Bad, bad DRM

Posted Sep 23, 2006 15:35 UTC (Sat) by man_ls (guest, #15091)
In reply to: Re: Kernel developers' position on GPLv3 by mingo
Parent article: Kernel developers' position on GPLv3

I value your contributions (or what little of them I gather from LWN) a lot, but I think you are way off here. Allow me to take this opportunity and answer you point by point.

One thing that made Linux so successful (over other GPL-licensed OS projects like Hurd): we do as little politics as possible.

Other GPL-licensed projects (gcc, emacs) do a lot of politics and are wildly successful. The Hurd's failure probably is more of a technical thing. Linux's success probably has more to do with good leadership.

DRM is a tool and a tool can be used in good, in evil and in neutral ways, so the answer is: it depends.

This moral relativism is good for some things; in particular, it works more or less for basic science. E.g. Richard Feynman used it to relieve his conscience in "What do you care what others think?"; the good professor had participated in the Manhattan Project, and then had to watch hundreds of thousands of people die because of his work, so it is understandable that he came up with this way of thinking. Is nuclear physics good or evil? It depends; it is a valid endeavour of the human mind, and there are good uses. It does not change that he thought the consequences were completely regretful.

This relativism does not work in other fields; most human tools have their purpose written upon them in neon letters. Are fingerprint-resistant automatic assault rifles good or evil? They are patently evil. Is the atomic bomb good or evil? They are devised to decimate whole cities; it is hard to see what good they can bring. Were the bombs dropped on Hiroshima and Nagasaki good or evil? They were utterly evil, and making them "the lesser of two huge evils" is not going to change that.

Is DRM good or evil? DRM limits what you can do with your digital devices for two purposes: imposing an artificial scarcity on digital media and therefore raising prices; complementarily, it is used to control said devices and limit what people can do with them. The laws to make them possible have proven many times to have unintended consequences, as professor Felten may tell you. DRM is eminently evil; it serves a bad purpose.

Is DRM used in support of abortion good or evil? [...] Is DRM used in the USA's atomic bombs good or evil?

It is neither good nor evil; it is absurd. Are you going to lock down pro-life propaganda films? Will you prevent the victims of your atomic bombs from doing radioactivity readings on the fissionable material you throw upon them, by selling them only DRM-laden Geiger readers?

Lets face it: some of these questions are really hard to answer

Yes, but the reason is that they are too bizarre. In a similar vein, "are atomic bombs used to avoid third-world child cannibalism good or evil?" is hard to answer too.

Hence the only practical solution is, even if you dont subscribe to this concept: dont try to dictate the moral decisions of others.

This is exactly what the GPLv2 tries to do: it tries to keep you honest by dictating some moral decisions. It does not force you to use GPL'd code; in fact it allows you to do a lot more than copyright law alone, which is almost nothing. But it sets some conditions, and when they are impossible to meet you are still free to go and get your code from somewhere else. Sometimes you can even purchase the same code with a different license. Still, if you use code under the GPL the decision of sharing modifications has been taken for you.

the GPLv2 already gives enough moral and legal background for people to build a community and great free software around - and people are already giving back much, much more to Linux and FOSS than the GPL forces us to do.

As I stated above, BSD-style licenses do not force you to contribute, and many still have excellent code, documentation and support. This is not a good argument.

In fact, trying to dictate the morals of people, like the GPLv3 i believe does (by suggesting that DRM is fundamentally evil), can be considered immoral in itself.

How come? The GPLv3 takes a position: restrictions on what software can do are evil. The GPLv2 is much the same: closed and locked-down code is evil and must be avoided. Since the only way to achieve DRM is through closed and locked-down code, it is no surprise that DRM is considered evil now; just a natural consequence.

So we must be very careful to not let our natural worry to "fight" leeches get overboard and damage the very foundation our community is built upon: freedom and fairness.

I would rather say that the community is built upon a balance; the GPLv2 states some moral goals (the famous four freedoms), and then tries to balance your freedom with some restrictions in order to achieve these goals. The GPLv3 is no different. You can disagree with those goals, but talking about generic freedom and fairness will not change that.

