or not. Most commercial entities will not go for the GPL, yet have valuable contributions to make. On the other hand, this license proliferation is just mucking the whole thing up, and licenses like the BSD license allow the code to be exploited too easily without return to the community. IMHO, we need to have just three licenses, the GPL, LGPL, and a third one which will make commercial entities more cofortable yet still enable it to work with GPL software and its devotes...something like a NQFPL (not quite free public license). make no pretense to be compelety free, but yet still works with the movement. In the NCL license defense, I think the two year expiration period is a decent idea...but, simply put, we have too freeking many licenses running around, their going to become meaningless soon. Perhaps dialogue on a third license of non-commercial origin, but with commercial entites in mind, could be developed. I would love to see FSF and OSI involved in this...an attempt to build consensus.
It is fully buzzword compliant, that is it uses the terms "Copyleft" and "Free" (while avoiding the legally protected term "Open Source"), but this is a very dangerous license to the Free Software community. I urge you not to treat this license as "Copyleft" "Free" or "Open Source", and avoid software licensed under the NCL.
I call this license dangerous because, while it pretends to be a Free license, it deliberately interferes with the redistribution of its software, one of the three pillars of Free Software.
Here's a hypothetical example, lets say the NCL becomes popular. There are fourty commonly used NCL packages out there. None of the authors are greedy, they just want $1 per copy sold for their two years. If a company like RedHat were to include these popular packages, the cost of their distribution will have gone up from $50 to $90. This will hurt sales, with no increase in their profit, thus hurting the free software community (a little).
Worse, when CheapBytes goes to repackage this version of RedHat, they will have to make their own arrangement with each of the authors, and their distribution will have gone up from $2 to $42, probably heigher, because they will need to hire someone just to manage all these agreements. This will destroy them, companies like them, and seriously harm the entire free software community.
Unencumbered redistribution is part of what makes Free software Free. This license mocks the community while asking us to accept them. I say throw them on their ear.
----
Open mind, insert foot.
The DFSG #1 states that the license "may not restrict any party from selling or giving away the software".
The NCL states: "You MAY NOT distribute this work for profit as an executable or in object form".
Does anybody else remember a little thing that programmers like called "source code"? The NCL doesn't appear to restrict source distribution, just binary. Thus, the DFSG only chokes if it defines "software" solely as "binaries". If, however, it includes "source code" in its definition of "software", there's no inherent conflict.
True, but the license differences ensure that different programs can't "interbreed". Which is a shame. Mozilla and KDE or GNOME could have beautiful offspring.
I values freedom of information and cool stuff that works well.
This basically makes me a GNU sort of guy.
What do I do about all these coporates jumping on my bandwagon? Nothing. I use stuff that meets what I value, and rememebr that we got this far without them. They didn't matter then and don't matter now, at least not in a way that guides what I do.
If what this community has done is so great, then the last thing we shoudl do is change as soon as somebody (or some corporation) agrees that it's great, but wants to change it.
People like Linux becuase it's good. If we keep making good things, world domination will happen naturally. If people don't like what we make, chances are we still will and we'll have the stuff we like.
What more do you want?
The NCL is not what I like, so I won't use it. I'll just go back to doing what I did before and save my public displays of opinion when somethat that actually tries to stop me comes up. (Like the US government's ideas about privacy or regulating the net.)
NEw liscences, new apprioaches, new coproate ideas about our community,... they all fall under the utlimate equalizer that is the internet. It will be only as big as it is cool in the eyes of the netizens.
--
Erskin
geek.
"distribution" here of course refers to the "Free distribution" that #1 is about.
DFSG #2 are part of the reason why e.g. Pine and qmail are not DFSG-free. Debian has source packages of them in non-free (pine396-src, qmail-src).
It's termed a "Copyleft", but to me "copyleft" always meant "GPL, or another free software license that uses copyright law to prevent hoarding into proprietary software". But NCL isn't a free license.
Look at what Stephen Williams says in his submission: Seems to be aimed at people who want to sell their free software. But software licensed under it isn't free software.
Look at what Stephen Williams says in his submission: Seems to be aimed at people who want to sell their free software.
Yup. I saw the word "copyleft" and immediately associated the license with "free software". I should have done more than skim it before submitting the story to Slashdot; then I wouldn't have used the word "free" in my submission.
I've sent a message about the NCL to the FSF. They might want to comment on the copyleftness or otherwise of this license.
