Slashdot Mirror
GPL 3.0 to Penalize Google, Amazon?
Michael Ferris writes "Is this the start of a shakedown by the GNU folks? Michael Singer writes that Eben Moglen and the folks rewriting the GPL are looking at a proposal where companies would be required to pay money if they use GPLed software, even if they don't redistribute the software." From the article: "The current version of the GPL, which was last updated in 1991, fails to trigger the open source license if a company alters the code, but does not distribute its software through a CD or floppy disk...the [current] rule does not apply to companies that distribute software as a service, such as Google and eBay, or even dual-license companies like Sleepycat."
Call me paranoid, but I wouldn't want even the faintest chance that some nasty corporation managed to litigate itself in the position of being able to release a future GPL version, as in bold below:
This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.
That's handing control of the licencing of your code over to whoever is allowed to write GPLv1851, if I'm reading it correctly.Get your own free personal location tracker
the [current] rule does not apply to companies that distribute software as a service, such as Google and eBay
Google and eBay distribute HTML. That HTML is created by software that uses GPL code. So if I modify a GPL Office Suite, would I have to distribute the code if I email someone a document I made with it? Seems like a bad idea, in general.
I guess people could fork the GPL2.0'd code if the software developers switched to GPL 3.0
I think requiring companies to pay is a big time mistake, and to me, it
hails back to the days of the old BSD style licenses. The GPL works
so well now, precisely because it is unobtrusive. My company runs
GPL'd software because we are able to use it and make modification
without either redistributing the source code and we aren't required
to pay for that ability.
Requiring people to open all their changes or pay for them will put a
lot of businesses off when it comes to dealing with GPL'd software. I
don't think that is a good trade off to make, and I don't think it
will be healthy for the open source community in general.
A move like this will make the newer BSD style licenses and / or
licenses like the Python license much more attractive imo. Now that
open source is finally turning the corner, and solid technologies are
finally moving into the enterprise, why would we even entertain making
changes that will certainly hamper open source adoption?
This isn't a consistent position in my opinion. If you are
developing free software, it should remain free. The idea that you
can get someone hooked on software, and then pull the rug out from
under them and start charging them is ludicrous. If this were to
happen, I can honestly see a major fork in the GPL happening.
Doug Tolton
"The destruction of a value which is, will not bring value to that which isn't." -John Galt
Shouldent that be FAIBFSF? Free as in beer free software foundation?
"It's so convenient to have a system where everyone is a criminal" - A. Hitler
I _REALLY_ hope this idea falls flat on its face. These companies are the brightest hope for Free Software. Screwing them over is incredibly idiotic.
I think this proposal is crazy. If you use software as a part of running your business, that software is benefiting you and indirectly providing services to your customers, even if they never see it. So where do you draw the line?
-73, de n1ywb
www.n1ywb.com
But who gets paid the money and who determines how much? Can rates increase as GPL'd software develops? Is money paid to Richard Stahlman [sic] to distribute through the FSF? Seems an easy way to get around this anyway would just be to have small closed-source scripts that would only be called from the GPL'd code. After all, what's wrong with that? (so long as they're not redistributing their code, just letting use of it as a service)
(\_/)
(O.o) This is Bunny. (> <)
"If you look at the market, Yahoo, eBay, IBM, Amazon, Google have all sunk millions into the GPL infrastructure," Olson said. "Not only are we changing the rules, we are changing them retroactively. With the new way, it lets the customer pay with either their source code or with their wallet."
Basically, in any other language: 'Now youve had time to build a good infrastructure on the current rules, prepare to be shafted'. If this comes to pass, then in my mind they are no better than Microsoft changing EULA terms with a service pack. Now that there are major companies with an infrastructure built on GPL software, the FSF are looking to essentially move the goalposts and if this is applied retroactively to current code (which from the articles wording I think it will be) then I personally think that its going to do more harm to the GPL community than benefit it in code donations as companies scramble to move away.
Please someone tell me that they cant do this retroactively, that its impossible under the current GPLv2 terms.
Just the implication of this terribly and inaccurately worded headline will be devastating to open source. This article is intentional flamebait. Zonk, please try to be a bit more responsible with the articles you post.
