OpenBSD Journal

Ask Undeadly: FreeBSD's deal with Intel

Contributed by deanna on from the sysctl click-to-agree=1 dept.

Nate writes in to ask:

While FreeBSD may lack the wpi driver itself, it now has its firmware. Announced today is the FreeBSD Foundation's compromise with Intel, FreeBSD has added a new mechanism in order to implement Intel's end-user licensing agreement. Because of this FreeBSD will now be authorized to redistribute the firmware images for Intel wireless cards.

Is this part of a growing acceptance of corporate interests? With the inclusion of binary blobs and its permittance of this new mechanism, FreeBSD is choosing the easy way to gain functionality, but is it to detriment of the system? Will FreeBSD soon be riddled with dozens of flags one must tweak before the system becomes usable? How will this affect OpenBSD's efforts to get these firmwares under liberal terms?

(Comments are closed)

Comments

  1. By Steve Shockley (66.28.244.19) on

    Gee, how come you never see OpenBSD announcements on PRWeb?

    Comments

    1. By Anonymous Coward (87.194.34.157) on

      > Gee, how come you never see OpenBSD announcements on PRWeb?

      Maybe because OpenBSD doesn't need to spin its compromises?

      Comments

      1. By Anonymous Coward (128.171.90.200) on

        > Maybe because OpenBSD doesn't need to spin its compromises?

        I don't think OpenBSD compromises

    2. By sthen (85.158.44.149) on

      > Gee, how come you never see OpenBSD announcements on PRWeb?

      And how come they didn't say where the driver was almost certainly going to come from?

  2. By Matthias Kilian (91.3.55.109) on

    > Is this part of a growing acceptance of corporate interests?

    IMHO, this is just another example for bending over for "convenience" and at the same time not realizing the consequences. It's not very different from accepting BLOBs or NDAs.

    All of this is just shortsighted. It may please users now, but it also encourages the vendors *not* to make their firmware freely distributable or give unrestricted access to their documentation.

    What's really sad is the fact that many people say they don't like it (BLOBs, NDAs, restrictive licenses) but in the same sentence claim that "it's better than nothing". Seen for BLOBs, seen in the NDA discussion (about Greg Kroah-Hartman's recent activities), and seen today in a german BSD webforum.

    Comments

    1. By Anonymous Coward (85.214.23.162) on

      http://www.allbsd.de/src/Kampagnen/StopBlob/StopBlob-de-poster.jpg

    2. By Daehlie Owns (216.126.226.56) on

      > > Is this part of a growing acceptance of corporate interests?
      >
      > IMHO, this is just another example for bending over for "convenience" and at the same time not realizing the consequences. It's not very different from accepting BLOBs or NDAs.
      >
      > All of this is just shortsighted. It may please users now, but it also encourages the vendors *not* to make their firmware freely distributable or give unrestricted access to their documentation.
      >
      > What's really sad is the fact that many people say they don't like it (BLOBs, NDAs, restrictive licenses) but in the same sentence claim that "it's better than nothing". Seen for BLOBs, seen in the NDA discussion (about Greg Kroah-Hartman's recent activities), and seen today in a german BSD webforum.
      >

      Time to separate reality from fantasy. Intel, being an entity for the purpose of profit, aka a business, has nothing to gain and everything to loose by giving their drivers or firmware source away for free. They paid developers to create them, hence without a return on that investment, no more business.

      That being said, hardware design is a cut-throat, beat the other guy to the paper launch business, therefore should Intel, or anyother company for that matter, give away the source of their firmware, containing trade secrets on both firmware design, and the intricate inner workings of their hardware, any company in Taiwan could take the hardware apart in no time and start building a cheaper equivalent. Like I said, nothing to gain and everything to lose.

      Conversely, these pre-compiled binaries give little or no trade secrets away and are the closest thing to a compromise you are ever going to get out of the big hardware makers. Frankly, as tiny a sliver of the market share OSS operating systems are, should we continue to preach this open-source purity dogma and refuse the pre-compiled solutions, the hardware makers are going to just stop writing anything for them entirely. Just like they used to...

