Slashdot Log In
RMS transcript on GPLv3, Novell/MS, Tivo and more
Posted by
Hemos
on Mon Nov 27, 2006 08:40 AM
from the years-of-adoption-and-version-battles-predicted dept.
from the years-of-adoption-and-version-battles-predicted dept.
H4x0r Jim Duggan writes "The 5th international GPLv3 conference was held in Tokyo last week. I've made and published a transcript of Stallman's talk where he described the latest on what GPLv3 will do about the MS/Novell deal, Treacherous Computing, patents, Tivo, and the other changes to the licence. While I was at it, I made a transcript of my talk from the next day where I tried to fill in some info that Richard didn't mention."
Related Stories
[+]
Stallman Absolves Novell 101 comments
A few days ago we linked the transcript of Richard Stallman's talk at the Tokyo GPLv3 meeting . Now bubulubugoth writes to point us to an analysis of what Stallman said in Tokyo. In particular, these quotes: "Microsoft has not given Novell a patent license, and thus, section 7 of the GPL version 2 does not come into play. Instead, Microsoft offered a patent license that is rather limited to Novell's customers alone." And, apparently resolving the conundrum of whether GPLv2 and GPLv3 licenses can be commingled: "There's no difficulty in having some programs in the system under GPL2 and other programs under GPL3."
[+]
Your Rights Online: What the GPLv3 Means for MS-Novell Agreement 161 comments
eldavojohn writes to mention IT Business Edge has a dry but interesting interview with a lawyer (Antoinette Tease) on the effects the GPLv3 on the Microsoft & Novell alliance. From her answers: "Unlike prior versions of the GNU General Public License (GPL), which did not address patent rights, the current draft of the GPL version 3 has several provisions that address patent rights. Section 2 states that the license to use the open source code 'terminates if you bring suit against anyone for patent infringement of any of your essential patent claims' based on any version of the open source program." She goes on to say "the GPLv3 as currently drafted would impose an obligation on Novell to somehow 'shield' its customers from patent lawsuits brought by Microsoft, or, alternatively, to make the source code publicly available..."
[+]
Your Rights Online: Perens Rains on Novell's Parade 277 comments
unum15 writes "This week is Novell's Brainshare conference. They are touting the Microsoft covenant not to sue as 'good for consumers'. However, Bruce Perens decided to take this opportunity to 'rain on Novell's parade'. Perens read a statement from RMS affirming the GPLv3 would not allow companies to enter deals like this and continue to offer GPLv3 software. Perens even goes as far as to suggest this move is an exit strategy by Novell. There are also audio and pictures of the event available."
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Full
Abbreviated
Hidden
Loading ... Please wait.

RMS is always right. (Score:5, Insightful)
But I deeply admire that since I (finally) understood that overkill is necessary to avoid doing things twice.
Re:RMS is always right. Mod parent up. (Score:5, Insightful)
Right now, software patents aren't being used against free software in any serious way, and DRM problems are localised to a few specific types of software. But that's all going to change with "trusted computing" and the inevitable war that corporations are going to wage against people who want to use free software on their machines.
Stallman is fighting back. Even if you think he's over the top, a free software fundamentalist who has gone too far in his preaching, you should still listen to him. He's talking a lot of sense, on behalf of you.
Re: (Score:3, Interesting)
Stallman is fighting back. Even if you think he's over the top, a free software fundamentalist who has gone too far in his preaching, you should still listen to him. He's talking a lot of sense, on behalf of you.
I do think he's over the top (GNU/Linux?
Re: (Score:2)
1) Everyone agreeing to a particular licence at the same time
2) Everyone agreeing to give someone (Linus? FSF?) the copyright to their
Re:RMS is always right. Mod parent up. (Score:4, Informative)
1) Everyone agreeing to a particular licence at the same time
2) Everyone agreeing to give someone (Linus? FSF?) the copyright to their submissions.