You sound like people who have had their moral decisions taken for them and now enjoy the freedom others have gained; but cannot cope with new situations. I think it is time we all thought about these moral decisions for our own sake. It is not just about theoretical aspects; they are practical matters that arise every time we switch on our devices, which by the way look more like computers as time goes by. Do we want those computers to work for us, or for the manufacturer? Ingo, do you like the fact that your code is used in Tivo devices? Would you like to see your code power other locked-down devices? What good is the GPL then if you cannot hack on them?

to post comments

GPL-ed projects and politics

Posted Sep 23, 2006 17:48 UTC (Sat) by mingo (subscriber, #31122) [Link] (2 responses)

Other GPL-licensed projects (gcc, emacs) do a lot of politics and are wildly successful.

Maybe my different viewpoint comes partly from the fact that i'm right in the middle of these projects. The kernel has lots of dependencies on both gcc and glibc, so we follow them with great interest and we very much want those projects to succeed. Gcc has struggled for years (commercial compilers were leagues better) because it was developed in such a political way for a long time. When the egcs and pgcc projects threatened a hard fork it has been depoliticized and gcc got alot more contribution-centric, and it is in a much better technical shape now.

For glibc i suggest you read the following announcement from Ulrich Drepper (who has contributed most of the glibc code and who has been doing this for ~10 years), from August 2001. I'd suggest for you to scroll down to the section that starts with "And now for some not so nice things": glibc 2.2.4 announcement . (This was all of course eclipsed by the sad events of 9/11.)

This stuff definitely takes some digestion, and i dont expect you to take this from me at face value, because you do seem to (honestly) believe in the opposite, but doesnt it at least raise some doubt in you, which doubt would justify some more investigation and some more pondering?

GPL-ed projects and politics

Posted Sep 24, 2006 6:45 UTC (Sun) by man_ls (guest, #15091) [Link]

doesnt it at least raise some doubt in you, which doubt would justify some more investigation and some more pondering?

Sure it does. In fact now that you say it I do not think that politics are that good in a software project. Still, there are times when people have to unite against really bad ideas.

GPL-ed projects and politics

Posted Sep 25, 2006 20:55 UTC (Mon) by k8to (guest, #15413) [Link]

Yes, Stallman can be a crazy nutter unwilling to accept outside input.
Still, the response I am seeing from you and the other kernel devs is
similarly simplistic. Because Stallman does "politics" which are bad,
changing proposed by Stallman are "political".

If you don't believe in politics, argue against the new controls being
added to GPLv3 from a practical viewpoint.

If you _do_ believe in politics, argue against the new controls being
added to GPLv3 from a political viewpoint.

Simply calling them "politics" comes off as a refusal to address them at
all.

DRM good or evil?

Posted Sep 23, 2006 18:19 UTC (Sat) by mingo (subscriber, #31122) [Link] (6 responses)

I dont think i should get into the business of trying to determine what is good, evil or silly - but you did this work for me, so my only job is to point out possible inconsistencies in your categorization:

most human tools have their purpose written upon them in neon letters. Are fingerprint-resistant automatic assault rifles good or evil? They are patently evil.

is it evil even if you found it on the street (honestly, some street gang left it there), and by accident you are attacked by a drunk maniac weilding an axe, who kills your dog with a single blow and now threatens to kill you, your wife and your son? So while i'd agree with you that the production of such a rifle is probably patently evil, actual use might still be considered "good", in special circumstances.

Is the atomic bomb good or evil? They are devised to decimate whole cities; it is hard to see what good they can bring.

Here again the answer is: "it depends". For example, which atomic bomb? The russian atomic bombs were never used against civilians, and they helped create a "balance of total mutual destruction", which resulted in no other atomic bombs being used after Hiroshima and Nagasaki. (although they were very much considered militarily) Were thus those, "defensive" atomic bombs "evil" too? Or did they save humanity from total destruction?

another question here: man would have figured out the a-bomb no matter what. If not the Manhatten project then some other effort. If you had the choice, and this discovery was inevitable, which country would you have picked to discover the atomic bomb? Nazi Germany? Communist China under the rule of Mao? The Soviet Union under the rule of Stalin? Or maybe the USA?

