LWN.net Logo

Here's the DRM support I could get behind

Here's the DRM support I could get behind

Posted Dec 12, 2005 23:20 UTC (Mon) by rknop (guest, #66)
Parent article: GStreamer to support DRM

Something like DeCSS.

That brought support for the DRM on current DVDs to Linux. (And it's not as odious as the DRM that's being stacked onto future media formats and future DVD formats, but it *is* a form of DRM, even if mainly directed at player developers rather than at users.)

If we had DRM support that explicitly had the purpose of breaking the DRM so that you could play the media with fully free software, well, that could only be a good thing.

-Rob


(Log in to post comments)

Here's the DRM support I could get behind

Posted Dec 13, 2005 1:24 UTC (Tue) by drag (subscriber, #31333) [Link]

Well...

At first I was like 'ok, that's weird, but the real solution is not to give money to people that use DRM'... which is the most correct stance I figure.

But later I though, 'well, should we realy be helping these people out? (think of the RIAA/itunes/etc) If they want DRM in Linux, then let them deal with it.' They don't need any help from OSS community to make it more acceptable to people. That is what keeps those advertisers in business.

But then I remembered that DRM type stuff can actually have legitimate uses.

Like gstreamer isn't just a multimedia _playback_ it's for multimedia creation, programming, and modification. So then with gstreamer-enabled programs like recorders or transcoders or whatnot then DRM has limited use for individuals protecting themselves and their privacy.

A couple for isntances are: Protecting recordings of confidential conversations. Like if you'd want a lawyer to keep a recording of meetings with yourself for criminal proceedings and be able to send those recordings to a colleague to a guy half way around the world.

Or encrypting VoIP stuff for a business... or being able to send 'sensitive' personal stuff from/to a girlfriend or stuff like that.

DRM by itself isn't going to keep it safe forever, but it's another layer of security that may be helpfull.

People use encrypted drives and loopback files for storing stuff, or used encryption in emails and such.. this is the same thing, but with audio/video.

People also use encryption for websites and plenty of free software supports that.

Now if that was to 'break' drm, then that is wonderfull... but Fluendo can't hardly support that since they'd get busted under the DMCA or at least they wouldn't be able to distribute anything in the USA and no propriatory codec creating company (Microsoft, Apple, etc) would have anything to do with it or Fluendo.

Here's the DRM support I could get behind

Posted Dec 13, 2005 2:49 UTC (Tue) by rjw (guest, #10415) [Link]

This makes no sense.

Look at what DRM actually is: A system whereby the "owner" of content can *trust* that any random machine that intercepts their data stream will be able to decrypt it, but will refuse to do things with it.

Its a security idea based on a nonsense - the idea that you can have a trusted client. DRM can *never* work even medium term : you are giving out the algorithm, the cyphertext, *and* the key. There is no way this can be secure against any vaguely capable attacker, ie people doing corporate espionage, commercial pirates, or dedicated warez groups.

What it can do is inhibit the free market in consumer electronics and software, and greatly inconvenience the average user. It will expressly not be suitable for any security other than these idiotic trusted client schemes.

Here's the DRM support I could get behind

Posted Dec 13, 2005 3:33 UTC (Tue) by drag (subscriber, #31333) [Link]

Well I never thought about it like that.

I always assumed they would finally create a sceme that worked reasonably well. Something based around hardware keying or whatnot. Or like were you are encryption a with a shared private key or something like that.

But if it is always going to be like what I say it is, then how can DRM possibly work? Are they depending fully on something like the DMCA to stop US citizens from having access to decryption tools?

The Sun Opera thingy is completely worthless?

As far as trusted computing goes.. I figure anything that allows me to have higher control over my bits and my computer is good; anything that is used that allows control by a external person/group is bad.

Here's the DRM support I could get behind

Posted Dec 14, 2005 10:34 UTC (Wed) by ekj (guest, #1524) [Link]

I always assumed they would finally create a sceme that worked reasonably well. Something based around hardware keying or whatnot. Or like were you are encryption a with a shared private key or something like that.

Most people never thougth about it. But the grandparent is rigth: DRM is, in it's essence, nonsensical. Bruce Schneier said it best: Trying to make bits non-copyable is like trying to make water not wet.

