Loading ...
Sorry, an error occurred while loading the content.

"Licences Wars"

Expand Messages
  • Shlomi Fish
    Hi all! (for linux-elitists : I m sorry about my last post about the KDE 4 rant. I thought it was acceptable, but it turned out to violate this list s
    Message 1 of 12 , Nov 5, 2008
    • 0 Attachment
      Hi all!

      (for linux-elitists :
      I'm sorry about my last post about the KDE 4 rant. I thought it was
      acceptable, but it turned out to violate this list's etiquette. This post is
      about something I haven't published (yet, though I intend to), and will
      contain my full thoughts in an attempt to get more input.
      )

      Software licences - what a flamatory topic. My post is going to be
      titled "Licences War" (as a tribute to "Languages Wars" -
      http://www.joelonsoftware.com/items/2006/09/01.html ), and will cover my
      opinion regarding it. The first part will explain the concepts and
      differences in ideology behind strong-copyleft-licences ("viral" ones like
      the GPL or the SleepyCat Licence), public-domain, non-copyleft licences such
      as BSDL, MITL, Apache, etc. and soft-copyleft licences in between (LGPL, the
      Artistic Licence, MPL, etc.). I'm not going to cover this explanation here,
      but I refer you to http://en.wikipedia.org/wiki/Copyleft and to other
      resources on the wikipedia.

      Now. The first thing I'll mention is that it is not entirely agreed upon which
      licences are free/open-source and which are not:

      The original Artistic Licence (authored by Larry Wall for dual-licensing perl)
      is:

      * considered non-free (and non-GPL-compatible) by the FSF

      * considered free by Debian.

      * considered non-free by RedHat.

      * considered free by Mandriva.

      * considered open-source by the OSI -
      http://www.opensource.org/licenses/artistic-license-1.0.php

      * was considered a contract in a certain judgement:

      http://lwn.net/Articles/246695/

      (Who are you going to believe?)

      The Artistic Licence was later clarified into the Clarified Artistic Licence
      which are the minimum changes to make it "free" and GPL-compatible, and now
      has a new and improved version - Artistic 2.0, which is considered free and
      GPL-compatible by everybody, and is generally preferable for newer projects.

      Moreover, the Apache 2.0 Licence was specifically phrased to be
      GPL-compatible, and is considered GPL compatible by the Apache Software
      Foundation, but the FSF concluded it was not compatible with the GPLv2
      (although it is with the GPLv3 that was published afterwards). Their legal
      teams still disagree on that matter. So who are you going to believe?

      The Creative Commons Attribution (CC-by) and Attribution-ShareAlike Licences
      (CC-by-sa) are considered free (but not GPL-compatible) by the FSF (see
      http://www.gnu.org/philosophy/license-list.html#OtherLicenses ), while the
      Debian legal team concluded that they were not free.[1]

      {{{{{{
      [1] - I should note that I have issues with the entire Debian policy of
      including only free-as-in-speech material in their distribution, regardless
      of its type. I don't feel that non-software-content should abide by the same
      rules as software, and even RMS said that computer games are morally allowed
      to have non-free graphics, sound, and plots as long as their engines are
      free:

      http://developers.slashdot.org/article.pl?sid=05/01/09/191257

      So Debian are trying to be holier than the pope here and try to coerce
      everybody into abiding by irrational rules.

      But this is besides the point of whether they are free or not.
      }}}}}}

      Another curious licence is the Affero GPL -
      http://en.wikipedia.org/wiki/Affero_General_Public_License - which aims to
      close the "Application Service Provider Loophole". What it means is that if I
      install an AGPLed program on a web-server and modify it then I must make my
      modifications public. However, the Free Software Definition (see
      http://www.gnu.org/philosophy/free-sw.html ), says that I must have "The
      freedom to study how the program works, and adapt it to your needs.". As a
      result, I personally don't consider the AGPL as free (because I may wish to
      run it on my web-server and modify it), but the FSF thinks otherwise. We have
      enough problems with the suitability of the GPL for embedded systems, that we
      don't need to kill the prospering web-apps market too.

      After I cover the controversy of the different licences, I'm starting with a
      few bad ideas of what not to do:

      1. Bad Idea No. 1: Choose a non-Open-Source Licence:
      ----------------------------------------------------

      Well, I'm not sure it is a bad idea, but choosing a licence for a non-FOSS
      program is out of the scope for what I'm trying to write. I'm not going to
      stop you from trying, but I'm not going to help you with the licence choice.
      My best advice is to consult a lawyer.