It's not really that hard. I wouldn't be surprised if 80% of the code was written by 20% of the contributors. Plus, some of the code in Linux is already GPLv3-compatible, since it's either taken from the BSDs, or it's licenced under GPLv2-or-later. Additionally, older code tends to be replaced over time. If a switch to GPLv3 became popular among today's kernel developers, and Linus imposed a policy of only accepting GPLv2-and-GPLv3-compatible patches, then I suspect that in 5 years, it would be fairly simple to replace the little bit of GPLv3-incompatible code that remains, and drop GPLv2 compatibility.
It could probably also be done in less time and/or without Linus' cooperation, but with substantially more work required.
Would it be a pain? Yes, but I'm very skeptical that it's the impossible task that you claim it to be.
Re:RMS is always right. Mod parent up. (Score:5, Insightful)
That's right... many of the people commenting on his views of DRM simply haven't thought through the real implications. They think that DRM is about stopping some kids from copying music and video.
DRM is about forcing you to run particular EXACT code, and either not running at all or refusing to talk to you if you are not. This also happens to be the essence of DRM too... so it's a happy coincidence for the pro-DRMer within the technology companies.
Treacherous Computing is hardware that is meant to allow person X to set their hardware up to refuse to run (or not cage) software that is not digitally signed by them. As you can imagine, this COULD be extremely valuable for security. HOWEVER, the collection of companies making up the Trusted Computing Group (most of the tech companies) decided that the capability to TRUST should be reserved for them. You do not get to override it. This "trust" (only running approved code) starts with the BIOS, moves up to the kernel and in stuff like Vista it then moves into the media subsystem. In a few years, it will move into applications like web browsers, or word processors. Not running Microsoft Word? No you can't open that document, sorry. Add into this nasty mix, the ability of Trusted Computing hardware to execute code IN SECRET, and you have a control freak's wet dream. Not only can you not change what your computer does, you cannot even know what it is up to. The instructions are/can be encrypted.
Imagine it this way: a prison is very useful for keeping criminals in. It's not so good if YOU are the convict. In trusted computing as it is formulated today, YOU ARE THE CONVICT. You do not have the keys. Don't let them treat you like criminals. Demand the right to control your own PC -- you paid for it.
Trusted Computing is a political issue because it is a massive power grab by the technology companies. People may think Stallman is over-reacting, but I assure you that he is not. Look into the implications properly.
Re:RMS is always right. Mod parent up. (Score:5, Insightful)
You know what? Treacherous Computing could be "fixed" by the addition of one thing: an "owner override." In other words, a mechanism to allow the owner of the machine to maintain control by having the ability to instruct the TPM to lie on his behalf. It would still allow all the nice "securing the critical systems" applications you mention, while making it useless for the totalitarian things like DRM.
So, does the specification for Treacherous Computing include an owner override? NO! Why? Because the purpose of it is not for your noble cause of "securing critical systems," but for enforcing DRM. And that's why it's evil.
Re: (Score:3, Interesting)
Then why aren't we currently holding users civilly liable for damages inflicted by their poorly configured machine? Maybe we should be.
Re:RMS is always right. Mod parent up. (Score:4, Insightful)
No, you don't understand. Treacherous Computing takes away power from you, the administrator (or owner) as well! The users won't be root, but neither will you. Microsoft will be root.
Let's use an analogy: say you're renting a house. Is it reasonable for the landlord to control the keys? I'd say yes, and I think you'd agree. Now, imagine that you own a house. Is it reasonable for the builder to control the keys? I'd say no. Do you agree?
The "cure" proposal (Score:3, Informative)
There was a bit in there about a way of handling GPL violators who want to be able to redistribute again. According to Stallman, this is because when you redistribute without following the terms of the GPL, you lose any further right you have to distribute it, even if you comply with the GPL later on. In the article, he says he's not sure where the idea originated, but I suspect it's somewhere in the KDE camp, as this is where the idea that you need "forgiveness" from the copyright holder was first seen [linuxtoday.com]. The "cure" proposal would be a way of granting that forgiveness automatically.