You want literally millions of devices to be able to play the DRMed media. This means they must posess all the nessecary information to decrypt it. The algorithm, *and* the key.

Yet, at the same time one assumes that *none* of these devices will ever do anything the publishers want. Display on screen ? Ok. Save unencrypted on disc ? Not ok. Output through headphones ? Ok. Output through digital-out ? Not ok. Output through headphones-connection that happens to be connected to the audio-in on another device capable of recording ? Not ok.

Notice that it's *not* enough to stop most of the people most of the time. They need to stop all of the people all of the time. If 1 in 10000 people are capable of ripping, mp3ing and uploading the newest Tittney-album, then the other 9999 people can download it like they do today, totally unaffected by the DRM. Infact they'd be *more* likely to download it than they are today because the downloaded drm-free mp3 will work on "all" devices (and give all freedoms) whereas the drmed crap will work on "many" devices and give you exactly those rigths (i.e. near none) that the publisheer wishes to give you.

Today, a good rip and an original audio-cd is equally good. (for most practical purposes anyway) In the drm-world the good rip is a lot *better* than the original plastic-thing-that-looks-like-a-cd-but-isn't.

Here's the DRM support I could get behind

Posted Dec 16, 2005 9:06 UTC (Fri) by ronaldcole (guest, #1462) [Link]

"How can DRM possibly work?" Fear, of course... fear of being plundered and pillaged by professional plunderers and pillagers: teams and teams of lawyers.

Here's the DRM support I could get behind

Posted Dec 13, 2005 4:21 UTC (Tue) by lutchann (subscriber, #8872) [Link]

DRM is not nonsense--it was never meant to protect against well-armed attackers like commercial pirates or dedicated warez groups. It only has to be good enough to defeat the typical college student who knows enough to flash a new firmware or solder a mod chip onto a board. There are few enough commercial pirates and warez groups that they (or their distribution channels, including P2P) can be shut down with legal measures, at least in developed countries. In the long run, DRM+lawsuits will probably eliminate 99% of unauthorized copying of popular content, and that sounds awful good to the major content owners.

Software-only DRM is pretty dumb, though, if that's what you were referring to.

Not good enough then

Posted Dec 13, 2005 7:38 UTC (Tue) by man_ls (subscriber, #15091) [Link]

It only has to be good enough to defeat the typical college student [...]
Then it's no good, since anyone can connect to the internet and download instructions / software written by the non-typical geeky college student who can crack it.

You can frighten the guy so he will not make the software public, but then Ed Felten or someone like him will pick the story up, host the software and defend the case in courts. And since code is a manner of speech, and free speech must be guaranteed by the US constitution, he will win. Or a similar story in Europe.

In the end it's hard to ban free speech.

Software-only DRM is pretty dumb, though, if that's what you were referring to.
DVDs were a combination of hardware + software, and they failed. Since software is infinitely flexible, it can simulate any hardware.

Not good enough then

Posted Dec 13, 2005 11:03 UTC (Tue) by job (guest, #670) [Link]

DVDs were a combination of hardware + software, and they failed. Since software is infinitely flexible, it can simulate any hardware.

You can't simulate reading a disc with a specific gap length, for example. All you need to do to "protect" your content is to make sure other people don't build the same things (eg. putting them in jail, invading their country etc.).

In practice you probably use protection bits or encryption keys because they are easier to control with laws. Sure, that's easier to bypass in software, but you can't simulate yourself out of jail.

Quite bad

Posted Dec 13, 2005 21:34 UTC (Tue) by man_ls (subscriber, #15091) [Link]

You are right. However, these laws are just making it clear that digital restriction measures are too draconian. Putting people in jail because they distribute a prime number which unlocks a DVD you purchased is hard to justify. If only people just cared a little bit more about things around them...

Not good enough then

Posted Dec 13, 2005 16:13 UTC (Tue) by lutchann (subscriber, #8872) [Link]

anyone can connect to the internet and download instructions / software written by the non-typical geeky college student who can crack it.

Assuming a software-only crack is possible, yes. I don't think that's an assumption that will always be valid.