Were the bombs dropped on Hiroshima and Nagasaki good or evil? They were utterly evil, and making them "the lesser of two huge evils" is not going to change that.

Did they save 3-4 hundreds of thousands of american lives, at the "expense" of 200,000 japanese lives? By simple, cold-blooded extrapolation from the casualty figures of Osaka's assault, probably yes. Is it evil to pick the lesser of two incredible evils? The answer depends on your fate. The Christian religion will most likely say: "yes, it was incredibly evil, man must not kill, let God decide". Under other religions it could be considered "good".

I think even these scenarios - although you picked them - are alot less clear-cut than you suggest. The same goes for DRM. It's a tool, and its morality depends on intent and other circumstances, not on the tool itself. DRM was not invented today, it was in use for more than a decade in probably every desktop chip that you used - and the use of that type of DRM was considered totally good. (the Intel microcode upload mechanism is DRM.) DRM has also been in use in probably almost every ATM that you used in your life, for over a decade. For a totally valid and non-evil purpose too. I believe you only consider DRM "evil" because you are seeing it used for evil things in things like DVD players. But even there it's not the use of DRM that is evil, but the intent of that use: the content mafia wants to preserve its monopoly.

DRM good or evil?

Posted Sep 23, 2006 21:47 UTC (Sat) by dlang (guest, #313) [Link] (4 responses)

actually, I've seen some reports that Truman authorized the use of the atomic bomb only after receiving a report that showed that dropping the bomb was expected to cost fewer Japanese civilian lives then an invasion of Japan would (again based on the behavior of the japanese at okinawa).

if these reports are correct (and from what I know of Japan during that timeframe I tend to believe them) then the use of the Bombs qualify as definitivly Good, not Evil.

DRM good or evil?

Posted Sep 24, 2006 2:46 UTC (Sun) by ianji (guest, #40710) [Link]

The idea that atomic bombs were dropped on Hiroshima and Nagasaki in order to saves lives is
widely believed but not neccessarily true. I have also seen reports that the terms of the eventual
Japanese surrender were essentially the same terms that they had already conceded to before the
nuclear attacks, and that the US only turned them down earlier because they wanted the world to
see a demonstration of the power they posessed. For further reading check out "The Hiroshima
Myth" by John V. Denson (I haven't read it but I have read a fairly lengthy synopsis).

Arithmetic of death

Posted Sep 24, 2006 6:02 UTC (Sun) by man_ls (guest, #15091) [Link]

Suppose I tell you I've also seen some reports that Stalin only authorized those Gulags after receiving a report that leaving those people alive would only have costed many more lives (based on the behaviour of the Russians at the previous revolution). Does it make the Gulags more palatable? Is Stalin a humanitarian now?

Also, imagine the report was not true. Maybe some guy in the military slipped it to Truman because he wanted to use his new toy. I tend to think that reports made during wartime are not very trusty, but what do I know.

I cannot avoid but think that those arithmetics of death are weak as a justification.

DRM good or evil?

Posted Sep 24, 2006 13:35 UTC (Sun) by drag (guest, #31333) [Link] (1 responses)

There is a huge political benifit for some people to vilify the American's use of nuclear weapons at the end of WW2. So be sure to know were your getting your information before you jump to conlusions.

---------------------

There are a few concrete things to keep in mind before making up your mind:

Based on historical evidance there is every sign that invading the Japanese homeland would of been horrific in terms of lives and damage to both sides of the war. The only time the U.S. engaged in a land conflict on traditionally Japanese territory (as per my understanding) was during the Battle of Okinawa.

Total casuality counts were 7,373 men killed and 32,056 wounded on the American side on land and another 5,000 killed and 4,600 wounded at sea.
(According to Wikipedia it's 12,500 dead and 32,000 wounded, total British and American )

On the Japanese side they had a estimated 130,000 troops stationed on the island. By the end of the battle they had 107,000 of them killed. With a possible another 20,000 killed, but completely incinerated and unaccounted for by the American's tatics of burning out the Japanese emplacements with flame throwers. (Wikipedia says it's 110,000 dead and 7,455 captured total). Typically soldiers would rush at Americans holding live grenades rather then surrendering.