Regardless of what GPL 3.0 turns out to be, developers are not forced to use it. They can continue to use GPL 2 if they wish, just as they can choose to use a BSD license, Apache license, creative commons, or any other license of their choosing. Furthermore, software that has already been released under GPL 2 cannot be retracted, it remains available under GPL 2 forever.
Such a clause will really ensure BSD never dies.
The submitter isn't clear about the fact that this would not apply to everyone who changes software for commercial use but does not redistribute. This applies ONLY when "redistribution" of the software sort-of-occurs, because the software is used to provide a service. For example: any open source software Google uses in its search engine interface does not count as "redistributed", even though it _interacts_ with users of Google, because no actualy _binaries_ were shared with those users. For another example, if I modified the GIMP and then let people use my modified version over the Internet, but not on their machines, I would not have "redistributed" my modifications. This is considered by many to be a "loophole" to be closed.
Why is everyone always saying "Software should be free, unless you're a business, in which case, get your checkbook"? The GPL shouldn't mandate anything except that code be made available. I think corporations should be entitled to the same rights and privileges as private citizens.
REM Old programmers don't die. They just GOSUB without RETURN.
Frankly, I don't think this is even the case. Yes, the FSF has been talking about making a GPL 3.0 for a while now, but the proposal they're offering makes no sense.
The problem is that it infringes on one of the four freedoms, specifically the freedom to use. If such a provision were to find its way into any license that made it so that companies and individuals were not subject to the same terms, the license would be both discriminatory and non-free.
Simply put, this is somebody making FUD about the GPL. Don't buy into it for a second.
Haec merda tauri est. Ceterum censeo Carthaginem esse delendam.
I heard the next version of the GPL will require everyone to throw pies at Bill Gates.
At least, that's what I heard anyway.
When whatever changes come up, they will be reviewed and we can rant and rave about it at that time.
Nothing is founded, no concrete written agendas were tossed out by the article and all we have is a hypothetical situation that would be very different from the current model.
Like many have said and will say, it's usually a good idea to specify the version of the GPL in which you release software. Unless you really don't care what kind of changes are made at a later date.
I'm not saying all of this isn't going to come true, but at this juncture we could very well be required to throw pies to comply with the next redistribution agreement.
"You should always go to other people's funerals; otherwise, they won't come to yours." -- Yogi Berra
Moglen has stated: "I do not believe that we will be reach consensus on this front, so I believe the license will have to accommodate options as to the question of Web services, but this must be squared with the ideological pursuit of freedom". [ref.]
And frankly, it's not really a loophole. Web services are not distributing software, they're running a service using software. That's obviously open to interpretation, but I haven't ever heard anyone distribute under the GPL and complain about someone using their software as a web service. There has been at least one derivative license which has addressed this issue.
In the end, GPL 3.0 will likely provide an optional provision which will 'trigger' GPL source distribution requirements for a web service, at the option of the copyright holder; that is really the best choice. Rather than getting into an enormous philosophical debate over whether the idea is "good" or "bad" or "punitive" or whatever, let's simply have two clear licenses and give the option to the copyright holders to decide under what terms they will license their property.
That new constraint would be a totally different principle for the license. The current principle is that if someone gets an executable from you, they also can get the source code, just as you got the source code from which you made your executable. Just using the source code, or customizing it for your own use, doesn't require distributing the source. The new principle would be requiring anyone who customized the source to release all customizations.
This will stop many programmers, many of the best programmers, from using source code under that license. We could no longer keep any of the value of the software we created to ourselves. What is a "web service"? Is my email-processing CGI a "web service"? Any software in the same workflow as any other software under this license would have to be released. So many developers won't make small customizations, because that would force us into the source code distribution business, with all its overhead. Or we might just ignore that provision, or the whole license, en masse.
The GPL is successful because it is a fair contract, even though it's revolutionary. Its enforcement teeth are rarely tested, because it's so close to an equitable agreement among peers. Which has resulted in lots of value contributed by profit-driven organizations, despite the claims of many that the license is anticapitalist. Upping the ante, to require private customizations to be published, could stop the rising tide of acceptance that is pushing GPL to be the default, and any proprietary license to be radical. And then the caution it would inspire: investing in GPL'd software might force acceptance of ever-more demanding licenses, like a GPL4.0 that required redistribution of even software that wasn't changed at all, just to get users "to pull their weight".