Re: (Score:2)
I'm just curious if the man has ever said anything good about ANY software license other t
Re:The "cure" proposal (Score:5, Informative)
It appears GPLv3 is trying to pull the best bits from other licenses into as generally applicable license as possible, rather than being specific to one software program. I doubt anyone will ever agree on how well it succeeds, but it's a good idea to try, I feel.
Re: (Score:3, Informative)
Mea Culpa.
Treacherous Computing (Score:3, Insightful)
There would need to be an incentive, say, the threat of being forced to use vi on Windows, or no technology at all, if he didn't dedicate himself sincerely to protecting shareholder value or expanding the market for the product.
Because, while I find myself admiring and agreeing with RMS quite often, I also feel that he fails to appreciate the merits of any opposing viewpoints. Experience beyond his catbird seat as chief agitator of the FSF might temper the fellow nicely.
Re:Treacherous Computing (Score:5, Informative)
Re: (Score:3, Funny)
Re: (Score:2, Informative)
Chaintech - old motherboards are not supported. No video drivers for builtin 7AIV S3 ProSavage video card. Last VIA KM133 drivers cause black sc
Re: (Score:2)
I've had my share of scanners "go brick" because the driver fell out of currency.
The point is that the demonization of companies would have more weight if, to pick from your list, RMS had actual executive experience runn
Re:Assumption (Score:4, Informative)
The whole "protect shareholder value" bit is vastly over-rated by the general run of small investors as a legal excuse. Name a successful suit by shareholders against a corporate management that didn't do something to maximize profits because it was unethical or illegal. The majority of shareholder suits lose, and that includes more vaguely reasonable ones such as the Disney severance pay to Eisner lawsuit.
I just did a simple Google lookabout for currently as yet unsettled shareholder lawsuits. Nothing fancy, just typed in "Shareholder lawsuits", and then looked to see if the case appeared still in process. There were none mentioned on the first four pages (which is where I quit) that did not involve the board having allegedly either restated earnings and projected profits (always with a big decline in projected earnings) or incurred SEC fines. There's not a single one I could find this way that is even about a board simply having picked a bad business approach, so long as they did so openly, let alone one about not having entered any ethically 'gray' areas that could have led to profits. There are no shareholder lawsuits over such decisions as not laying off personnel in a decline, not selling non-exportable tech to nations on prohibited lists or other such actions revealed (at least in this search) either. I've heard of both such cases before, and am actually surprised that I couldn't find such a gase in the quick search. Still, I think this says something useful about how uncommon such cases are, at least.
In many states, the only way a court will even proceed with any shareholder lawsuit is if there is a SEC related decision first, and in some states, these decisions have to result in fines of various minimum values. In such cases, violating the law to maximize profits is the only way to become vulnerable to lawsuits over shareholder profits. That's right, in many cases, you have to break the law in an attempt to increase profits before you can be sued for the attempt failing! There, boards that use this as an excuse are saying, in effect, they had to give in before a threat that only becomes a threat if they give in! Amazing!
In over half of lawsuits of this general type, the shareholders sueing didn't do the most basic steps required before the court would even hear the case. For one example, many consistently failed to either try to get the board of directors to act on complaints before resorting to lawsuits, or to show the board was in some way biased or not sufficiently disinterested before proceeding to litigation. Of lawsuits that make it past such hurdles, most of the remainder failed automatically when the petitioner in effect asked the court to ignore that a violation of criminal law would be required for the board of directors to be in civil compliance. This frequently resulted in the suit being dismissed with prejudice.
A few years ago, a major shareholder in the pharma industry sued over just such an issue, claiming a corporation should have violated U.S. law by bribing Chinese (PRC) officials, since the fines would have been less than the profits. The arguements advanced got his lawyer disbarred, his own LLC dissolved and him convicted of racketeering so he could never form another one (at least in most jurisdictions).