Okinawa had a civilian population of about 450,000 people. Of that by the end of the battle (according to Wikipedia) they suffered at least 150,000 in terms of 'losses'. Much of it was from people simply killing themselves to avoid the 'American barbarians'.

In comparision with after effects and radiation poisoning taken into account there were about 213,000 people died as a direct result with both atomic bombings.

-----------------------------

From Truman's perspective I think the question was much more simple:

Invade Japan and expect U.S. militiary casualties numbering easily over a hundred thousand even if it turned out to be a short land war. (and easily several times that if Japanese decided to drag it out)
OR
Drop the bomb, end the war now, and loose nothing. No ships. No planes. No americans dead, wounded, or dying.

The correct course of action would of been pretty obvious during that time. I think that is the simpliest and most logical explaination as to realy why he dropped the bomb.. and most likely the correct one.

We were definately going to occupy Japan one way or another. There was no way the Americans were going to accept a conditional surrender. No way that they'd give Japan a chance to rebuild itself outside of their complete control. Just not going to happen anymore then they'd let Germany rebuild itself outside of their countrol.

The real delima historically, as I understand it, is Russia. The Russians would of been poised for invading Japan along with the U.S.. Weither or not Japanese would of allowed the Russian army to invade Japan before surrendering is the deciding factor on weither or not the Nuclear bombs were nessicary. There would of been a lot of rascism and ancestoral stuff going on between the Russians and Japanese. (meaning it would of sucked worse for the Japanese then it did for the Eastern European countries.) But to me this is only a question realy asked in hindsight as a historical debating point.

Would of Japan done a unconditional surrender to the Americans just based on the threat of having part of their country subject to Russian occuption? Maybe, I don't think the answer is very obvious (one way or another) and it was much less obvious in 1945. Germany didn't follow this path. The allies had to fight through the entire country until they destroyed the seat of government before the Germans surrendered. And even then after that there was resistance groups that fought against the occupation for years and years before they finally gave up.

As for scaring Russians with nuclear weapons.. a demostration on a small island would of been enough. I think that it's likely the display of weapons on the Japanese was hoped to leave a big impression the Russians, but that would of been a tertiary goal. Primary being to scare Japan to surrender, and secondary to reduce enemy resources if that didn't work.

German resistance AFTER WWII?

Posted Sep 25, 2006 13:24 UTC (Mon) by morhippo (guest, #334) [Link]

Never heard of it. AFAICS, the Germans very pretty docile once they had surrendered. Could you give me a source that there has been German resistance for years?

I am German and I have never heard of such a thing.

DRM good or evil?

Posted Sep 24, 2006 6:35 UTC (Sun) by man_ls (guest, #15091) [Link]

is it evil even if you found it on the street (honestly, some street gang left it there), and by accident you are attacked by a drunk maniac weilding an axe, who kills your dog with a single blow and now threatens to kill you, your wife and your son?

What can I tell you, yes, it is evil but it can have some good uses. Not that leaving automatic assault rifles on the street is generally a good idea, with all those little kids running around and such, but if some responsible adult found it and a drunk maniac attacked said adult at the same time, it could have a good use, yes.

In my case I would probably kill the drunk maniac, the wife, the kid and some other pedestrians based on my lack of expertise. A simple, non-automatic weapon would be better. Now, before you advocate that "combat weapon training is a good thing in some situations", just give me a good old axe and make it even. After all the maniac is drunk and I am not.

Were thus those, "defensive" atomic bombs "evil" too? Or did they save humanity from total destruction?

Total destruction from bombs from the other side? Yeah, really good bombs then.

If you had the choice, and this discovery was inevitable, which country would you have picked to discover the atomic bomb? Nazi Germany? Communist China under the rule of Mao? The Soviet Union under the rule of Stalin? Or maybe the USA?

Doing simple extrapolation from casualty figures, I would answer "anyone but the USA". No other country has used it against humans. This is not some communist opinion, it is cold-blooded extrapolation.

DRM was not invented today, it was in use for more than a decade in probably every desktop chip that you used - and the use of that type of DRM was considered totally good. (the Intel microcode upload mechanism is DRM.)