The GPL2.0 isn't broken. Let's not "fix it" in a way that could destroy its success, and our chances to benefit from one another's work without onerous burdens.
--
make install -not war
"If you look at the market, Yahoo, eBay, IBM, Amazon, Google have all sunk millions into the GPL infrastructure," Olson said. "Not only are we changing the rules, we are changing them retroactively. With the new way, it lets the customer pay with either their source code or with their wallet."
The buisness world is on a very good wicket. They are getting us to make the tools they use to enslave us.
Just becasue something is free as in beer doesnt mean it is devoid of all responsability from the users behalf.
If a company's buisness model is dependent on free software then its in their companies best itnerest to be very generous to the programers who maintain the software they require.
It sounds like the GPL v3.0 is trying to make the commercial world a little bit more responsable.
future submitted patches can be GPL v3 only. the original author already gave permission for all of their work to continue and be relicensed under v3 in the future if the overall project decides to go that way. if that author doesn't like it they don't have to contribute to the project that has moved on to operate under the new version.
With the advent of the internet, it's now easier for computer programmers to freelance (see the well known Rent-a-Coder website). My impression is that under the current GPL, a programmer possibly could legally do contract work by adding to a GPL'd project without being required to release the work -- so long as the buyer will never distribute the program via CD, floppy, or internet (i.e. keep the program "in-house"). Someone, correct me if this is not true.
It appears that closing this loophole will also close the doors for programmers to freelance in this manner. That is they won't be able to sell their programming service of enhancing a current GPL'd project -- unless, of course, the solicitor agrees to either pay or release the code. The other option is to force the programmer to pay the GPL fee and roll this expense into the contract costs. I think this issue has to be debated and discussed at length, because we can't go about and make a gut decision of saying this provision is a good addition to the GPL just because we want to make big companies like Google, Amazon, etc. pay. After all, it could affect freelance programmers -- this could very well be their bread-n-butter. Let's not get into the question of whether personal enhancements of GPL'd programming tools (e.g. IDE's) are required to be paid or released under the proposed provision, that could be a sticky situation.
Linux at home
My god, I hate it when HP,MS or whomever does this. Now the GPL goes this way, UGH. It's not the fact that they are changing the GPL, that's bad enough but the fact they are retroactivly changing is what makes it so bad. This is the kind of shit the the bad boys do.
"They" can't change the license retroactively. Any software licensed under GPL is either "version 2 only" or "version 2 or later, at your option". So, any GPL software that exists at the moment the GPL-3 is unveiled, you may continue to use as long as you want under GPL-2 terms. Of course, this will not be true of new or updated software released afterward under "version 3 only" or "version 3 or later" terms.
I hope this idea goes no where or you can see any the profesional devs go elsewhere really quickly.
If the GPL-3 turns out to be as implied in the article, I completely agree with you. And I think that most free software authors will as well. If the terms of GPL-3 are that bad, no one will use it and it will quickly become irrelevant - nothing to worry about.
But let's wait until we actually see a draft of a proposed GPL-3, not get upset over third-hand rumors.
- Kevin B. McCarty
Google and eBay allow end users to run programs on Google and eBay servers that create HTML for the end user. If I go to Google and do a search, INPUT occurs on my computer, PROCESSING occurs on Google's computer, and OUTPUT occurs on my computer.
That's not the same as you distributing a document you created. In that case, INPUT occurs on your computer, PROCESSING occurs on your computer, and OUTPUT occurs on your computer.
The question is: What's important in determining if a program has been "distributed"? Clearly if I give you a CD with the program and you run it on your computer, I've distributed the program. Clearly if I take GPL software and modify it and I run it on my computer only, I have not distributed the program.
But if I take GPL'd software, modify it, and then let YOU run it on my computer...
Did I distribute it, or not?
Your immediate answer is probably "Who cares?" But now what if I charge you to run this modified program on my computer, for example, by charging a fee if you use my auction program? Now I'm using software provided by the open source community for my financial gain, but not returning the modifications I made to that software to the community.