      Comments

      1. By Anonymous Coward (193.63.217.208) on

        > > > Is this part of a growing acceptance of corporate interests?
        > >
        > > IMHO, this is just another example for bending over for "convenience" and at the same time not realizing the consequences. It's not very different from accepting BLOBs or NDAs.
        > >
        > > All of this is just shortsighted. It may please users now, but it also encourages the vendors *not* to make their firmware freely distributable or give unrestricted access to their documentation.
        > >
        > > What's really sad is the fact that many people say they don't like it (BLOBs, NDAs, restrictive licenses) but in the same sentence claim that "it's better than nothing". Seen for BLOBs, seen in the NDA discussion (about Greg Kroah-Hartman's recent activities), and seen today in a german BSD webforum.
        > >
        >
        > Time to separate reality from fantasy. Intel, being an entity for the purpose of profit, aka a business, has nothing to gain and everything to loose by giving their drivers or firmware source away for free. They paid developers to create them, hence without a return on that investment, no more business.

        They're not being asked for driver source at all. They're not being asked for firmware source, simply re-distribution rights of firmware binaries and documentation to interact with the firmware to write our own drivers. Their return on investment comes in sales of devices.

        > That being said, hardware design is a cut-throat, beat the other guy to the paper launch business, therefore should Intel, or anyother company for that matter, give away the source of their firmware, containing trade secrets on both firmware design, and the intricate inner workings of their hardware, any company in Taiwan could take the hardware apart in no time and start building a cheaper equivalent. Like I said, nothing to gain and everything to lose.

        Again, no firmware source is being requested. AFAIK OpenBSD wants firmware redistribution rights and documentation only for released products. Since they're released products, there's nothing to stop people buying a device and reverse engineering the whole thing anyway. It's hard to do but it can be done. Nothing at all is lost by giving OpenBSD what they want (what to put into what registers to get the device to perform operation X).

        > Conversely, these pre-compiled binaries give little or no trade secrets away and are the closest thing to a compromise you are ever going to get out of the big hardware makers. Frankly, as tiny a sliver of the market share OSS operating systems are, should we continue to preach this open-source purity dogma and refuse the pre-compiled solutions, the hardware makers are going to just stop writing anything for them entirely. Just like they used to...

        You've just acknowledged that binary firmware gives nothing away, so there can be no problems with giving OSS re-distribution rights can there? If we don't push for open source purity then no-one else will. Open source operating systems are widely used, and increasingly so especially now Vista has turned out to be a wildly over-priced dog's breakfast. We're not asking for vendors to write anything especially for us, they should have full, accurate documentation for their chipsets (OK, OK, should and do are not the same :P) so why not just release it?

      2. By Anonymous Coward (202.45.125.5) on

        Daehlie,

        > Time to separate reality from fantasy.

        None of your arguments are valid.

        > Intel, being an entity for the purpose of profit, aka a business, has nothing to gain and everything to loose by giving their drivers

        They DO give their drivers away for "free". Although the driver development costs may be factored into the final product price.

        >or firmware source away for free.

        The OpenBSD project is NOT asking for firmware SOURCE.

        > They paid developers to create them, hence without a return on that investment, no more business.

        There is NO logic here. Nothing is lost by providing programming documentation. In fact, the business gets OSS drivers written for FREE or practically FREE (if printed documentation is provided).

        The companies make money by selling hardware. The drivers are merely the required glue between that hardware and the OS. Programming documentation is not much good to anyone who has not purchased the actual hardware, which is the item being sold. So there is NO loss to the company if driver source or programming documentation is revealed.

        > That being said, hardware design is a cut-throat, beat the other guy to the paper launch business, therefore should Intel, or anyother company for that matter, give away the source of their firmware, containing trade secrets on both firmware design, and the intricate inner workings of their hardware, any company in Taiwan could take the hardware apart in no time and start building a cheaper equivalent. Like I said, nothing to gain and everything to lose.

