Become a fan of Slashdot on Facebook


Forgot your password?
Hardware Hacking Open Source Hardware Technology Build Linux News

If You're Going To Kill It, Open Source It 245

ptorrone writes "MAKE Magazine is proposing big companies like Cisco and Sony consider 'open sourcing' their failed or discontinued products. The list includes Sony's AIBO and QRIO robots, IBM's Deep Blue chess computer, Ricochet Wireless, Potenco's Pull-Cord Generator, Palm, Microsoft's SPOT Watch, CISCO Flip Camera and more. MAKE is also encouraging everyone to post about what products they'd like to see open sourced."
This discussion has been archived. No new comments can be posted.

If You're Going To Kill It, Open Source It

Comments Filter:
  • by clemdoc ( 624639 ) on Friday April 29, 2011 @08:10AM (#35973282)
    Opensource the Space Shuttle :)
    • Re:The Space Shuttle (Score:4, Interesting)

      by Rik Sweeney ( 471717 ) on Friday April 29, 2011 @08:15AM (#35973316) Homepage


      • by Richard_at_work ( 517087 ) <> on Friday April 29, 2011 @08:41AM (#35973476)

        What needs to be open sourced about Concorde? The principles are well known, its the economics that are the deal breaker. Airbus, Boeing, Lockheed, Embraer and Bombardier could all produce a supersonic civil aircraft if they so wished - but it would have such a small market, it wouldn't make financial or business sense for them to do so.

        • Re: (Score:2, Insightful)

          by Anonymous Coward

          Hence the definition of a "failed" project. Open Source it so the everyman can study it and break it appart reuse any pieces they find interesting. It's not just so someone can use it in business it's about knowledge sharing and general interest and possible unforseen resuses of technologies.

        • What needs to be open sourced about Concorde?

          The ignition keys.

        • That's taking a rather narrow view of what benefits can result from open-source. If they open sourced the whole design, who's to say what aspect of the design someone might learn something useful from in doing some other project. I'm quite certain there are parts of the design where engineers solved a particular problem in a way which could be applicable or instructive to any number of other engineers, not just aircraft engineers working on a supersonic civil aircraft. The value of open source isn't mere

    • All crypto products should obviously be open source, that'd cover many VPN solutions. Wuala should be open source for the same reason.

      • by npsimons ( 32752 ) *

        All crypto products should obviously be open source, that'd cover many VPN solutions. Wuala should be open source for the same reason.

        All crypto should be open source, but for different reasons. Schneier wrote on this a bit; unfortunately, I don't have the links at hand, but here are some quotes:

        As a cryptography and computer security expert, I have never understood the current fuss about the open source software movement. In the cryptography world, we consider open source necessary for good security; we ha

  • I thought I kept up with news for nerds, but I missed that one. When did AIBO die? Was there an outcry, like a great disturbance in the force or anything?

    • by Moryath ( 553296 )

      No, just a bunch of electronic yipping.

      Actually, it was over 5 years ago [] that it happened.

      Sony have just been jerkholes about people trying to continue to use and improve the toy they spent a buttload of money on since.

    • There have been several different models since their introduction on May 11, 1999 although AIBO was discontinued in 2006. *snip* On January 26, 2006 Sony announced that it would discontinue AIBO and several other products as of March, 2006 in Sony’s effort to make the company more profitable.

      From TFA.

  • Won't Happen (Score:5, Insightful)

    by Anonymous Coward on Friday April 29, 2011 @08:17AM (#35973332)
    Not going to happen for two reasons:
    - More often than not, technology or techniques developed from said projects are used in future or ongoing projects.
    - Only one thing worse than your project failing is releasing it in the wild and having another company or group making it successful without you.
    • Re: (Score:2, Redundant)

      by Bibz ( 849958 )

      In theory it's a good idea and would benefit everybody, but like parent said, it probably won't happen for many reasons.