The extreme of this problem is that eventually, the internet becomes so fast and clients become so dumb that software is never "distributed" at all. I take an open source office suite and then modify it. According to the GPL, if I then sell that software on CD, or by download, so that people can actually run it on their computers, I must provide the source to it as well. But what if I'd rather just make money off of the GPL'd software I've taken without giving anything back to the community?
Well, then I just put the software on my own server, and instead of selling CDs or downloads, I let people provide input and receive output over a remote connection to the program running on my server.
And wala! People can modify and essentially provide GPL'd software without having to provide source.
paintball
Currently the GPL forbids redistribution of GPL-covered code linked with other code to form a larger work, when that other code is not also distributed under the GPL: the whole larger work must be so distributed. It is a "derived work", and copyright law is clear on this: one can not redistribute derivatives of copyright works without permission. It's that permission that the GPL grants, when you abide by its terms.
Of course, you are not prevented from redistributing GPL code *aggregated* with non-GPL code.
The problem arises from distinguishing between derived and aggregated works: what if I distribute a GPL app that I write, with a bunch of GPL shared libraries that I didn't write (complete with source), and a nonh-GPL proprietary library, essential for the application, that I did write, without which the application is useless.
Is that an aggregate, or a derived work of all the GPL code I did not write?
On the one hand, there's no (legal) requirement that my app even run, and, since I distribute source, one can implement a version of the non-GPL library. The fact that it works when my non-GPL library is installed in the right directory (LD_LIBRARY_PATH, anyone?) is a happy coincidence. Redistributing my GPL app, with modifications is fine, but you can't redistribute my "essential" non-GPL library, rendering the app rather useless without it. You add value to the app, I license more copies of the essential library.
On the other hand, the proprietary library, the foreign GPL libraries, and my app together, constitute a derived work of the GPL libraries, and must be redistributed en masse under the GPL.
If the libraries are statically linked into a monolithic executable, there is little argument that the whole source must be redistributed. The situation is more controversial if a mere aggregate of files is distributed: some argue that not distributing the proprietary library under the GPL is a violation, other's don't.
I tend to believe that if the GPL-covered peices can be redistributed indivudually, without the proprietary library, regardless of whether the result is useless, there is no violation: the fact that the code is actually only useful when the proprietary library is present is but a happy coincidence. What if I distributed a GPL version of the library, and offered a proprietary replacement, with far better performance?
This controvery gets even muddier when one considers alternate ways of effecting program linkage. In effect, the "functional derived work" exists only at run-time, and, indeed, the maner in which the parts of the aggregate are combined, can, itself, be the subject of restrive licencing, and patent. What about linkage via a remote procedure call mechanism? Y'all remember RPC/XDR over TCP/IP, right? Suddenly, self-assembling functional derived works become a reality. Protocols like SOAP, used to support "Web Services" exacerbate the problem.
This leaves a big, gaping hole in the GPL: socket wrappers were a common "hack" to "get around" the GPL: just layer a RPC mechansism around the proprietary library, and a GPL wrapper to call it from the app, and you were all set. And, that didn't even address the issue of inter-machine communication: 127.0.0.1 and Unix sockets are ubiquitous.
Of course, the minute one's app "links" with proprietary code on a different server in this manner, the GPL loses all force, for one did non distribute the proprietary part.
It is this area that the GPL v. 3.0 tries to address, IMHO, perhaps by more differentiating between linkage and aggregation. This can be done, of course, but then the license starts to lose some of its roots in copyright law.
The problem with such an approach, though, is, what constitutes linkage: does an exchange of HTTP requests and responses? What if there is ultimately GPL code satisfying that request? Surely, if I use a program provided by a vendor
You could've hired me.
Well, it's wrong in almost every case, and Olsen should be ashamed for saying that. (Though I wouldn't be surprised if an article came out tomorrow saying he was "misquoted").
Nearly all GPL software is released under "v2 or (at your [the user's] option) any later version". If not that, then it's "v2 only".
Never seen "v2, unless there is a later version, then the latest applies" So he's wrong about the retroactivity.
Paying with your wallet isn't even an option with a lot of projects either, most projects with mixed code bases (i.e. they accept patches from many devs) are not at the liberty to relicense under a different license, unless they can contact each and every dev that ever submitted more than a 1 line patch.