        Like I said, the OpenBSD project are not asking for firmware source code or any trade secrets. They are just asking for programming documentation to USE the hardware. Programming documentation should not be giving away any trade secrets.

        > Conversely, these pre-compiled binaries give little or no trade secrets away and are the closest thing to a compromise you are ever going to get out of the big hardware makers.

        Odd, the OpenBSD project has made achievements beyond your "closest thing" claim, with other vendors.

        > Frankly, as tiny a sliver of the market share OSS operating systems are, should we continue to preach this open-source purity dogma and refuse the pre-compiled solutions, the hardware makers are going to just stop writing anything for them entirely. Just like they used to...

        Oh please, go play with the FreeBSD people. You clearly do not understand the issues involved with this. Please don't forcast what is going to happen, without first knowing what IS happening now.

        The OpenBSD project is merely asking for programming documentation to USE these products. Not blueprints to recreate the products. And not firmware source code.

        Please don't comment if you are only going to confuse the issue for others.

      3. By Anonymous Coward (83.5.224.179) on

        > Conversely, these pre-compiled binaries give little or no trade secrets away and are the closest thing to a compromise you are ever going to get out of the big hardware makers. Frankly, as tiny a sliver of the market share OSS operating systems are, should we continue to preach this open-source purity dogma and refuse the pre-compiled solutions, the hardware makers are going to just stop writing anything for them entirely. Just like they used to...
        >
        >
        bollocks. it's not oss dogma, it's redistribution rights for firmware that has been shunted from being onboard to being loaded.

        If self-interest is the name of the game, what exactly is the point of you supporting intel's interests? do you think they need your help? do you have share options? or do you just like to pick the stronger side as a matter of principle?

        it's ridiculous that as consumers, we allow companies to function through our purchases and then let them dictate how we can use the hardware we have bought. would you find it acceptable to buy a car that can only tank at certain predefined stations?

      4. By Anonymous Coward (69.207.171.114) on

        Do not feed the trolls.

      5. By Anonymous Coward (151.188.247.104) on

        This person is obviously a scared MCSE who is scared for his job if F/OSS platform knowledge becomes required where he works. I've met many of them before, and every one of them presents such pseudoarguments, which have already been aptly refuted by other posters here.

        Also remember that Microsoft has "suggested" to its employees that they post to sites like this. Throw enough stock at the Microserfs, they start obeying.

        lker@cmosnetworks.com

        Comments

        1. By Daehlie Owns (216.126.226.56) on

          > This person is obviously a scared MCSE who is scared for his job if F/OSS platform knowledge becomes required where he works. I've met many of them before, and every one of them presents such pseudoarguments, which have already been aptly refuted by other posters here.
          >
          > Also remember that Microsoft has "suggested" to its employees that they post to sites like this. Throw enough stock at the Microserfs, they start obeying.
          >
          > lker@cmosnetworks.com

          I manage a network of around 60 RHEL servers, OpenBSD routers, and several hundred ubuntu desktops. Having a realistic view of the place of oss does not imply swallowing all of its doctrine, or rather seeing the benefits of certain aspects while not seeing benefits in other aspects. I however ranted on about something unrelated and I realized after the first reply.

          My Apologizes.

      6. By Anonymous Coward (128.171.90.200) on

        > Intel, being an entity for the purpose of profit, aka a business, has nothing to gain and everything to loose by giving their drivers or firmware source away for free.

        Apart from the fact more people can use their devices on more systems, meaning they can sell more units and thus make more money.

        They have everything to gain and nothing to loose apart from letting everyone know just how shitty their devices really are.

      7. By Lars Hansson (203.65.246.12) lars@unet.net.ph on

        > Time to separate reality from fantasy.
        Indeed, lets do that and lets start by talking about reading comprehension, something you seem to lack.

        > Intel, being an entity for the purpose of profit, aka a business, has nothing to gain and everything to loose by giving their drivers or firmware source away for free. They paid developers to create them, hence without a return on that investment, no more business.

