Slashdot Mirror
Proprietary Blobs and the Pursuit of a Free Kernel
jammag writes "Ever since the GNewSense team pointed out that the Linux kernel contains proprietary firmware blobs, the question of whether a given distro is truly free software has gotten messier, notes Linux pundit Bruce Byfield. The FSF changed the definition of a free distribution, and a search for how to respond to this new definition is now well underway. Who wins and what solutions are implemented could have a major effect on the future of free and open source software. Debian has its own solution (by allowing users to choose their download), as do Ubuntu and Fedora (they include the offending firmware by default but make it possible to remove it). Meanwhile, the debate over firmware rages on. What resolves this issue?"
For things like wireless drivers the vendors can hide behind the FCC's restrictions and not release open source firmware for their hardware. This is among the worst forms of lazy regulation as it treats all users as criminals, shifts complexity to the masses, and results in products of lesser quality.
Get rid of the bad government policies and our computers would start working better. And we'd have more freedom, both on and off the expansion bus.
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
If the complicated parts of the drivers that they don't want us to know about were in ROM instead of binary blobs, and the drivers were very simple then it would solve the problem, because anyone could write drivers for whatever OS they want. As it is, you have to be using the operating systems that Nvidea allows you to use. I prefer not to have to wait around for device manufacturers to decide we should be able to use their hardware on a specific system.
Is it any more free than having a distro that's free but not having the freedom to run it on your hardware because it's completely useless?
I understand the moral conflict, but it's not like I could buy a complete set of open hardware, and even if I could, I'd just be compromising on a different front.
I am typing this from a Gnewsense system. I really appreciate the position Stallman holds - that the sole reason he would ever use unfree software would be to write free software to replace it. Thus, until he wrote the GNU system, he used proprietary systems and components until he could write his own free one. I am not able to go that far, but for non-work related things, I usually avoid non-free software, and even at work, I am working with Red Hat and other free software a lot of the time.
I guess I wasn't following things closely as one thing I was surprised at when I started using Debian (and later Ubuntu) was that there was no free Java out there. Gcj/gij and Kaffe are out there, but neither is at a level that can run most modern Java programs. Sun said in 2006 they were releasing Java as GPLv2, but that is still going on as far as I know. No full-featured Java means problems for packages I use like Eclipse or Vuze or Freenet.
Video players also have a lot of problems. Mplayer and Debian had a long history (of no Mplayer), but over the past two years it has been brought into Debian (but not Gnewsense). Flash videos from places like Youtube is a problem as well, I use Gnash, which can see some videos on Youtube and can't with others. It's also a whole rigmarole for me to watch Youtube videos on Gnewsense, I actually paste URLs into a shell script instead of watching them through my browser.
I figure if I'm going to put binary blobs, Java, and so forth on, I might as well being using Microsoft Vista. I agree with Stallman that a system is not 100% free if it allows an automatic method of installing non-free things. I personally think Debian, while not 100% free, is still close enough to suit myself in terms of allowing the option of installing non-free stuff. I don't use Debian any more but I can appreciate their position. With regards to Fedora and Ubuntu, I do not think the "you can remove non-free stuff if you want" argument holds water. That is a slippery slope as far as I'm concerned.
I appreciate Stallman's position very much. The problem with technical people is they tend to think very logically and practically and technically and don't really appreciate what Stallman's stance does. For every Stallman out there, there are thousands of guys in suits out there who want to see Vista, or at the very least some Suse hybrid on everyone's desk. I think we are very lucky to have Stallman around. I have to admit he has been helped by the Linus's and Debian's out there which are a little more practical, and a little less ideological (although to the average suit, they seem as ideological as Stallman). But stepping too far away to me is on a slippery slope to Vista land. It's an old story - if you can't beat it, then sue it for patent crap, start making Suse Linux/Microsoft hybrids and all of that.
While the Tivoization is allowed via GPL v2, it has been argued that it was never intended.
This is obviously not the case with the BSD license, and if it was they would have released another BSD license that fixed it.
Is it any more free than having a distro that's free but not having the freedom to run it on your hardware because it's completely useless?
Having a distro like that serves at least one practical purpose: I can use it to evaluate a given set of hardware for compatibility. That can inform future purchasing decisions.
For instance, having used Linux, I now know that I will never knowingly buy a Broadcom wireless card -- or, very likely, anything from Broadcom -- even for devices I don't plan to run Linux on.
This is just taking that one step further.
it's not like I could buy a complete set of open hardware
Actually, under certain, limited circumstances, you can. I believe the OpenMoko Freerunner was such a device.
Don't thank God, thank a doctor!
"I don't understand why people don't want others have the freedom to install proprietary software on Linux system."
Because some see OSS as a political movement, not a tool.
Personally I see it as a tool. Open source allows many different things that I could not do with closed source and, even back when it wasn't as technically sound an option it still often won because I could do what I needed with it.
As a tool many OSS projects have been a great successes - better than most would have believed ten years ago. It has been so mainly for the reasons above plus it was *really* easy for companies to adopt into their corporate structure (after all, it was not only free as in beer but free as in speech). ESR "won" in this sense.