It disregards the role of the unpaid collaborator who would add features to your program, because the initial developer has an advantage that the unpaid collaborator can neither obtain nor circumvent. This is a disincentive to the unpaid collaborator because instead of contributing work to the community they are now contributing work that someone else will be paid for no matter what they do. Contrast this to indirect revenue methods. If I don't like Red Hat, I can circumvent them and make my own Linux distribution, obtaining what would have been their profit for myself for some (possibly small) number of customers. Consider this in the case of a product like Linux, where the initial developer contributed a small amount of the total work and his services as an architect and coordinator, while the lion's share of the work was done by others. It makes collaborative development unwieldy. If every developer insists on their own revenue capture, you would soon have a too-expensive product or a paperwork and procedural mess. Who decides how much each developer gets? Who decides who is worthy as a developer? Do they all make that decision for themselves and then compete with each other in some way? It gives the initial developer a lock that causes a disincentive to "fork" a product. If Linus had direct revenue-capture from Linux and I decided to make a fork of it because I felt I could engineer it better, I might be able to do an excellent job, but Linus would still be compensated for my effort.
So, to sum it up, I think that direct revenue capture works to the detriment of collaboration.
Thanks
Bruce Perens
Bruce Perens.
This license and others like it can be classsified as "Mandiatory Direct Revenue-Capture for the Initial Developer."
It disregards the role of the unpaid collaborator who would add features to your program, because the initial developer has an advantage that the unpaid collaborator can neither obtain nor circumvent. This is a disincentive to the unpaid collaborator because instead of contributing work to the community they are now contributing work that someone else will be paid for no matter what they do. Contrast this to indirect revenue methods. If I don't like Red Hat, I can circumvent them and make my own Linux distribution, obtaining what would have been their profit for myself for some (possibly small) number of customers. Consider this in the case of a product like Linux, where the initial developer contributed a small amount of the total work and his services as an architect and coordinator, while the lion's share of the work was done by others.
It makes collaborative development unwieldy. If every developer insists on their own revenue capture, you would soon have a too-expensive product or a paperwork and procedural mess. Who decides how much each developer gets? Who decides who is worthy as a developer? Do they all make that decision for themselves and then compete with each other in some way?
It gives the initial developer a lock that causes a disincentive to "fork" a product. If Linus had direct revenue-capture from Linux and I decided to make a fork of it because I felt I could engineer it better, I might be able to do an excellent job, but Linus would still be compensated for my effort.
So, to sum it up, I think that direct revenue capture works to the detriment of collaboration.
Thanks
Bruce Perens
Bruce Perens.
Thanks
Bruce
Bruce Perens.
What we really can't tell in advance, though, is which of these putative issues will prove to be a real problem and which will remain purely theoretical. And until we have several years of experience with different "commercial open software" licenses, any conclusions about how the hacker community (or the lawyer community, or indeed the user community) will behave under conditions X, Y, and Z will be purely speculative.
Netscape, after a year, seems to offer a couple of tentative observations:
All we can really do is enjoy the new flood of source washing in and try not to spend too much time quibbling about exact license terms. (By and large, corporations will listen more closely to the lawyer they're paying hundreds an hour to than to J. Random Hacker anyway, and there's not much we can do about that.) It'll be fascinating, though, to watch the commercial license situation shake itself out. But it'll take patience.
Craig
I don't know about you, but I write code for a living. I write a lot of it. I spend a lot of time at work. The more I work, the more money I make. You see, I make money so I can eat and live in a decent place.
I would much rather work on my hobbies than my professional stuff, but I need money. So, in the few minutes of free time that I have, I hack up a few little GPL'd programs.
If I could sell those programs, I might work a lot harder on them, and not spend so much time at work writing proprietary code. The problem is nobody wants my code unless it's GPL'd. It's a religion. Must... have... GPL.... Look at what happened to Qt. It had a free license (not GPL free, but free enough for free software to benefit.) Nobody wanted Qt until it went QPL, and most people still don't like it.
We wouldn't have Ghostscript without licenses like that (AFPL).
Does it really matter if the FSF or Debian say it's not free? It will be.
This license is great because it *forces* those that use software licensed under it to give back to the community instead of people just leeching off of the community. Granted, it needs a little revising, but the spirit is good. Feed the coders; They give you free stuff.
I agree with the folks who think this thing is bad juju. It's basically the same as "Free for noncommercial use" only with that clause expiring. That's not free software.
Anyway my question is, how will this thing work in real life? Say I write SuperWidget 1.0 under the NCL, Bob comes along and modifies it, Larry modifies that. Then Will Fences decides he wants to make a commercial app from Larry's code: does he have to pay Bob and me, too? Seems like that could snowball quickly, making the license something businesses would want to avoid, not embrace.
--B
Because you raise some red herrings, and it answers some of your legitimate questions.
It clearly does not "disregard the role of unpaid collaborator". Depending on how the steward chooses to divvy up the money (if any), the all collaborators are paid. If not, the collaborators are free to choose a new steward who will divvy up the money they way they want it divvied. (Although the exact details of this seem to me one of the weaker areas of the license.)
You ask "who decides how much each developer gets?" The license answers this very clearly: the steward, who is elected by the developers.