      An other reason :
      - There might be some trade secrets embedded in the products

    • I think there is some room here for forced hostile takeovers. Say an open source consortium forms and a pool is created to buy a company and release its code.

      Forget old and failed stuff. I think the first target should be quickbooks.

      • Uhh, I'm afraid that you are dreaming, more than anyone else here. How 'bout a guick list of the comanies most likely to form such a consortium, who actually have the money to do forced hostile takeovers? I think the wealthiest company that is freindly to open source is IBM, but they have their own ideas on open source. Then, there's Oracle, with their Open Office and Java - oh, wait. Not really that freindly, right? Going down the list - well, there's Red Hat. Wonder how large a company they could ea

    • Re:Won't Happen (Score:5, Insightful)

      by CharlyFoxtrot ( 1607527 ) on Friday April 29, 2011 @08:48AM (#35973520)

      This and also : patented technologies used that might leave a company liable and similarly licensed technology used that cannot be open sourced. They're asking companies to take a product they are about to kill and spend a lot of money on it to go through the code weeding out anything that might expose them to lawsuits. In exchange for what, exactly ? It might be a boon to customers using legacy products but you want those using your new products, there's zero upside for companies on this.

    • by methano ( 519830 )
      I read the title differently from its intended meaning. I thought it meant that if there is a project that you want to die for sure, then open source it and everyone will sit around waiting for someone else to work on it. I suspect there are some examples of that happening also.
    • Not going to happen for two reasons: - More often than not, technology or techniques developed from said projects are used in future or ongoing projects. - Only one thing worse than your project failing is releasing it in the wild and having another company or group making it successful without you.

      Another reason - liability. If something goes wrong with a project they developed and then open sourced, they may find themselves the target of a lawsuit since they will have the deep pockets.

    • by LWATCDR ( 28044 )

      Exactly. PalmOS is probably still being used in some industrial, military, or medical device some where. It works and they see no reason to develop and debug a new one.
      There is also liability issues. For all a company may know there might be a Stupid software patent they didn't know about that they used in their code. Of course on the plus side there could be prior art in their code as well.
      A lot of the rest of the stuff just didn't make a lot of sense to me like Open sourcing the Flip? Get a CMOS camera se

    • Re:Won't Happen (Score:4, Insightful)

      by Digital Vomit ( 891734 ) on Friday April 29, 2011 @12:11PM (#35975748) Homepage Journal
      You forgot a third reason:
      - Corporations have no intention of ever fulfilling their obligation to the public domain as demanded by copyright law.
  • IP is the problem (Score:3, Insightful)

    by Anonymous Coward on Friday April 29, 2011 @08:17AM (#35973334)

    All products would most likely need an audit which would take both time and avoid any legal trouble that could happen. Something I doubt either company would do for the sake of giving people free shit. But you never know, maybe they have higher moral fiber than I think :)

  • Nope. (Score:5, Insightful)

    by cpu6502 ( 1960974 ) on Friday April 29, 2011 @08:20AM (#35973342)

    These companies don't want to compete against their own products (released to open source). They'd rather make these products disappear forever, and force customers to buy the newest gadgets.

    Basically it's the same strategy Microsoft follows when it refuses to open source Windows 3 or 95 or XP.

    • by ArcherB ( 796902 )

      These companies don't want to compete against their own products (released to open source). They'd rather make these products disappear forever, and force customers to buy the newest gadgets.

      Basically it's the same strategy Microsoft follows when it refuses to open source Windows 3 or 95 or XP.

      Exactly. Companies don't want the public to improve their old products, preventing them from buying new ones. For example, let's say Cicso opened up the software of all of their old routers. The open source community would take those routers and improve on them, giving them features only available in new routers. Now companies will upgrade their old routers instead of buying new ones.