As a political movement OSS has been an abject failure. It didn't achieve any of the goals of that drive the various founders. Some that were - hmm, not sure an actual term that fits term - but a mix of anti-corporation, anarchism, anti-capitalism and a few other political movements didn't see the fall of corporate structures. Some - which would be Stallman - didn't see a wave of community based software production where we all gave come about.
It turned out that when you give people "freedom" they often do things you do not like. Indeed their vision simply strengthened those policies and companies they were fighting.
In the end the problem with using it as a political tool is that there is still other choices. If I am going to have to choose between using an OSS product (even assuming I like the vision Stallman had) chances are I will go with Microsoft and all it's ills and have my company function instead of let it die and be "pure". Not allowing *any* non-open binary, not allowing any company that patents things you do not like to use your software, and a whole host of other things that many OSS projects are moving towards is a fine political statement - I have no issue whatsoever with someone doing that.
If you are looking for your software to also be a tool you can't do that - after all if your hammer comes with a long list of stuff you can't build my bet is that you will go spend the money to get one you can build anything with - even if the former hammer is free money wise. Indeed, few would consider the hammer that you could not build anything the hammer's maker didn't like to be "enforcing freedom".
OSS first hurdle was back when the decision was finally made to allow corporate interests to contribute with both code and direction. Many fought it but, in the end, a greatly improved set of software won out.
OSS is in the next of it's critical times where it will morph into something that can truly beat Microsoft or become a mostly hobbiest's tool. It's been building for some time - at least the last five years. It's still not to a head but it is getting there.
Personally I've of a mind that it is too late now - it will just cause a fork. Redhat, IBM, Debian, and many others will choose to keep their companies afloat over other entries and the "truly free" options will become hobbiest tools. GPL3 is pretty much as far as is going to be allowed and still be acceptable in the cooperate world (and even that one is hard to chew and has pushed a number of companies back to Microsoft).
------- Sorry about the spelling, I suffer from two problems. Dyslexia makes it difficult to spell well, lazy makes it
I do not necessarily disagree with that - though I think that scenario is now very unlikely (OSS systems have enough acceptance that it would be ... difficult for Intel to do that).
That is why you encourage Intel to Do The Right Thing. They may - one day - do so. However, if you tell them do it or else chances are they will take or else. Further when your OS can no longer run those extensions then your OS will not be run. Not even microsoft can take that hard line a stance and get away with it - I fail to see why many that have a comparatively minuscule market share think that they can.
So, lets take another scenario (which is much more likely). Intel produces a closed 16 core CPU that requires proprietary microcode. Linux vendors demand it be fully open or they refuse to support it all. Customers needing the 16 CPU core (or wanting it) have two choices: purchase MS products and have it supported or figure out how to hack it into the system yourself through unapproved patches and probably paying someone to re-write what is needed to get it to work (guess which one will be picked). Intel then releases a 32 core processor and noting that few used their last product they decide to not even support OSS at all. While yours *may* happen if I get my way, mine *will* if you get yours.
Of course, what will really happen is option three - RedHat (and several others) will ignore Stallman and do what they need to sell product. Many of the purists have somewhat woken up and have started to use what they are fighting against (licenses, patents, and such) to *force* OSS to what they want but the thing is just too easy to fork. Distro's that go the "pure" way will live only in hobby land.
Of course, that is part of why companies like Redhat are both loved and hated - they brought Linux to the commercial success that it is today but "betrayed" those political/social ideals that many in the OSS community started with. Of course, having never truly believed those (like me, they read ESR and thought that buy made a lot of sense) they didn't really betray anything, they more or less showed that one side could gain a larger market and mind share than the other (which is probably even more infuriating than an actual betrayal).
Now, of course, when HURD is ready then it will sweep the world - but until then I suspect that ESR's view of OSS will win pretty much every time it comes in conflict with Stallman's.
------- Sorry about the spelling, I suffer from two problems. Dyslexia makes it difficult to spell well, lazy makes it
I can't believe a totally illogical comment like yours is "+5, insightful". There's no censorship to not recommending a distro as "free" software when said distro itself recommends non-free software. It's only a matter of policy for whom and what the FSF wants to recommend. That's no more censorship than if Amnesty stated they would not recommend a political party that recommends torture. OH BUT THAT'S AN ATTACK ON FREEDUM OF SPEACH! No, it's not, idiot.
Fuck, this site is so full of morons that it makes me sick.
The UltraSPARC T2 is open source. You can download the Verilog for it, and with a big enough FPGA you can run it. Or you can run it in a simulator, or send it off and get it fab'd. You can take the core and incorporate it into a larger design with some custom accelerators (it has a very clean coprocessor interface for doing this) and get that fab'd. It has good performance and good performance per Watt for a lot of workloads.
Open source does not been designed by hobbyists and does not mean without commercial backing. If you make improvements to the T2 design and release your changes, then Sun may well incorporate them in the T3, or if they don't, you can get someone else to manufacture the chips for you.
I am TheRaven on Soylent News