Ahem. It is most definitely not DRM. There are no "rights" there to manage at all. I think you are confusing encryption and trusted computing with DRM. According to the wikipedia, TC is "an enabler for DRM", but it is not the same. I do not think that TC is evil all the way, even if Stallman does.

I believe you only consider DRM "evil" because you are seeing it used for evil things in things like DVD players.

I believe I consider DRM evil because it is the wrong solution to the wrong problem. First you have to make TC pervasive, then you have to lock down all computers all the way for DRM to work; in the process you must shut down all those pesky researchers, or any loophole will get exploited to death. Distribution of source code is not possible and deployment of new binaries is unthinkable.

When all music is locked down, it is the day that we go back to pre-CD-ROM days. We lose the convenience of digital music, and we lose fair use. So we probably stop buying music.

Bad, bad DRM

Posted Sep 23, 2006 18:40 UTC (Sat) by mingo (subscriber, #31122) [Link] (6 responses)

The GPLv2 is much the same: closed and locked-down code is evil and must be avoided. Since the only way to achieve DRM is through closed and locked-down code, it is no surprise that DRM is considered evil now; just a natural consequence.

actually, that is not what DRM does. DRM is a method for hardware to only execute software from a "trusted source". DRM by itself does not "close down" software: in the case of Tivo you still get the full source code, you still get the binaries, and those binaries will still run on the DRM-ed device (and probably on other devices). What the DRM-ed device does not allow is to run /other/ binaries. Not yours, not other people's software. It is a special-purpose appliance. The Tivo manufacturer, besides having made a decision to limit your software to 128MB of RAM, by limiting you to have access only to a single IDE port, by limiting you to not have a keyboard and by creating a small form factor which has no expansion slots, he also decided to in essence burn the whole OS into "virtual ROM". Note that this "virtual ROM" (the DRM-ed OS) can still be upgraded by the manufacturer rather cheaply, but not by you. The manufacturer never sold this box to you under the pretense that it's a general purpose computer. It's a PVR. The Tivo manufacturer _never made it intentionally hackable_ and never benefited from its hackability. They never wanted an "enthusiast" community around the Tivo - and while you might disagree with their decision, it's their decision to make. They could have burned it all on ROM as well, with easy-to-remove ROM cartridges, where the ROM cartridges have some weird physical form factor that only Tivo produces - which would be just as "unmodifiable" as a DRM-ed OS is. Furthermore, if you still want to hack your Tivo, you can solder out the flash ROM that included their DRM-ed bootloader and can probably replace it with some free BIOS. Furthermore, Tivo is not a monopoly in any market, and you can readily buy other appliances with similar functionality, or you can run MythTV on your PC.

So tell me please, what evil thing did Tivo do?

Bad, bad DRM

Posted Sep 24, 2006 14:24 UTC (Sun) by drag (guest, #31333) [Link] (5 responses)

Are you sure that is legal, replacing the BIOS like that?

I don't own one, and I don't realy want one, but isn't part of reason TiVO signs the kernel is so that it protect userland from fiddling?

Hasn't TiVO entered into legal aggrements with media companies so you can do Pay per View and stuff like that and those companies require DRM-like stuff to 'protect' their content? I mean, for instance, if I hack the board by replacing the bios and I distribute bios chips for the TiVO and this allowed users to 'unprotect' protected content then I figure this a violation of the DMCA in the united states.

Are you prepared to allow people to be arrested by hacking on machines using your code if digitally signed versions of your software are part of a DRM sceme?

If your willing then that's fine. It's your code and such. (Just like I ain't going to 'player hate' BSD license using programmers) Just as long as you realise that it's pretty likely that it will end up being illegal to hack the Linux kernel in many situations were the user owns the devices in question.

Bad, bad DRM

Posted Sep 25, 2006 15:45 UTC (Mon) by mingo (subscriber, #31122) [Link] (4 responses)

But didnt you have the noble goal to freely modify the OS so that you could learn and be free? Or was the goal of that "hacking/modification" of the Tivo to go against the wishes of the content copyright holders and to "unprotect" their stuff, and to not pay? If it's the latter then i have no sympathy for that. If it's the former, i doubt there would be many grounds for suing you. (sure, you can be sued over just about anything and the DMCA makes it particularly easy - but the content owner could hardly claim that you did actual damage to him.)