That's doubtless an extreme case, but it shows the real problem here. Some people will sue over absurd, totally meritless claims, and concoct the most bizarre legal justifications. Some people will even go to civil court and argue their own case in such a way as to make it abundantly clear they are committing criminal acts, all the while thinking they are going to win big bucks. Some people will do the equivalent of shooting their parents and then absolutely demanding that the judge must show them mercy because they are an orphan.
Tivoisation? (Score:2)
Re: (Score:2)
Re:Tivoisation? (Score:5, Insightful)
Disclaimer: I own two Tivos and a home-built MythTV box.
Re: (Score:2)
Re:Tivoisation? (Score:5, Insightful)
The whole point of the GPL - see the printer driver story [blogspot.com] - is that end-users should be able to modify AND USE software released under the license. The deal is "if you use this software in your product, you're expected to give users that right.
Tivo have taken advantage of the benefits of Linux and found a legal loop-hole with which to weasel out of their responsibilities over the code. The cost of using GPL software on your hardware offerings is that users SHOULD be able to replace it on the hardware, regardless of how you feel about that. The Xbox is irrelevant in this case - it's fair enough that Microsoft deny Linux installations on their hardware, because they haven't benefited from the GPL in the first place.
The sum of the software I've released publicly amounts to a few Bash scripts, but if I were a Linux kernel developer who had purchased a Tivo I would be LIVID that my work had been used in this way, preventing me from modifying & using my own code, now resold to me installed on hardware I'd PAID good money for!!
Sure, the GPL v2 permits Tivo to restrict what kernel can be run on their hardware, but it's an unforeseen loophole - certainly against the original spirit of the GPL - and the contract should be changed.
Stroller.
Re: (Score:3, Interesting)
Re: (Score:3, Informative)
My understanding is that (under US law, anyway -- you don't say where you and your pot are) if you draft a contract, you'r
Re: (Score:3, Interesting)
Free Systems (Score:5, Interesting)
Is that actually true? I don't use any non-free software. I know other people who don't, and other people who do. I know people who use mostly non-free systems. But what about the percentages? Time for a Slashdot poll?
How free is your software?
- Mostly free (some proprietary)
- Completely free
- Mostly proprietary (some free)
- Completely proprietary
- CowboyNeal hasn't written an OS yet
PS. The first part of the statement (There are two basically free operating systems: GNU/Linux and BSD) isn't right, there are more than two. Syllable, Haiku, ReactOS, FreeDOS,
Re: (Score:2)
Well, I have a Flash player installed, so it's not entirely free. And doesn't mplayer include proprietary libraries? And depending on how your definitions are, is any mp3 player free? I think not. Also, nVidia drivers on one box, ATI drivers on another...
Re: (Score:2)
Re: (Score:2)
Well, if you consider that even the kernel provided by Debian includes non-free firmware, and those few distros that satisfy RMS's free standards are currently marginal (Utoto and Gnewsense), I'd guess most users of GNU/Linux are in fact using mostly free
Re: (Score:2)
Re: (Score:2)
These days distros can install them automatically. (not a move I'm against. In fact I voted for it (successfully
Re: (Score:2, Informative)
``He was talking about usable operating system.''
No, he wasn't. At least, it doesn't say so in TFA.
``How ma
Do It Right the First Time (TM) (Score:2)
IANAL, but the software developer in me says: if you want to ensure that the four freedoms aren't taken awa
Re: (Score:2)
RMS (Score:5, Insightful)
I wonder sometimes how computing would be if he wouldn't be so hard fighting for digital freedom. GNU/Linux would not be here today. We would have to use mainly proprietary software. Even for your web applications it would be hard to get one free.
Even if you don't agree with RMS you always should at least read and listen what he has to say. And make arguments why you don't agree, don't simple say he is ******.