      2. Bad Idea No. 2: Choose a non-GPL-Compatible Licence:
      -------------------------------------------------------

      David A. Wheeler wrote about it here:

      http://www.dwheeler.com/essays/gpl-compatible.html

      Now, since the GPLv2 is incompatible with the GPLv3 (and vice-versa) you
      shouldn't choose GPLv2-only or GPLv3-only or GPLv3-and-above, because some
      projects may be licensed only under one of them. Also, the LGPLv3 is
      incompatible with the GPLv2, so you shouldn't choose it either.
      GPL-version-2-or-above or LGPL-version-2.1-or-above should be OK in this
      respect.

      3. Bad Idea No. 3: "Same terms as Perl"
      ---------------------------------------

      I originally wrote about the problem with the "licensed under the same terms
      as Perl itself" here:

      http://use.perl.org/~Shlomi+Fish/journal/36050

      (It was also covered here:
      http://perlbuzz.com/2008/04/the-problem-with-same-terms-as-perl-licensing.html ).

      4. Bad Idea No. 4: "Under the Public Domain"
      --------------------------------------------

      See "Public Domain" here:

      http://linuxmafia.com/kb/Licensing_and_Law/

      The concept of "Public Domain" is not recognised in many jurisdications, and
      it is not clear whether one can place a software under the public domain in
      the first place. If you're interested in making your software
      as-close-to-PD-as-possible, you should choose the X11 Licence (
      http://en.wikipedia.org/wiki/MIT_License ).

      5. Bad Idea No. 5: Use the GPL or the LGPL:
      -------------------------------------------

      The GPL/LGPL contain many additional restrictions to the concept of copyleft,
      and are very mis-understood, over-hyped, and don't maintain compatibility
      with newer versions. Even the LGPL is reportedly problematic:

      http://discuss.joelonsoftware.com/default.asp?joel.3.642822.36

      The GPL and LGPL are of more political nature than other similar FOSS
      licences, and as such should be avoided. I recommend using the SleepyCat
      licence ( http://en.wikipedia.org/wiki/Sleepycat_License ), which is a
      strong-copyleft licence, that is compatible with GPLv2 and above, instead of
      the GPL and the Artistic 2.0 (or above) licence instead of the LGPL.

      I read the GPLv2 originally once and couldn't understand it. The LGPLv2 or the
      GPLv3 would likely prove to be more problematic.

      --------

      Finally, not a statement of a bad idea in itself, but a call to use the X11
      Licence (or similar FOSS licences). That's because

      - It's simple.
      - It's easy to understand.
      - It's practically PD which is what people innocently expect.
      - Much fewer worries on who can violate your licence.
      - Still gives you protection against litigation.
      - It's GPL-compatible.
      - You can boast that your program is Public domain / BSD-style.

      ----------------------

      Regards,

      Shlomi Fish


      -----------------------------------------------------------------
      Shlomi Fish http://www.shlomifish.org/
      What Makes Software Apps High Quality - http://xrl.us/bkeuk

      Shlomi, so what are you working on? Working on a new wiki about unit testing
      fortunes in freecell? -- Ran Eilam
    • Omer Zak
      ... It is my understanding that you do not have to make your modifications available to people, who don t surf your Web site. So if you install the stuff on
      Message 2 of 12 , Nov 6, 2008
      • 0 Attachment
        On Thu, 2008-11-06 at 01:33 +0200, Shlomi Fish wrote:
        > Another curious licence is the Affero GPL -
        > http://en.wikipedia.org/wiki/Affero_General_Public_License - which aims to
        > close the "Application Service Provider Loophole". What it means is that if I
        > install an AGPLed program on a web-server and modify it then I must make my
        > modifications public.

        It is my understanding that you do not have to make your modifications
        available to people, who don't surf your Web site. So if you install
        the stuff on your own PC and allow only localhost to access it, then you
        don't have to release your modifications to the public.

        But IANAL and I may be factually mistaken about this.

        > 5. Bad Idea No. 5: Use the GPL or the LGPL:
        > -------------------------------------------
        [... snipped ...]
        > The GPL and LGPL are of more political nature than other similar FOSS
        > licences, and as such should be avoided.

        I agree that GPL and LGPL have political agenda.
        I disagree that they should be avoided because of this reason.