So I'm not sure if Olsen's words got very twisted up by the idiot journalist, or he just doesn't understand the GPL. If it is the latter case, that's kinda sad.
I've had enough abrasive sigs. Kittens are cute and fuzzy.
This is an idea that has been floating around for several years now. The GPL works by giving provisional permission to do things that you're otherwise prevented from doing by copyright law. If you do those things, then you have to accept the GPL or be in violation of copyright law. Pick your poison.
Copyright law gives IP owners the right to control (within the limits of fair use) the following actions:
The GPL version 2 grants the right to do the first three things. It does not grant the right to do the last 2. The legal idea is that a site like Google is a public performance of the software. To date no court has ruled on this notion, but at least some lawyers think that courts are likely to rule this way. Therefore the GPL version 3 would also grant the right to publically perform and display the software.
If the GPL version 3 comes out with such terms, and a court rules in the way that Stallman hopes that they will, then companies are caught on the horns of a dilemma. They are committing the wrong of publically performing a copyrighted work. Version 2 of the GPL grants them no permission to do so, so they are open to lawsuits. Much of that software comes with terms saying that they can choose to use it under the terms of version 3. If they follow the terms of version 3, then they can avoid those potential lawsuits.
This is, incidentally, a fundamental flaw in the GPL as it is currently constructed. Changing laws and interpretations for copyright may lead to people needing to seek permissions not granted in the GPL. There is no good "upgrade" mechanism in the license as it stands. (Unless you've done what the FSF wants and given the FSF the ability to come out with a new license that people can choose to accept. But in that case your gesture will be likely to be used as the FSF wants, not as you want. As in this instance.)
The FSF board consists of respected people like Eben Moglen and Larry Lessig. They aren't going to allow the betrayal of the FSF.
Copyright law doesn't cover use of the software - only distribution. If someone is not distributing the software, copyright law for the software does not apply. So how can you make a service provider release code for something they aren't distributing?
For this to work as described, the user would have had to sign a contract. Otherwise, it will be just as unenforceable as a EULA (you already had a right to click the Next button on your own computer, etc).
It specifically states there is no restriction on running the program, or what you can do with the output.
Changing from a distribution license to a usage license is a VERY significant change in the spirit of the license.
Quote Last para, Term 0
The act of running the Program is not restricted, and the output from the Program
is covered only if its contents constitute a work based on the Program
Term 9
Such new versions will be similar in spirit to the present version,
Wow, thanks FSF folks. You think the minor and stupid threat of legal action against Linux using corporations by SCO chilled use of Linux? Even talking about a change like this could do far more damage. It is incredible that they would even talk about something like this. If I'm a company like Google, why would I go forward using GPL software when the people in charge of the license talk about changing it to make me pay when I'm not redistributing my code? What's next? I run a website on Apache, I have to pay the FSF for the privilege? Sheesh.
I saw a present by Moglen at the OSDL conference and this is not consistent with what he said would happen with GPL v3. I suspect this much ado about nothing...
Although the intent behind the GPL is laudable, and the wording of the GPL in its current form achieves this goal reasonably well, it is always a bad idea to have a small cabal of minimally accountable people control such an important lever.
The GPL is a single point of control over the vast majority of the FOSS movement (~60-75% of all projects according to Wikipedia). The wording of the GPL impacts not only the fate of the FOSS movement itself, not only the fate of the work product of each individual participating in this movement, but also the fate of all the companies (large and small) that have chosen to assume the risk of depending on this software.
Given the GPL's extreme importance to such a large and growing audience, we should all take a hard look at who really controls it. The GPL is controlled by the FSF. The FSF is a 501(c)(3) non-profit; it has a board of directors who have responsibility for oversight of the President who runs FSF day-to-day (RMS is the president). The board of directors is elected by "Members" (NOTE: If you join the FSF off their website, you are an Associate Member which is a NON-VOTING position). I'm not sure how one gets to be a full Member.