        NO-ONE is, and never have been, asking for Intel to open-source their firmware. Really. How hard is it to understand this simple thing? Seriously?

  3. By Duffman (82.226.59.105) on

    I think that the next OpenBSD release should use the 'samurai spirit' aka bushido for the comics.

    Puffy as a samurai: the only one that follow its way with no compromise.

    Just my 2 cents ^^

    Comments

    1. By anonymous pedro (82.135.30.22) on

      > Puffy as a samurai: the only one that follow its way with no compromise.

      it is good to remember that all samurais are gone...

      Comments

      1. By CODOR (CODOR) on

        > it is good to remember that all samurais are gone...

        So? Pufferfish don't use operating systems, either...

        Comments

        1. By anonymous pedro (82.135.30.22) on

          > So? Pufferfish don't use operating systems, either...

          you totally missed the point of my post :)

          the guy said samurais 'followed their way with no compromise'

          and i pointed out that didn't bring them any further from their destruction

      2. By Duffman (82.226.59.105) on

        > > Puffy as a samurai: the only one that follow its way with no compromise.
        >
        > it is good to remember that all samurais are gone...
        >
        >

        So ?

        Following your mind, we shouldn't have used Gaul people for the last release, they are all dead too ...

        Comments

        1. By Anonymous Coward (74.114.186.65) on

          > > > Puffy as a samurai: the only one that follow its way with no compromise.
          > >
          > > it is good to remember that all samurais are gone...
          > >
          > >
          >
          > So ?
          >
          > Following your mind, we shouldn't have used Gaul people for the last release, they are all dead too ...

          Asterix and co. are still alive in the books and I'm pretty sure that's the only place the magic-potion-drinking variety of Gauls ever existed in the first place.

        2. By Anonymous Coward (82.135.31.244) on

          > Following your mind, we shouldn't have used Gaul people for the last release, they are all dead too ...

          sure, definitely. that's the whole logic behind my comment. thanks for making it even more clearer for everybody to understand.

  4. By Anonymous Coward (80.140.210.130) on

    setting a env var is very liberate

    Comments

    1. By Matthias Kilian (91.3.55.109) on

      > setting a env var is very liberate

      you mean it's liberate in the same sense that most people don't hesitate to click on "yes, i accept these conditions" without actually having read the conditions they accept?

  5. By Anonymous Coward (216.68.198.57) on

    Corportate "compliance" is a bad deal, introduces new "negotiation"
    measures. Do you want Intel or Lawyers, etc, to later require verification,
    of EULA and use of firmware? Reaks of ISP and backdoors
    to measure compliance. Disaster. Etc, etc, with whatever!

    The OSS world perhaps is too open and perhaps outsourced to corporations too much.

    DRM and TS [Trusted Systems] power is emerging, and FreeBSD is going for market share.

    Market share might be very powerful with negotiation of useability in time.
    Legal threats and enforcement seem to be the Achilles heel, hopefully
    OpenBSD will pick up more market share, being tainted less. Perhaps,
    corporatations might feel more comfortable with these legal hooks and negotiations?
    Intel's corporate power is nothing to take lightly.

    Kudos to OpenBSD team and supporters!

    Barbarians of the world, get ready, the legal riddles of source
    and power are getting ready to fight.

  6. By Anonymous Coward (203.10.110.131) on

    Is OpenBSD the only OSS OS which actually believes in the virtues of OSS?

    It makes you wonder why any of the others are even bothering. If they don't have convictons behind what they're doing, then they may as well drop everything and get into the Vista line up before current stocks run out!

    At least we can rely on OpenBSD to be true to its goals.

    Thank you OpenBSD developers and fellow users! I feel like donating some more...

  7. By Anonymous Coward (85.214.23.162) on

    > With the inclusion of binary blobs

    FWIW, these are no blobs, just firmware binaries.
    But I don't think this is a suitable way either...