I'm very happy that once again he defends FREEDOM with GPL 3 and will make the company's who want to use copyleft software play nice with you and me. I hope we (nerds, programmers, it-users) will use GPL 3 and help RMS. It is after all in our best interest to do so.
Gosh, I thought RMS came off as totally reasonable (Score:5, Insightful)
All I can see here is him talking about cleaning up the language of the GPL so that is works better in various countries and making sure it's properly compatible with other important free licenses, like the Eclipse and Apache license. That's important stuff. Some stuff about making unclear things clear, and setting it up so that you can more easily and clearly add additional rights, such as if you are using the GPL for a font set you made you can explicitly say that documents using that font can be under any license the document creator wants (which isn't a problem really, but it makes some people nervous so you can be explicit if you want).
I'm not sure what all the ranting here about RMS not having to work for a living is coming from, except maybe jealously. He's making a decent living at doing what he loves, which I thought was what we all want. Good for him, I hope someday to have a career as successful and important as his.
Re:Very Easy... (Score:5, Interesting)
From his Wikipedia article:
Stallman maintains no permanent residence outside his office at MIT's CSAIL Lab,[28] describing himself as a "squatter" on campus.[29] He owns neither an automobile, common in pedestrian-friendly Cambridge, nor a cell phone, having stated his refusal to own a device with proprietary software.[28] Because his "research affiliate" position at MIT is unpaid,[30] he supports himself financially with speaker fees and prize money from awards he has won.
Re: (Score:2)
That's his lifestyle choice. (Gross as it may be.) Even burger flippers can get a basement apartment.
Re: (Score:2, Funny)
Re:Very Easy... (Score:5, Insightful)
As is the software running the trains and planes he rides in, the software used by the government that collects his taxes, the software that... you get the idea. It's not possible for RMS to control the software used by other people, he can only control what software he uses and encourage others to use non-proprietary software. I suspect that if he could find a bank that used only Free Software, he'd use that bank. Actually, so would I... not out of opposition to proprietary software but because a bank that is so-enlightened would probably be a good choice in other ways. I'd expect them to have better security than most other banks, for example.
He'd fight at least twice as hard. (Score:2, Insightful)
You're not forced to release your code under any of his licenses. If you're doing open source development
If you have to update your license reactionarily (Score:2)
Re: (Score:2)
Re: (Score:2)
Sure, unless there happen to be some major kernel contributors who wrote huge chunks of the kernel who are dead set against the idea, that could work.
I don't think you can call something a "fork" anymore when you have to rewrite the entire thing fro
Re: (Score:2, Informative)
You have always been able to download any version of SUSE and distribute the GPLed components according to the GPL, for the last several versions the non-oss compone
Re: (Score:2)
Re: (Score:2)
Re:What RMS should address (Score:5, Informative)
The root cause is that the GPL allows for the existence of non-free distros (Novell and RedHat are the ones I know)
Dude, Red Hat Enterprise Linux is provided as source at Red Hat's site. There is at least one large, easy to use distribution which takes those sources and rebuilds them after removing any Red Hat trademarked logos. That distribution is CentOS [centos.org]. Red Hat Enterprise Linux (RHEL) is completely free and GPL compliant. I recommend the contract-supported RHEL for customers that need that support, or need the certification on particular hardware for compliance issues and I recommend CentOS for customers that are comfortable dealing with (or paying me to deal with) a lot of extra issues.
Your statement that RHEL is non-free is thus completely false and demonstrably so.
Further RHEL, unlike SLES, is not complicated by an unclear patent-deal with Microsoft which seems to open Novell and Novell's customers to arcane legal threats due to implicit admission of the existence of infringement of Microsoft patents.
Add to this that Fedora Core is almost completely paid for by Red Hat in terms of infrastructure and developers and is also completely Free and I think that your comment if not a troll is unbelievably off-base.