Take the Linus/Linux example which you raise. If Linus (as steward) chose to keep all the money himself, no doubt an election for a new steward would quickly occur. The new steward might decide that Linus gets no more money, since his contribution in terms of lines of code is small compared to the total size.
Of course, the details on a project with that many contributors gets hairy very quickly, which is why I think this is the real weakness of the license. But that's a detail.
If you're going to attack this license, read it and attack it on its real flaws, not on some strawmen you conjure up. A license something like this can significantly add to the availabilty of free software, by encouraging those who would not release under GPL or BSD. Let's fix its real flaws, instead of wasting breath on imagined ones.
-- Al
-- Alastair
So, to sum it up, I think that direct revenue capture works to the detriment of collaboration.
The problem with the lack of a direct revenue capture system is that it discourages commercial companies from investing large amounts of money into upfront research (for open source projects anyway).
How many companies are going to invest millions of dollars into research if the only revenue they can see in the future is from support? New companies coming in have the advantage of not having to recover the cost of developing the product in the first place.
Yes, its possible for companies to survive and propser by selling support for GPL'd products, but these companies are nearly always established after a lot of free time has been donated where the risk is a lot lower that a useful product is not eventually developed.
The lack of a direct revenue capture system may in fact discourage funding for new innovative development.
I don't think anyone at *Netscape* is saying the license doesn't matter. There has been a lot of code we can't use (GPL), and a lot of people we'd like to collaborate with who can't use our code (again, GPL issues).
Those are two separate issues. Netscape will never be able to incorporate GPL code because of crypto restrictions. Our lawyers have agreed to let us take baby steps toward fixing the second problem by dual-licensing the JavaScript engine. If it works out well we may be allowed to dual-license more of the code eventually.
(Keep in mind that it has taken a year to get this far, and that we now have a new set of lawyers to deal with.)
Contrary to what Bruce suggested, this liscence compensates the "unpaid contributer" more than the main author. According to the liscence each author (meaning anyone who has contributed code to the work) gets an equal vote on who the steward is, so noone has the leverage to keep most of the profits to themselves.
Of course, there do seem to be problems with this liscence regarding the details. I could get my friends and family and pets to submit code I wrote so that we could get a bigger share of the vote (and thus the pie). But I suppose the other authors could always fork the project and refuse to let me or my family and pets submit any code, once they realized what I was doing...
--Artemisia
Bruce
Bruce Perens.
Just think about how the taxes would be handled. What a book-keeping mess.
Thanks, but I'm going to continue writing GPL-ed stuff and not get into squabbles over who gets paid what.
Bruce Perens.
> This license and others like it can >be classsified as "Mandiatory Direct >Revenue-Capture for the
> Initial Developer."
> It disregards the role of the unpaid >collaborator who would add features to your >program, because the
> initial developer has an advantage >that the unpaid collaborator can neither obtain >nor circumvent.
> This is a disincentive to the unpaid >collaborator because instead of contributing work >to the
> community they are now contributing >work that someone else will be paid for no matter >what they
> do.
Bruce, you should have read my license before opening your mouth. It makes it quite clear in the definitions section that an "Author" is anyone who contributes code to the project. Under the stewardship section it then specifies that each author has just a single vote, regardless of contribution. So even if Linux were licensed under the NCL, Linus would have no special rights under it.
> It makes collaborative development >unwieldy. If every developer insists on their own >revenue
> capture, you would soon have a >too-expensive product or a paperwork and >procedural mess. Who
> decides how much each developer >gets?
Again, as a leader in the free software community you have the responsibility to actually *read the entire license*, rather than just rely on the quickly written preamble. All of these decissions rest in the hands of the steward of the work who is elected by the the authors, each with one vote.
>So, to sum it up, I think that direct revenue >capture works to the detriment of collaboration.
I understand your fears about collaboration and free reuse of code, that is the sole reason I included the stewardship system in the license. Without a guarantee of these freedoms the license couldn't be called free. I believe that this license will protect these freedoms, although only experience will tell.
I didn't write the license out of some kind of selfish grab for money at the expense of the freedoms of others. I wrote it because:
1) The free software community is bigger than the FSF, yet they are the only beneficiaries of any donations from people like Cheapbytes. The distribution of reward should be more even throughout the community.
2) There are a great many programmes that just do not lend themselves to offering consultancy services. How do you offer consulting for a game, or a genealogical programme? Home users will never purchase consulting services, so the only way to make money in this environment is to sell the software. My intention in this respect is to provide some motivation for programmers to write these sort of programmes professionally.
As I said, only experience will show if the license works properly, but I believe that it is atleast worth a try. The GNU GPL itself was a radical idea at the time, so I don't see how a supporter of it can then turn around and refuse to even support a trial of a license that contains a new idea.
Matthew Parry
Bowerbird Computing.