    Comments

    1. By phessler (209.204.157.106) on

      > > With the inclusion of binary blobs
      >
      > FWIW, these are no blobs, just firmware binaries.
      > But I don't think this is a suitable way either...

      firmware binaries are acceptable. the firmware part means they are loaded onto the device itself rather than into the kernel. all non-trivial devices have firmware somewhere. this just makes it uploadable from the OS. the problem is not the firmware, its the invasive license that is required.

      freebsd has blobs elsewhere in the system, which is why that was mentioned.

  8. By Anonymous Coward (89.54.90.155) on

    good so you're better. rah rah. I'm missing a slew of features in OpenBSD, but it's getting better. I've been patient for 10 years now.

    Comments

    1. By djm (65.57.245.11) djm@mindrot.org on

      > good so you're better. rah rah. I'm missing a slew of features
      > in OpenBSD, but it's getting better. I've been patient for 10
      > years now.

      "patches welcome"

    2. By djm (65.57.245.11) djm@mindrot.org on

      > good so you're better. rah rah. I'm missing a slew of features
      > in OpenBSD, but it's getting better. I've been patient for 10
      > years now.

      "patches welcome"

  9. By Karol Swietlicki (89.78.4.108) on

    Gee, I love this.

    Maybe we could make it easy to make deals with the devil too?
    No need mess around with all that dirty blood anymore!

    legal.i_want_to_sell_my_soul.license_ack = 1

    See? So simple for our users!

    Comments

    1. By Cabal (Cabal) on http://www.enginuity.org/

      > Gee, I love this.
      >
      > Maybe we could make it easy to make deals with the devil too?
      > No need mess around with all that dirty blood anymore!
      >
      > legal.i_want_to_sell_my_soul.license_ack = 1
      >
      > See? So simple for our users!

      Did anyone read the article at all (besides one poster above)? These are firmware files, the same thing that is already loaded in most devices that we don't have access to. These are not binary blobs.

      Comments

      1. By Chris (24.76.100.162) on drclaw.ca

        > > Gee, I love this.
        > >
        > > Maybe we could make it easy to make deals with the devil too?
        > > No need mess around with all that dirty blood anymore!
        > >
        > > legal.i_want_to_sell_my_soul.license_ack = 1
        > >
        > > See? So simple for our users!
        >
        > Did anyone read the article at all (besides one poster above)? These are firmware files, the same thing that is already loaded in most devices that we don't have access to. These are not binary blobs.

        The fact that they're binary isn't the problem, it's the necessity for the user to say "yes" to an EULA and the lack of freedom of distribution that's stupid.

        Comments

        1. By Anonymous Coward (87.79.237.121) on

          > it's the necessity for the user to say "yes" to an EULA and the lack
          > of freedom of distribution that's stupid.

          Besides, it's stupid, the user basically said "yes" when he
          bought the device in the first place.

          Comments

          1. By sthen (85.158.44.149) on

            > > it's the necessity for the user to say "yes" to an EULA and the lack
            > > of freedom of distribution that's stupid.
            >
            > Besides, it's stupid, the user basically said "yes" when he
            > bought the device in the first place.

            They didn't buy the device, they (or maybe their employer/etc) bought a laptop which happened to have the device pre-installed.

            Ask laptop vendors for an alternative...

          2. By Anonymous Coward (66.92.65.117) on

            > Besides, it's stupid, the user basically said "yes" when he
            > bought the device in the first place.

            Really? You mean the EULA is printed there on the package for all to see?

            You don't need to agree to a EULA when you don't use the vendor's drivers, and then you can do whatever you want with your hardware, which is the way it *should* be. I have this now with OpenBSD. Why would I want to give it up?