      Also, there is a liability issue. In the example above, what if someone found a security hole by examining the software

  • by PhilHibbs ( 4537 ) <> on Friday April 29, 2011 @08:20AM (#35973344) Homepage Journal

    The company doesn't necessarily own all the rights to all the components. My dad and I wrote a BASIC interpreter for the PC in the 80s, but when we decided we wanted to release the source, we realised that Walter Bright owned the code that we had licensed to do the floating point arithmetic.

    If anyone wants to take on an MS-DOS BBC BASIC interpreter written in assembly, and fancies writing a new module to do floating point to replace the code in question, let me know and I'll talk to my dad about it again.

    • Re: (Score:2, Interesting)

      by Yvanhoe ( 564877 )
      You know what ? Release the code, give a liberal licence on your own copyrighted code, mention that the floating point was Walter Bright and that he reserves all rights to it.

      Then if someone wants to have a complete open source implementation, they will remove this code and replace it.
    • Re: (Score:3, Insightful)

      by noname444 ( 1182107 )

      Just release what you legally can. If someone is interested they can replace the floating point parts.

    • by LWATCDR ( 28044 )

      Actually it might not be too bad. Just change the calls to 8087 calls and say that you must have an FPU. The problem comes down how well the code is documented. Of course part of me is thinking just how freaking fast this would be. Using freedos on modern PC running this everything would probably fit in the L1 cache! Egads.

  • Once they are - say - 10 years old - the complete source and schematics of the ECU, as well as all other parts of the car are revealed.

  • Too many licensing issues and agreements with other companies. At least that's the excuse IBM gave against open-sourcing OS/2. Damn, linux would be amazing with a modernized WPS :(

    • OS/2 has also been renamed and is still being sold by a different at around $260 a seat. []

      If IBM was to open source OS/2, not only would Microsoft be all over them (it was, remember, a joint development effort), but they'd probably be in violation of the agreement with eComStation which allows that company to modify and continue selling the thing.

  • I thought that the .NET Micro Framework (the platform that the SPOT watch was based) is currently open source. At least, you can port it to the platform of your choice. []

    • You are correct. There are several open source companies that are selling hardware based on the .NET Micro Framework, including TinyCLR (FEZ product line) and the Netduino.

      Microsoft's official site on the .NET Micro Framework is at [].

      I've personally used it for several projects with great success... they really did a nice job on it and you can even use Visual Studio to develop for it, which makes it incredibly easy to debug as well. (Attached debugger to the hardware, for example.
  • That should be the law..

    • Nah, R&D might just be working the kinks out.

      Putting decent limits on copyrights and patents, and making it easier for bullshit patents to get the old heave ho, will go quite far by themselves.

  • Even though something is "failed or discontinued", that doesn't mean that there are a lot of patents based on it. Open sourcing some of these would probably raise the wrath of the legal departments. So I guess a lot of companies would rather decide to just sit on the stuff, instead of opening some other can of legal worms . . .

  • If you put an alternative for people to use instead of your new and improved pay-for version, you're not going to sell as much.

  • So I'm assuming it's a coincidence this story about releasing "abandoned" products was posted by timothy right after the "Nokia Outsources Symbian OS Work" story right? Wishful thinking? ;)
    • Symbian was released under the EPL (which was later changed [] to a mostly-closed license) in 2010.

      Full source code dump is available here [] and some other stuff are available here [].

  • Did Deep Blue cheat? (Score:4, Interesting)

    by Atari400 ( 1174925 ) on Friday April 29, 2011 @08:50AM (#35973532)
    I think the only way to find out would be to see if Deep Blue would make the same move again, and what the code looked like that would prompt it to do so.
    • Really dumb. The whole IBM cheated conspiracy theory came from GMs who saw one move and thought, "wow that didn't look like a computer-y move at all!" Well guess what, modern chess computers would find that sort of move easily, and the old idea of what a computer-y move looks like has been dead for a decade just in the software chess comps.

      Another thing people fail to realize is that Deep Blue was a hardware research project. IBM doesn't sell chess computers or software, and never had any interest in it. An

  • The reason these companies will never open-source even their 'failures' is because the greed is so consuming that they will squat on the IP of even the failed projects hoping to some day milk some extra cash from it.