        In an ideal world, in which all software vendors act fairly and compete
        only over the merits of their products, politics would not be necessary.

        But in today's world, politically-charged licenses - on at least part of
        the software running the world - are essential to keep the unfair
        software vendors in check. For, without viral licenses, they can take
        over any Free Software and create closed source derivatives out of it.

        Unrelated to the above, consider the fragmentation of the Unix world in
        the 1980's vs. the natural tendency to converge in the Linux world. The
        GPL played an important role in making this convergence happen.

        > Finally, not a statement of a bad idea in itself, but a call to use the X11
        > Licence (or similar FOSS licences). That's because
        >
        > - It's simple.
        > - It's easy to understand.
        > - It's practically PD which is what people innocently expect.
        > - Much fewer worries on who can violate your licence.
        > - Still gives you protection against litigation.
        > - It's GPL-compatible.
        > - You can boast that your program is Public domain / BSD-style.

        However the X11 License still allows unfair software vendors to use your
        software to maintain their stranglehold over the software world.


        --
        Did you shave a yak today?
        My own blog is at http://www.zak.co.il/tddpirate/

        My opinions, as expressed in this E-mail message, are mine alone.
        They do not represent the official policy of any organization with which
        I may be affiliated in any way.
        WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html
      • Shlomi Fish
        Hi Omer! ... No, you are right. I should have phrased it as an on a public web-server . However, this misphrasing still doesn t make it a better licence, or
        Message 3 of 12 , Nov 6, 2008
        • 0 Attachment
          Hi Omer!

          On Thursday 06 November 2008, Omer Zak wrote:
          > On Thu, 2008-11-06 at 01:33 +0200, Shlomi Fish wrote:
          > > Another curious licence is the Affero GPL -
          > > http://en.wikipedia.org/wiki/Affero_General_Public_License - which aims
          > > to close the "Application Service Provider Loophole". What it means is
          > > that if I install an AGPLed program on a web-server and modify it then I
          > > must make my modifications public.
          >
          > It is my understanding that you do not have to make your modifications
          > available to people, who don't surf your Web site. So if you install
          > the stuff on your own PC and allow only localhost to access it, then you
          > don't have to release your modifications to the public.
          >
          > But IANAL and I may be factually mistaken about this.

          No, you are right. I should have phrased it as an "on a public web-server".
          However, this misphrasing still doesn't make it a better licence, or make it
          any more free accrding to my interpretation of the Free Software Definition.

          >
          > > 5. Bad Idea No. 5: Use the GPL or the LGPL:
          > > -------------------------------------------
          >
          > [... snipped ...]
          >
          > > The GPL and LGPL are of more political nature than other similar FOSS
          > > licences, and as such should be avoided.
          >
          > I agree that GPL and LGPL have political agenda.
          > I disagree that they should be avoided because of this reason.
          >
          > In an ideal world, in which all software vendors act fairly and compete
          > only over the merits of their products, politics would not be necessary.
          >
          > But in today's world, politically-charged licenses - on at least part of
          > the software running the world - are essential to keep the unfair
          > software vendors in check. For, without viral licenses, they can take
          > over any Free Software and create closed source derivatives out of it.

          The SleepyCat licence is a viral licence, but is much less politically charged
          than the GPL, and I didn't have a problem understanding it. As a result I
          recommend using it instead of the GPL, should you decide to use a viral
          licence.

          Similarly, the Artistic 2.0 licence is LGPL-like, and should be used instead
          of the LGPL because it's less politically charged and problematic.

          >
          > Unrelated to the above, consider the fragmentation of the Unix world in
          > the 1980's vs. the natural tendency to converge in the Linux world. The
          > GPL played an important role in making this convergence happen.

          Possibly - I'm not sure. It preceded my time. However, there's a strong
          convergence in the BSD, X11, and Apache worlds today, and they are all based
          on BSD-style licences. GPL doesn't positively prevent fragmentation, as it
          explicitly allows forking.

          >
          > > Finally, not a statement of a bad idea in itself, but a call to use the
          > > X11 Licence (or similar FOSS licences). That's because
          > >
          > > - It's simple.
          > > - It's easy to understand.
          > > - It's practically PD which is what people innocently expect.
          > > - Much fewer worries on who can violate your licence.
          > > - Still gives you protection against litigation.
          > > - It's GPL-compatible.
          > > - You can boast that your program is Public domain / BSD-style.
          >
          > However the X11 License still allows unfair software vendors to use your
          > software to maintain their stranglehold over the software world.