DVDs were a combination of hardware + software, and they failed. Since software is infinitely flexible, it can simulate any hardware.

CSS was designed, what, 10 years ago? It was implemented in dozens of software-only products and yet it took five years to break? CSA was standardized in 1994 (two years after the AHRA, in which a "copy-disallow" bit was considered state-of-the-art content protection) and it took until 2002 for even the algorithm to be fully understood. All publicly-available attacks still require a source of key material; it can't be brute-forced like CSS.

The media companies have learned their lesson, and can also rely on a decade of hardware advances to design more secure protection schemes. I don't know if I would be so smug about their inevitable failure.

Not good enough anyway

Posted Dec 13, 2005 21:29 UTC (Tue) by man_ls (subscriber, #15091) [Link]

Good point. And yet remember that secrets cannot stay secret for ever, however well you guard them. One leak is enough to break any encryption scheme; and once in the open, you cannot recall all devices and media in the field to change them. Therefore, these restrictions systems are inherently brittle.

I think legal threats (as job outlines above) are more credible. And more scary.

Not good enough anyway

Posted Dec 15, 2005 7:15 UTC (Thu) by zblaxell (subscriber, #26385) [Link]

DRM can require players to phone home (or provide an Internet connection that the GPU in your video card can use to phone home) for firmware updates on a regular basis.

Some of the proposed DRM schemes disable *authors* or *players*, not media. Some of the proposed media types require someone to sign (and therefore approve) every author's or player's key. Playback hardware (which doesn't have an unencrypted native data format, not even on the PCI bus) will refuse to accept the data unless it comes with a valid, unrevoked signature.

The idea is to force everyone trading things on the net to have their keys signed, then have video cards that can receive lists of revoked author keys, and refuse to play anything that came from someone who displeases their DRM masters.

Can't require phoning home

Posted Dec 15, 2005 15:37 UTC (Thu) by felixfix (subscriber, #242) [Link]

No way can DRM manufacturers require or even expect that all units will phone home regularly. iPod? Car system? Even trying to convince people to plug their home stereo into a phone is problematical. Can you imagine when Joe Sixpack sets up his new stereo, he won't read the manual, he'll plug it in, plug in speakers, put in a CD and hit the play button ... and nothing will happen. He's going to get mad and frustrated when he can't make it play his CD, he's going to take it back or call the 1-800 number, and he's going to really come unglued when they tell him his stereo won't play a CD unless it is connected to the Internet.

Even if only 1% of the Joe Sixpacks demand a refund, the message will come through loud and clear.

Can't require phoning home

Posted Dec 15, 2005 17:49 UTC (Thu) by zblaxell (subscriber, #26385) [Link]

Don't be so sure. An average iPod is fairly often connected to an Internet-connected PC of some kind, which can act as a conduit for control messages from Apple.

It's not actually necessary to phone home if the device listens to broadcasts from home. Plastic discs (even blank recordable media) can come with embedded directives to the players from the DRM vendors. A DVD holds 9GB of data but many movies are much smaller...each one could contain a million author/player revocation certificates. These can easily reach the home and car stereos. The car stereo can also receive updates via its radio receiver.

Can't require phoning home

Posted Dec 22, 2005 18:58 UTC (Thu) by beoba (guest, #16942) [Link]

I have difficulty seeing many users going along with "you must plug your ipod into a computer every N days or it will stop playing your music"

Can't require phoning home

Posted Dec 23, 2005 21:18 UTC (Fri) by zblaxell (subscriber, #26385) [Link]

I think it would be more like "if you want to put more music on your iPod, you must plug your iPod into a computer." At that point it can require Internet connectivity to its DRM masters since the whole point of the exercise is to transfer data from the same DRM masters to the iPod. The user experience might be "sometimes when you put new music onto your iPod, it stops playing some of the old music any more."

Of course if someone writes a free version of the software on the PC that talks to the iPod, there's no way they can fail to notice that for some reason the wire protocol to the iPod involves bouncing blocks of unintelligible bits back and forth to some Internet site that nobody has ever heard of.

On the other hand, if Apple starts embedding iPod DRM directives into standard formats (e.g. MP3) as audio watermarks, then Apple still owns the iPod unless you are a guru of stenography countermeasures.