    Case in point: the 1990s DOS game Ascendancy. It was developed by a tiny outfit named The Logic Factory; not at all Big Corporate Business even. Its source has never been released. A sequel was promised for over a decade (Duke Nukem Forever, anyone?), though it never materialized. The game e

    • Re: (Score:2, Insightful)

      by Anonymous Coward
      I actually have no problem with this. They dusted off their IP and are using it again. If they were to leave it to languish and still sent out the C&D letters, then you would have a point of it being pointless.
  • The reason a lot of these things will never be open sourced is simply because the technology is still economically viable, and will be used for other things, even if the PRODUCT involved isn't. The AIBOs and Deep Blues of the world aren't the "endgame", they're a way of getting the tires on a given technology to be kicked for a bit.

    • Bringing back dead projects mostly create software zombies, the ghosts of the old project might come back to scare you (IPs, bugs, patents, people..). It is always better to start again and build from ground up a better product, even if you need to reinvent the wheel here and there, but that is mostly for the better anyway.
  • Allow me to introduce you to the elephant in the corner that is owned by IBM - OS/2 Warp. Remember that? You know, the 32bit GUI OS that ran windows applications faster and more securely than the version of windows that was available at the same time?

    I think I just came across another ATM recently that was running a specialized version of Warp; so I guess we can't call it completely dead yet, even though IBM won't sell it for any amount of money.
  • Just because a product dies doesn't mean people stop using it.

    So if a software product is killed off, and the code made available for everyone (not just the good guys) to inspect, who pays the cost of patching any security vulnerabilities that are found as a result?

    It's not that the holes weren't there before (you never know, they may *be* the reason the product got canned), just that until it was handed to the world on a plate, there were easier vulnerbilities in other products to exploit. I have to sa

    • by ron_ivi ( 607351 )

      Sounds like a feature rather than a bug.

      By releasing the code when it goes out of support, any customers who depended on the product can hire someone to do a code-review/security-audit rather than continue using it with the holes in place.

  • The people in control of this code are living in a outdated and dismal delusion.

  • Like Competition...

    Look at Blender for example. It has became a MAJOR contender in the 3d space. the last release has taken steps that are starting to pass horribly overpriced commercial products like Maya.

    The hair and smoke simulations in Blender are just short of magical. and it's constantly getting better.

    • This, I think, is the main problem. Companies are afraid of having to compete with their own old products.

  • Yeah, like there arenâ(TM)t enough dead open source projects out there already. You may not like that comment but it's an undeniable truth.
    • by LWATCDR ( 28044 )

      There are a lot plus a lot of good projects that don't get enough help.
      For instance Firebird and PostgreSQL are both really good database projects. MySQL gets the most attention and is available on more web hosts so most projects make that the prime Database with often PostgreSQL as an after thought. [] isn't dead but is almost invisible.
      And then you have Lazarus + Freepascal which offers a very Dephi like system. It runs on Linux, Windows, and OS/X and there is a lot of cool code wr