          So do the LGPL/Artistic/etc. licences within some legally-complying limits.
          And even the GPL can be used as long as one does not directly link to GPLed
          code. In any case, I'm not too worried about trying to undermine "unfair
          software vendors". If they pose a threat (which I don't think they do), then
          no amount of keeping my code GPLed will prevent that threat. If they don't
          pose a threat, then making my code viral will not help them and may likely
          pose a problem to many benevolent developers (also of FOSS). To quote Alan
          Kay, "Don't worry about what everybody else are doing. The best way to
          predict the future is to invent it.".

          ----------

          One thing I forgot to note is that the GPL often stands against the Hacker
          Ethics as presented by ESR in http://www.catb.org/~esr/faqs/hacker-howto.html
          (and which I agree with), that says that "No problem should ever have to be
          solved twice.". For example, the FSF now started the GNU PDF project so it
          will be a GPL version 3-licenced code because both ghostscript and xpdf are
          GPLv2-only. Or the story here - http://www.osnews.com/story/7241 that said
          that:

          {{{{{{{{
          Once before, someone had contributed a patch to add boolean operations, but
          that patch relied on a polygon clipping library provided under an
          incompatible license. There's little more frustrating than having a solution
          in hand, only to be hamstrung by legal problems. Even though it was an
          important feature for us, we regretfully postponed development of it into the
          distant future on our roadmap and proceeded with other work.
          }}}}}}}}

          Or the fact OpenBSD is now re-implementing a lot of GNU GPL code from scratch
          because they find the GPL too restrictive.

          And I should also note that often one can build upon non-GPL-like-code and not
          release the derived work as FOSS, without it causing any harm, and actually
          causing a lot of good. Which is otherwise prevented if the code is GPL.

          Regards,

          Shlomi Fish

          -----------------------------------------------------------------
          Shlomi Fish http://www.shlomifish.org/
          http://www.shlomifish.org/humour/ways_to_do_it.html

          Shlomi, so what are you working on? Working on a new wiki about unit testing
          fortunes in freecell? -- Ran Eilam
        • Amit Aronovitch
          ... Possibly all. The reason is that freeness is not a boolean parameter. Not even a single parameter real-valued scale. There are multiple criteria,
          Message 4 of 12 , Nov 6, 2008
          • 0 Attachment

            On Thu, Nov 6, 2008 at 1:33 AM, Shlomi Fish <shlomif@...> wrote:

            Now. The first thing I'll mention is that it is not entirely agreed upon which
            licences are free/open-source and which are not:

            The original Artistic Licence (authored by Larry Wall for dual-licensing perl)
            is:

            * considered non-free (and non-GPL-compatible) by the FSF

            * considered free by Debian.

            * considered non-free by RedHat.

            * considered free by Mandriva.

            * considered open-source by the OSI -
            http://www.opensource.org/licenses/artistic-license-1.0.php

            * was considered a contract in a certain judgement:

            http://lwn.net/Articles/246695/

            (Who are you going to believe?)

            Possibly all. The reason is that "freeness" is not a boolean parameter. Not even a single parameter real-valued scale. There are multiple criteria, possibly contradicting each other. Much like human rights...
            It is much more practical to talk about specific definitions, such as OSI-free/DFSG-free/FSF-free etc.
            As your examples clearly demonstrate, these do not comply to a total ordering (in the mathematical sense: http://en.wikipedia.org/wiki/Total_order ).

            You should decide how you *want* your work to be distributed and choose the license accordingly (effectively decide which criteria should take priority). CC has a nice mechanism for it.


            The Creative Commons Attribution (CC-by) and Attribution-ShareAlike Licences
            (CC-by-sa) are considered free (but not GPL-compatible) by the FSF (see
            http://www.gnu.org/philosophy/license-list.html#OtherLicenses ), while the
            Debian legal team concluded that they were not free.[1]

            Notes:
            (1) Version 3.0 CC licenses *are* considered DFSG-free.
            (2) As for 2.0 and 2.5, seems that the issue is undecided, and under negotiations (between CC and Debian-legal)
            http://wiki.debian.org/DFSGLicenses#head-67131dbee718a24e1d227819dd818784aceb9cc9
             Personally, I think this is a case of unreasonably paranoidic interpretation of the license and they *should* be considered DFSG-free.
             