Now let's not kid each other here: We all know how sketchy the oversight of a Board of Directors can be; we all know that groups of theoretical "equals" can be strongly influenced by a small number of strong personalities; and we all know that "Strong Personality" is a very accurate two word description of RMS (and I doubt Eben Moglen is far behind). I think those among us who are objective (and especially those among us who have personally interacted with RMS) can agree that RMS is on a personal jihad and he's using the force of law to achieve his ends. The question is, do we all want to participate in that same jihad?
Of course, one can respond by saying: "There is a competitive market for licenses - no one is forcing developers to use the GPL." This might be true on paper but it's not true in practice. We're software developers, not lawyers. I'm willing to bet that the vast majority of developers release their work under the GPL simply because "that's the one I've heard the most about and that's what everybody else does", and NOT because they researched other available licenses and arrived at an understanding of their implications, and especially an understanding of who controls these licenses.
So where does that leave us? We have a single pseudo-fanatic who has substantial control over the single most important component of our movement. This should scare the shit out of all of you.
Me? I'm going to either change to a non-GPL license or stay with GPLv2.
It's funny how the FOSS movement parallels Russian history: RMS (Lenin) started a revolution to overthrow proprietary software (the Czars). But the new regime became a dictatorship (Communism). It's now time for the proletarians (FOSS developers) to revolt again and achieve self-determination (free, accountable, and transparent democracy).
Whilst there seems to be a lot of alarmist over-reaction to this story, the principle behind it seems valid. The following is a real world example. crossfire is a MORPG (it can't fairly be called massive at this time, since it has dozens rather than hundreds of players). It is released under the GPL 2+ It has been around for years. Graal online is a proprietory MMORPG which charges monthly fees to access. It runs a (now heavily modified) version of the crossfire server. They took the server, tagged on a pretty front end and charged some amount each month. The client is also proprietory. As far as is known there is no stolen code in the client (this is based soley on screenshots though, since the source code to the client is hidden away). However, with the exception of the graphics, it is apparent from the forums that large parts of crossfire are still being served to these clients. Parts of it they have rebranded, but kept otherwise identical. Unfortunatly under the GPL2 there does not appear to be way to get these guys to contribute back, they have taken a Free software game, made some code tweaks and run their own server, which they base their business model around. The clients won't store much if any of the game content, but they are served it by the server. But, since the server binaries are not distributed, nor is the source to it. Unless the clients use some of the original code (and they might do, though this can't be proven), the GPL 2 is not sufficiantly powerful enough to get them to play fair. The GPL 3 however would seem to be. It's just a shame that it wasn't around 3 years ago.
the plaintiff can still attempt discovery to determine intent, this affects damages.
;)
but regardless of any actual litigation, the water has been poisoned.
here's the thing, or actually several things...
a. companies are very risk averse - litigation is expensive and potentially deadly.
b. The GPL has never been formally defined.
b1. The GPL reads like the Port Huron statement - it's not a standard legal document.
b2. This makes it nearly impossible to render a confident legal opinion on exactly what is and is not a valid condition/restriction of the GPL.
b3. the FSF is now claiming that activation of the GPL doesn't actually require redistribution.
c. The FSF is apparently declaring their intent to impose this new interpretation RETROACTIVELY.
d. as you ask, IS THIS EVEN LEGAL - probably not, but no one really knows at this point.
e. If it's not binding the GPL is invalid.
f. If it is binding the GPL is a HUGE LIABILITY.
g. The GPL is either a joke or a HUGE LIABILITY.
h. companies don't want to invest in jokes a/o HUGE LIABILITIES - though they've been known to
i. this begs the question - IS THE FSF A HUGE LIABILITY ?
j. a-i = HUGE LIABILTY = RISKY = DONT USE FREE SOFTWARE.
What I'm getting at is that this event will signal to those interests who currently invest in software that GPL software can't be trusted, there are too many legal issues, the governing authority is capricious and likely to act in bad faith, they resent commercial applications of the license, and are litigious. Whether or not this results in a court case, the use of GPL sofware will almost certainly decline.
Perhaps he's just managed to read the Affero General Public License v1 and has decided that that's the way that the GPL v3 is going to look? But apparently he hasn't already read the coverage of this rather crappy license that debian-legal gave in 2003 and then informed the FSF (and RMS), explaining that it couldn't possibly be DFSG Free, let alone satisfy the 4 freedoms?