  10. By Ian McWilliam (220.240.54.229) on

    Copyright (c) 1992-2004 The FreeBSD Project.
    Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
    The Regents of the University of California. All rights reserved.
    Copyright (c) FOREVER and EVERAFTER INXEL, TeXas InstrXXents, MicrXXoft,
    AdaXtec, AtXeros
    you name it, we own it, you don't.
    FreeBSD 6.66-SNAPFROZEN #0: Thur Mar 08 16:03:05 UTC 2007
    root at weownyou.com.au:/usr/obj/usr/src/sys/BINARYBLOB
    panic: license: not agreed. FreeBSD redefining the Free in BSD.
    at line 932 in file /usr/src/sys/kern/licenses.c
    Debugger("panic")
    panic: from debugger
    #0 doadump () at /usr/src/sys/kern/licenses.c:236
    #1 0xc050d439 in boot at
    /usr/src/sys/kern/licenses.c:370
    #2 0xc050d838 in __panic () at /usr/src/sys/kern/licenses.c:548
    #3 0xc04601c2 in db_panic () at /usr/src/sys/kern/licenses.c:453

    Comments

    1. By Anonymous Coward (201.244.209.173) on

      > Copyright (c) 1992-2004 The FreeBSD Project.
      > Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
      > The Regents of the University of California. All rights reserved.
      > Copyright (c) FOREVER and EVERAFTER INXEL, TeXas InstrXXents, MicrXXoft,
      > AdaXtec, AtXeros
      > you name it, we own it, you don't.
      > FreeBSD 6.66-SNAPFROZEN #0: Thur Mar 08 16:03:05 UTC 2007
      > root at weownyou.com.au:/usr/obj/usr/src/sys/BINARYBLOB
      > panic: license: not agreed. FreeBSD redefining the Free in BSD.
      > at line 932 in file /usr/src/sys/kern/licenses.c
      > Debugger("panic")
      > panic: from debugger
      > #0 doadump () at /usr/src/sys/kern/licenses.c:236
      > #1 0xc050d439 in boot at
      > /usr/src/sys/kern/licenses.c:370
      > #2 0xc050d838 in __panic () at /usr/src/sys/kern/licenses.c:548
      > #3 0xc04601c2 in db_panic () at /usr/src/sys/kern/licenses.c:453

      JEJE

  11. By Chris (24.76.100.162) on drclaw.ca

    This is too bad.

    By the way, would any of the trolls like to explain to me how the right to distribute binary firmware without restrictions (which is what OpenBSD wants, nothing more or less) could possibly hurt Intel's business? I'd appreciate it.

    Comments

    1. By Darrin Chandler (dwc) on http://www.stilyagin.com/darrin/

      > By the way, would any of the trolls like to explain to me how the right to distribute binary firmware without restrictions (which is what OpenBSD wants, nothing more or less) could possibly hurt Intel's business? I'd appreciate it.

      Why would Intel want to, when they can count on "Free" projects bending over and working with their horrid license as is? That's the whole problem, and why it's an issue what other projects do. The willingness of other projects to accept this bullshit undermines everyone's freedoms by removing incentives for companies like Intel to use reasonable licenses.

      Comments

      1. By Anonymous Coward (198.208.251.21) on

        > The willingness of other projects to accept this bullshit undermines everyone's freedoms by removing incentives for companies like Intel to use reasonable licenses.

        Yare yare. This is *exactly* what is at issue here.

        I don't know why the word 'Source' occurs in this page over a dozen times...

  12. By Anonymous Coward (81.146.40.192) on

    I always prefer cable to wireless - in fact, I never use wireless if I can avoid it. So I can see that this kind of sullies the purity but as a pragmatic move I can't get too worked up about it; its easy to circumvent any possible risks of the blob by using a cable. It would be sad if this type of thing became a lot more prevalent though. Such a shame that hardware manufacturers keep their hardware programming details secret, I really cant see what they have to lose by publishing. The original ibm pc is the archetypal example of how successful an open hardware architecture can be.

Latest Articles

Credits

Copyright © - Daniel Hartmeier. All rights reserved. Articles and comments are copyright their respective authors, submission implies license to publish on this web site. Contents of the archive prior to as well as images and HTML templates were copied from the fabulous original deadly.org with Jose's and Jim's kind permission. This journal runs as CGI with httpd(8) on OpenBSD, the source code is BSD licensed. undeadly \Un*dead"ly\, a. Not subject to death; immortal. [Obs.]