Of course this doesn't do anything useful if the only data you put into your iPod comes from clean sources (e.g. ripped from standard audio CD's or analog sources by unencumbered DRM-free software) with no network access, but at this point you're maintaining a firewall around the iPod--if you ever stop paying attention or loan your iPod to someone who does, your iPod could escape its confinement and update itself.

Here's the DRM support I could get behind

Posted Dec 13, 2005 11:58 UTC (Tue) by rjw (guest, #10415) [Link]

Yes, the only thing that can work is the *legal measures*. This has nothing to do with DRM, or technology. Any DRM scheme, in a market without braindead laws like the DMCA, would immediately lead to multiple vendors selling hardware not respecting the scheme.

The thing that is protecting the 'content providers' is the laws, not the technology. DRM from a security standpoint is a joke. If you can play it you can copy it.

Here's the DRM support I could get behind

Posted Dec 14, 2005 10:39 UTC (Wed) by ekj (guest, #1524) [Link]

DRM is not nonsense--it was never meant to protect against well-armed attackers like commercial pirates or dedicated warez groups.

If the dedicated warez-groups can crack the drm, then the undrmed media will continue to float freely over the internet, like it does today, probably in a better anonymized network. And they've gained nothing.

There are few enough commercial pirates and warez groups that they (or their distribution channels, including P2P) can be shut down with legal measures, at least in developed countries.

First, I very much doubt this is true. Secondly, even if it was, this still wouldn't help. It would only mean that the neweest Tittney-cd became available maybe an hour later, since it'd get cracked it some non-developed country and thereafter spread over the networks exactly as it does today.

Third, you seem to assume that all people are either amateurs or comercial operations. This neglects the large group of people who have lots of skills and expertize, but no interest in doing pirating comercially. These are a lot harder to catch than the comercial pirates, because the latter can be caugth by following the money-trail.

If I *don't* require any payment, it's pretty easy to make some file available to the world in such a way that it's pretty darn hard to proove it originated from me.

Here's the DRM support I could get behind

Posted Dec 13, 2005 4:32 UTC (Tue) by rknop (guest, #66) [Link]

A couple for isntances are: Protecting recordings of confidential conversations. Like if you'd want a lawyer to keep a recording of meetings with yourself for criminal proceedings and be able to send those recordings to a colleague to a guy half way around the world.

So encrypt it with gpg and send the thing. DRM isn't going to help you here... and a fully free-software solution will.

-Rob

Here's the DRM support I could get behind

Posted Dec 13, 2005 4:42 UTC (Tue) by drag (subscriber, #31333) [Link]

good point. I fogot the drm stuff would probably be closed source.

gpg encryption is pretty much the same thing, just without the buzz-word-ness. (drm is essentially a buzzword, or so it seems to me)

I was thinking along that that line, except something that is designed as a standardized encrypted media format instead of more general encryption. Something you could stream with or something that you could play directly into a media player (hardware based or software based) without decrypting it and making multiple copies first.. and having a standardized way of dealing with keys and such.

but what your saying probably makes more sense.

Here's the DRM support I could get behind

Posted Dec 13, 2005 10:54 UTC (Tue) by job (guest, #670) [Link]

gpg encryption is pretty much the same thing, just without the buzz-word-ness

Not really. Encryption is about you hiding your data so unauthorized people can't view it.

DRM is about some people deciding what you are allowed to do with their data (and the abstract idea that they can give you some data while they retain ownership). This is typically something you do with laws. If there is encryption involved it's either just for show or something useless the lawyers require.

Here's the DRM support I could get behind

Posted Dec 31, 2005 18:20 UTC (Sat) by petegn (guest, #847) [Link]

The only form of DRM that is acceptable is one that is opensource and you have the choice to use or not personally i think it should all be scrapped instantly it only serves one purpose that is to make money for the likes of the RIAA and friends .

We should not be bieng controlled by the lokes of them We should be Controling Them and if the music industry and co dont like it we as they say "suck it in " , And lets not make a mistake here it IS the Music and Film industries that are the Main protagonists in all this crap ..

Pete .

Copyright © 2013, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds