Weekly Edition
Return to the Press page
| |||||||||||||
|
| |||||||||||||
MS to eradicate GPL, hence Linux (Register)
As a followup to the the recent CNET article about Microsoft revealing the source code
to Palladium, The Register believes that it is a
ploy to destroy the GPL and Linux.
"It, or a companion chip, will interface with some manner of PKI, current or
future, so that only 'authorized' applications may run with privileges."
(Log in to post comments)
MS to eradicate GPL, hence Linux (Register) Posted Jun 26, 2002 17:34 UTC (Wed) by zlynx (subscriber, #2285) [Link] All of the Open Source supporters seem to think this idea will be the end of the world. It isn't. It's a good idea, and it can be made to work for everyone, even Open Source.In order for this to work well, there will need to be PKI registries. I imagine companies like Verisign and Thawte will fill this role. An identity certificate would almost certainly have a reasonable cost, just like SSL certificates do now. So this isn't going to block Open Source developers. Software distributed by RedHat will be signed with a RedHat key. Just like RPM packages are now. Debian software will be signed with a Debian key, or an individual developer's key. People compiling their own software can sign with their own key. To repeat: This isn't the end of the world, and may even be a good idea!
Re:MS to eradicate GPL, hence Linux (Register) Posted Jun 26, 2002 23:51 UTC (Wed) by tony_peden (guest, #531) [Link] It's clear that you understand free-as-in-beer. Now go learn about free-as-in-freedom.
Re:MS to eradicate GPL, hence Linux (Register) Posted Jun 27, 2002 13:30 UTC (Thu) by zlynx (subscriber, #2285) [Link] This looks like a reply to my comment, but doesn't seem to apply to anything I said. If you have a point here, you'll need to use more words.
Palladium targets non-M$ operating systems! Posted Jun 27, 2002 8:47 UTC (Thu) by beejaybee (guest, #1581) [Link] You don't get it do you?One aspect of Palladium (according to my reading) is that code will not be allowed to execute without the blessing of the whatever rules are embedded in the chip. This _may_ be appropriate in critical business applications, and _may_ be desireable (from the point of view of publishers) for ensuring compliance with "digital rights management", but is it really neccessary for general computing systems? Just remember, the rules are being written by Uncle Bill's evil empire. Do you really think they will not take advantage of anything which could disadvantage competitors, by adding extra development work to "jump through hoops", having to modify free/open licences, fail compatibility with commercial systems or delay development by forcing time-wasting workarounds to be developed? M$ then try to draw teeth from complaints by announcing that they will publish Palladium's source code. This _is_ positive news in that it will help eliminate the usual M$ security compromises, but unfortunately is useless to developers - they need access to the rules database, as well as the code used to access it. Palladium does have some positive aspects, but I think we in the "free (as in spirit) software community" really need for it to be possible to turn this feature on and off on a per-process (or per-thread?) basis. Intel's climb down on the processor ID was not good enough as a hard system reset is neccessary to enable or disable this feature. Brian Beesley
Palladium targets non-M$ operating systems! Posted Jun 27, 2002 13:42 UTC (Thu) by zlynx (subscriber, #2285) [Link] *I* don't get it? Unless you're working for Microsoft, there isn't enough information available for you to "get it" either.You are assuming a worst case here, with Microsoft defining unchangable rules designed with the intent to destroy free software. I doubt this is what will happen. I think it is more likely that these rules will be defined by the operating system. Rules like: Only run code signed by company X, Y, Z. Or: Only allow this document to be modified by identity W. Documents such as a music file will probably carry rules of their own, like: Only play this music for identity W, or maybe Only play this music using a program signed by X. Is that such an awful thing? Anyway, my main point of my original comment was that this knee-jerk reaction by Open Source / Free Software is silly. It doesn't have to be bad, you just assume it will be.
Palladium targets non-M$ operating systems! Posted Jun 28, 2002 8:20 UTC (Fri) by beejaybee (guest, #1581) [Link] "I work for Microsoft" - NO, I have too much self-respect."I assume the worst case" - YES, definitely. And with good cause. When was the last time that M$ ever did anything which was not aimed at stamping out any code or service which did not result in revenue flowing directly into M$ coffers? Can you please tell me how an executable code trap in a CPU is going to be of any use for digital rights management if you can turn it off or bypass it completely by running a different OS? If you think that Palladium is designed for DRM, you have a terrible grasp on reality. Digital rights belong to the owner (or licensee, or subscribee) of the media being played, not to the hardware on which it is being played. This makes a total nonsense of protection embedded in the hardware itself. The OBVIOUS conclusion is that M$ is designing a system to prevent the hardware executing any code not writtem by M$, or a M$ licensed developer. And we all know that M$ is not going to license developers who wish to release code under the "viral" GPL. If my original reaction is incorrect, perhaps it's because the knee-jerk wasn't strong enough. The more I think about this, the scarier it seems to be. Brian Beesley
The Best of all Possible Worlds Posted Jul 1, 2002 1:27 UTC (Mon) by GreyWizard (subscriber, #1026) [Link] Yes, we live in the best of all possible worlds and we shouldn't assume that Microsoft would do anything to make life inconvenient for free software competitors. We should wait until they put the chains around our necks -- when it's too late to do anything about it -- before complaining. Seriously, pay attention. The only way a system like Palladium can provide a meaningful kind of fine grained control of digital copying to content owners is if the processor ring zero runs only "trusted" software. Want to run an unsigned kernel for any reason? Want to write an application capable of playing movies or sound? Signing with your own key won't get you in to the DRM club. Programmers will need to choose whether to give up all access to mainstream digital works or to leap through expensive beaurocratic hoops and obey the copyright priests who will be looking over their shoulders. No doubt vendors like RedHat would find ways to make properly signed Linux kernels, but this will increase the time and cost of creating new revisions dramatically. Operating systems without as much mainstream popularity will probably face yet another barrier. And that's really an optimistic view. Nothing prevents Microsoft from using patents and business pressure to obstruct Linux support. Nothing prevents copyright holders from using increasingly obscure Microsoft formats for thier offerings. Ordinary consumers are too ignorant to object and the free software community is too marginal to motivate the entertainment industry. What benefits, exactly, does this bring to ordinary people? Less control of their computers, more expensive upgrades, slower security corrections, a permanent revocation of fair use rights along with de-facto corportate ownership of bytes that never expires are just the beginning. Only a modern day Pangloss can regard this as a good thing.
|
|
||||||||||||