Think about it this way: we, Linux copyright holders are content owners of a valuable piece of work. Even though I dont agree with Hollywood's monopoly position and their tactics, i do believe in their freedom of licensing too.

Bad, bad DRM

Posted Sep 25, 2006 21:22 UTC (Mon) by man_ls (guest, #15091) [Link] (2 responses)

You really think you don't have the right to tinker with your Tivo or any other device that enters your house -- to learn what it does and then modify it to better suit your needs. Even if it is necessary to do things other than what those content owners will let you, for example the nefarious idea of skipping commercials.

You probably don't watch DVDs on your Linux desktop, since those content owners do not want you to. They explicitly protected their valuable content with CSS which you would have to, again explicitly, circumvent to watch your legally bought DVD on your legally bought computer. As you would not be using a sanctioned program, that would make you effectively an outlaw.

You probably don't listen to MP3 music either, since:

on Linux you would be infringing upon Fraunhoffer's valuable patent portfolio, and
even if you used Ogg Vorbis, the RIAA and friends have repeatedly stated that you do not have authorization to rip and transfer the music tracks from CD's to computers and MP3 players, and without a license you are forbidden to do so by copyright law.

So you only listen to CDs and patiently change them every 50-60 minutes. Occassionally you may play CDs in shuffle mode, just to feel a little adventurous: "Is this really allowed?".

Sorry, too much for me. Freedom to tinker is freedom to tinker. If you think "content" licensing is so important that they can limit what you do with your stuff, then this is probably why DRM does not look so evil to you. But this is precisely why some other people, like Stallman and Moglen, must do things which maybe you don't understand now, but will in some years' time when we see the consequences.

Bad, bad DRM

Posted Sep 26, 2006 6:36 UTC (Tue) by man_ls (guest, #15091) [Link]

Sorry, that was a bit patronizing. I don't know what else to say, I should have shut up.

I also think your work is great.

Bad, bad DRM

Posted Sep 26, 2006 11:03 UTC (Tue) by mingo (subscriber, #31122) [Link]

I think i repeatedly asserted that i find the actions of the content monopoly deplorable.

All that i'm trying to point out is what i already wrote about in great detail: that (unlike the anti-DRM propaganda suggests) not all uses of DRM are evil, and that instead of worrying about the effects of other people's creative works we should rather concentrate on making our body of creative works appealing enough. Trying to fight DRM that tries to protect other people's creative works is misplaced in that respect. By doing that we'll be easily handled with in the policy debate by intentionally confusing us with "pirates who want to steal pay-for content". We are fighting the wrong war in the wrong place and at the wrong time.

I find the idea that we'll suddenly find no tools at our disposal to put free software on very far-fetched. DRM used for content is cumbersome, expensive and slow to every party involved.

Bad, bad DRM

Posted Sep 26, 2006 2:34 UTC (Tue) by drag (guest, #31333) [Link]

The DMCA doesn't cover intent.

If I was to sell preflashed BIOS to users who wish to modify their Linux kernel on the TiVO it would put me in jail just as fast as if I sold it intending to allow users to steal content.

It does not matter. Just for the fact that it CAN be used to steal content is what matters.

Anyways what TiVO is doing now is technically illegal according to the GPLv2. It's implied. You guys just aren't going to call them on it, just like you not going to stop people from distributing closed source binarie in their Linux modules. It's up to you.

You can do what you want. I still love your software and appreciate what you guys are doing.

Bad, bad DRM

Posted Sep 23, 2006 21:39 UTC (Sat) by ibukanov (subscriber, #3942) [Link] (1 responses)

> Is the atomic bomb good or evil? They are devised to decimate whole cities; it is hard to see what good they can bring.

See http://en.wikipedia.org/wiki/Lake_Chagan

Bad, bad atomic bombs

Posted Sep 24, 2006 10:41 UTC (Sun) by man_ls (guest, #15091) [Link]

Given that the USSR funded a huge program to justify atomic weapon development (as did the US) and Lake Chagan is its best result, I would say it further proves my point.

Have you ever heard about construction of dams? They also create artificial water reservoirs, and the best point is: they do not turn water radioactive! You can drink it afterwards!