            {{{{{{
            [1] - I should note that I have issues with the entire Debian policy of
            including only free-as-in-speech material in their distribution, regardless
            of its type. I don't feel that non-software-content should abide by the same
            rules as software,

            In principle, I tend to agree. However you should specify which of the 9 guidelines ( http://www.debian.org/social_contract#guidelines ) if any, could be lifted for non-software contents.
            This might not be as easy as it seems. For example, the most common (and annoying) cause for removed documentation in Debian is rejection of the GFDL. However, this can not be avoided. Because of the GFDL's reuirements on availability of opaque sources (which is a "restriction on redistribution"), one would not be able to (automatically) remove deprecated documentation packages from Debian repositories without legally violating the GFDL.
            {{{ The part that I find most annoying is not the rejection of the GFDL by itself, but the tendency of some package maintainers to *delete* the GFDL docs instead of splitting the package into "free" and "non-free" parts }}

            and even RMS said that computer games are morally allowed
            to have non-free graphics, sound, and plots as long as their engines are
            free:

            http://developers.slashdot.org/article.pl?sid=05/01/09/191257

            So Debian are trying to be holier than the pope here and try to coerce
            everybody into abiding by irrational rules.

            As I said, you can't compare Debian and the FSF with a "greater than" relation. And DFSG freeness is not strictly a question of morality. For example, mandatory clickwrap might not be immoral, but allowing it would place an unreasonable burden on Debian users.


            5. Bad Idea No. 5: Use the GPL or the LGPL:
            -------------------------------------------

            The GPL/LGPL contain many additional restrictions to the concept of copyleft,
            and are very mis-understood, over-hyped, and don't maintain compatibility
            with newer versions. Even the LGPL is reportedly problematic:

            http://discuss.joelonsoftware.com/default.asp?joel.3.642822.36

            The GPL and LGPL are of more political nature than other similar FOSS
            licences, and as such should be avoided. I recommend using the SleepyCat
            licence ( http://en.wikipedia.org/wiki/Sleepycat_License ), which is a
            strong-copyleft licence, that is compatible with GPLv2 and above, instead of
            the GPL and the Artistic 2.0 (or above) licence instead of the LGPL.

            I read the GPLv2 originally once and couldn't understand it. The LGPLv2 or the
            GPLv3 would likely prove to be more problematic.


            True - simplicity is a valid consideration for choosing the license. If it is more important to you that your *users* understand your license than the lawyers of some software corps - don't choose GPL.
            The FSF seems to take the position that it is more important to protect the users and authors from IP lawyers finding loopholes to limit usage or redistribution of the software.

            --------

            Finally, not a statement of a bad idea in itself, but a call to use the X11

            (just dont confuse with XFree86 1.1 license ;-) )


            Licence (or similar FOSS licences). That's because

            - It's simple.
            - It's easy to understand.
            - It's practically PD which is what people innocently expect.
            - Much fewer worries on who can violate your licence.
            - Still gives you protection against litigation.
            - It's GPL-compatible.
            - You can boast that your program is Public domain / BSD-style.

            Good choice in many cases.
            However, note that some people who spent a lot of time developing free software, want to ensure it is used on fair terms ( as Linus said: "I give out code, I want you to do the same."
            http://www.forbes.com/technology/2006/03/09/torvalds-linux-licensing-cz_dl_0309torvalds1.html )
            and consider this more important than the "hacker ethics rule" of problems not having to be solved twice, or the desire to increase the user base.
            Copyleft is completely legitimate and there is nothing morally wrong with it.