  • by pruss ( 246395 ) on Friday April 29, 2011 @09:55AM (#35974146) Homepage
    Unfortunately, it can take a fair amount of work to properly open source a large commercial project. The commercial project may well have bits of code and other assets from various sources under various restrictive licenses and either permission would need to be obtained (which makes work for the legal department) or documentation for the restricted code would need to be written so that somebody in the company or a volunteer could do a clean-room rewrite. And even if there is in fact no such code or asset in the project, I assume due diligence would require someone at the company to go through the project carefully to make sure that they have the right to release all of it. Plus, even after all that was done, there may be issues with required proprietary build tools--though that issue could be left for the community to work around (one can release a tarball that doesn't compile and let someone try to figure it out)--and, as many people mentioned, there may be issues with patents. Last year, I tracked down and persuaded the author of the now defunct but excellent PalmOS astronomy app 2sky to release it under the GPL. But open sourcing it wasn't easy, even though this was a much smaller project than some of the ones mentioned. There were a large number of chunks of code to be rewritten because the author had obtained them under a GPL-incompatible license. And for me to be able to generate binaries and debug, I had to switch it to an open source toolchain from Codewarrior. And finally I had to reverse-engineer some of the author's database formats because he couldn't track down the documentation for them and the data needed to be updated (new daylight-savings rules, new comet data). It all works now ( []), but it was more work than I expected. The point is that to open source a large project is more work than inserting GPL notices and tarring. A company needs to make sure that everything they can't open source has been removed, and they may feel reasonably hesitant about releasing an obsolete project that doesn't successfully build. I still wish they would release. :-)
  • As the software curator at the Computer History Museum, the compromise that works most often is releasing
    code for non-commercial use. From a software preservation standpoint, it does put it in an institutional
    environment where the code can be saved and studied in the future. The most recent agreement is with PARC
    releasing the code for the Xerox Alto.

  • Companies are in the business of making money. If they can't make money from it, no one else can have it.

    Case in point MAME. MAME lets you play old arcade games (along with old console games). Some of the games haven't been available for decades and still companies like ATARI go after websites providing MAME downloads. Why? Because they have a HUGE stick up their @ss about someone else using their property.

    • Why? Because they have a HUGE stick up their @ss about someone else using their property.

      And why shouldn't they? Also, do you mind if I borrow your car and crash on your couch for a few months?

  • Companies exist to make money, what's the business case for open sourcing your failed products? Quite apart from all the other issues with proprietary pieces of technology that you might still be using, the fact that you might yourself resurrect a project if the market for it changes, and the fact you'd rather you competitors keep guessing about exactly how advanced you are.
  • This has been a particular chorus in the world of computer games.

    Quite literally, the code for some games is sitting forgotten in a drawer somewhere, "property" that will never - ever - be exploited. It's too old to be of any use whatsoever for commercial products, while there is a niche of old-time gamers who would love to port/rewrite/develop it for opensource use. But no, someone "owns" it, and can't give up the idea of squeezing that long-dried-out teat for a few more drops of wealth.

    For example, the

  • by rabun_bike ( 905430 ) on Friday April 29, 2011 @10:37AM (#35974624)
    When IBM killed OS/2 there was tremendous pressure for the company to open source the operating system. At the time, the vast majority of the banking industry ATM machines ran on OS/2. After doing some analysis IBM concluded they simply could not open source the operating system. Not because they didn't want to but because of all the 3rd party licensed technology embedded in the system that IBM did not own. Without agreements from these 3rd parties IBM concluded it was not a legal option for them to publish the source code. Even today there is pressure on IBM to open source OS/2. Conversely, one could also concluded the company has no upside to open sourcing. It would take a tremendous amount of legal and technical experience, time, and money to get all the agreements in place to put such a system in the open source domain. I would argue this would be a great treasure for researchers as well as computer scientists as well as corporate customers but IBM has different ideas. Likewise, other complex systems also are bound to many different patent and 3rd party agreements as well as internal propensity to keep secrets in house. []
  • by PPH ( 736903 ) on Friday April 29, 2011 @10:59AM (#35974944)

    I know several people who would dearly love to grep the source code of some closed source products to look for their misappropriated IP.

  • You know what I want open-sourced? Old video games. Mario, Zelda, Sonic, Ecco, Sim City 2000, Starcraft, etc.

    The publishers can remove all the artwork for so that they don't step on their own trademarks for all I care. Those games all deserve to live forever, independent of the hardware they require.

    I want Sim City 2000 native on Windows and Linux, I want Starcraft that runs natively on big screens, I want Zelda for my PC and a level editor to go with it.

"The number of Unix installations has grown to 10, with more expected." -- The Unix Programmer's Manual, 2nd Edition, June, 1972