Oh, right. Must not have actually checked all that out. Gee, does Mike Olson even use the GPL at all? Why would he be reviewing it anyway? Well, lets see: hrm... this sure looks like the 3 clause BSD license to me. Yerp. No GPL in sight at all. Ok, so someone who doesn't even use the GPL, (to my knowledge) isn't a lawyer, and isn't a prominent member of the copyleft side of the Free Software movement is reviewing a license that no one else has seen?
I mean, I can understand slashdot editors missing this bit of trivia in their rush to approve/reject a story... but surely Michael Singer at internetnews would have bothered to actually check if Mike Olson was the "insider" he was claiming himself to be?
http://www.donarmstrong.com
I am very surprised that Slashdot would actually approve such a misleading story. Misleading stories have appeared on Slashdot before, but at least when it comes to free software, the stories used to be truthful and informative. Well , this is an unfortunate exception.
First of all the story purports to tell what Eben Moglen is doing but mostly includes quotes from that Olson fellow. The obnly Moglen quote basicly says "let me finish the draft first and then we can discuss it". Then the article continues on pure speculation of what Moglen might be thinking.
Second, the slashdot blurb completely misrepresents the actual referenced article. The shashdot blurb makes it look like the main purpose of the FSF is to hit up google, yahoo, etc. for money. In reality the whole damn controversy is whether the code should be released or not.
Of course if there ever emerges a legal requirement to release the code and the code is not released the FSF might be entitled to damages -- thats just the way the US judicial system works. But that does not mean that the FSF are looking to hit up these companies for money. In every dispute so far the FSF has made it quite clear that they will forgo money damages when the code is actually released.
Now whether GPL derived code used for providing internet services should be released is an very interesting (and increasingly important) issue. However, this slashdot article completely misses the whole issue by making it look like the FSF is just trying to get rich on Google's behalf.
In theory /. threads are about the article cited in the thread. That means that we make references to parts of the article. The article states:
You are right in my reference to the article I did not say "CD and floppy disk" as is stated in the article. However, in the contxt of this /. debate, I beieve that alluding to the section of the article is sufficient. The assumption is that we all read the article.
The article seems to be about code that is being distributed by means other than floppy or cd. For example, there are conversations in this overall discussion about Google distributing fully functional turnkey computers that contain some GPL code. Stallman would want all of the code on that computer open sourced.
Now then, the next logical jump after distributed code brings us into the hornets nest of client server computing. The wording of the article really makes it seem that server side code is the next big issue on the GPL discussion block.
I realize that you are probably new to the internet technology. The Internet uses an idea called "client/server" architecture. Some people use the term "thin client". In this architectural design the main logic resides on the server. The server sends to the client small packets of HTML that the thin client (the web browser) formats and displays for the user.
When you "view source" you only see the HTML produced by the server. The main logic is hidden. HTML is pretty brain dead.
People developing server side applications have a completely different view of copyright than those who deliver their works as a deliverable product. Copyright, after all, was designed to protect distributed works. Web programmers really don't care that much about the HTML. The meat of their program is the logic hidden on the servers.
In most complex web sites, the meat of the program is the computer code that generates the HTML. For example, I could write a program that generates the first hundred thousands decimals of pi and displays it on my web site. Stallman would protest and say that he wants to see the code, and not the result. The code exists solely on my computer, I never distributed it and can claim exemption.
Now, I suspect that the article in question is about closing the loophole that allowed for certain means of distributing compiled computer code without the source. The real battle will come down the road when GPL wants to demand the publishing of all server side code. This would, of course, include all of the business logic so carefully guarded by businesses.
The question of closing a few loopholes that might allow the distribution of compiled code is one issue. The demand to see all code that exists on computers that contain GPL code will be another can of worms altogether.
Having developed programs with object oriented techniques. I know that there really is not a clear delineation between code and data. Code is just a bunch of data. Data really only has meaning within the context of the logic that interprets it, and so on. It is not difficult to escalate the demand to see the code on a computer to a demand to see all of the data as well.
Stallman disciples will say that the demand for full access to computers only applies to servers. To make life even more interesting, every computer on the internet has the potentional to function as a server. Well, you can take this argument where ever you like.
It's "voila"