               AA

          • Shlomi Fish
            ... Well, generally speaking one should strive for endorsement as free (and preferably GPL-compatibility) by everybody. Otherwise, you may be facing problems.
            Message 5 of 12 , Nov 7, 2008
            • 0 Attachment
              On Friday 07 November 2008, Amit Aronovitch wrote:
              > On Thu, Nov 6, 2008 at 1:33 AM, Shlomi Fish <shlomif@...> wrote:
              > > Now. The first thing I'll mention is that it is not entirely agreed upon
              > > which
              > > licences are free/open-source and which are not:
              > >
              > > The original Artistic Licence (authored by Larry Wall for dual-licensing
              > > perl)
              > > is:
              > >
              > > * considered non-free (and non-GPL-compatible) by the FSF
              > >
              > > * considered free by Debian.
              > >
              > > * considered non-free by RedHat.
              > >
              > > * considered free by Mandriva.
              > >
              > > * considered open-source by the OSI -
              > > http://www.opensource.org/licenses/artistic-license-1.0.php
              > >
              > > * was considered a contract in a certain judgement:
              > >
              > > http://lwn.net/Articles/246695/
              > >
              > > (Who are you going to believe?)
              >
              > Possibly all. The reason is that "freeness" is not a boolean parameter. Not
              > even a single parameter real-valued scale. There are multiple criteria,
              > possibly contradicting each other. Much like human rights...
              > It is much more practical to talk about specific definitions, such as
              > OSI-free/DFSG-free/FSF-free etc.
              > As your examples clearly demonstrate, these do not comply to a total
              > ordering (in the mathematical sense:
              > http://en.wikipedia.org/wiki/Total_order ).
              >
              > You should decide how you *want* your work to be distributed and choose the
              > license accordingly (effectively decide which criteria should take
              > priority). CC has a nice mechanism for it.

              Well, generally speaking one should strive for endorsement as free (and
              preferably GPL-compatibility) by everybody. Otherwise, you may be facing
              problems. As for CC, the standard CC licences
              (CC-by/CC-by-sa/CC-by-nc-sa/CC-by-nd/etc.) are not suitable for software due
              to their attribution clause:

              http://www.gnu.org/philosophy/bsd.html

              In fact, the Creative Commons does not let you choose "software" as the type
              of artwork, when filling their form.

              >
              > > The Creative Commons Attribution (CC-by) and Attribution-ShareAlike
              > > Licences
              > > (CC-by-sa) are considered free (but not GPL-compatible) by the FSF (see
              > > http://www.gnu.org/philosophy/license-list.html#OtherLicenses ), while
              > > the Debian legal team concluded that they were not free.[1]
              >
              > Notes:
              > (1) Version 3.0 CC licenses *are* considered DFSG-free.

              I see. That's god.

              > (2) As for 2.0 and 2.5, seems that the issue is undecided, and under
              > negotiations (between CC and Debian-legal)
              > http://wiki.debian.org/DFSGLicenses#head-67131dbee718a24e1d227819dd818784ac
              >eb9cc9 Personally, I think this is a case of unreasonably paranoidic
              > interpretation of the license and they *should* be considered DFSG-free.
              >

              OK.

              > > {{{{{{
              > > [1] - I should note that I have issues with the entire Debian policy of
              > > including only free-as-in-speech material in their distribution,
              > > regardless of its type. I don't feel that non-software-content should
              > > abide by the same
              > > rules as software,
              >
              > In principle, I tend to agree. However you should specify which of the 9
              > guidelines ( http://www.debian.org/social_contract#guidelines ) if any,
              > could be lifted for non-software contents.
              > This might not be as easy as it seems. For example, the most common (and
              > annoying) cause for removed documentation in Debian is rejection of the
              > GFDL. However, this can not be avoided. Because of the GFDL's reuirements
              > on availability of opaque sources (which is a "restriction on
              > redistribution"), one would not be able to (automatically) remove
              > deprecated documentation packages from Debian repositories without legally
              > violating the GFDL. {{{ The part that I find most annoying is not the
              > rejection of the GFDL by itself, but the tendency of some package
              > maintainers to *delete* the GFDL docs instead of splitting the package into
              > "free" and "non-free" parts }}
              >

              Interesting. Wouldn't one be able to remove the packages from the Deian's
              repository and let it rot on a public (and accessible) web-site of deprecated
              packages? IANAL, so I don't know.

              > and even RMS said that computer games are morally allowed
              >
              > > to have non-free graphics, sound, and plots as long as their engines are
              > > free:
              > >
              > > http://developers.slashdot.org/article.pl?sid=05/01/09/191257
              > >
              > > So Debian are trying to be holier than the pope here and try to coerce
              > > everybody into abiding by irrational rules.
              >
              > As I said, you can't compare Debian and the FSF with a "greater than"
              > relation. And DFSG freeness is not strictly a question of morality. For
              > example, mandatory clickwrap might not be immoral, but allowing it would
              > place an unreasonable burden on Debian users.
              >

              OK.

              > > 5. Bad Idea No. 5: Use the GPL or the LGPL:
              > > -------------------------------------------
              > >
              > > The GPL/LGPL contain many additional restrictions to the concept of
              > > copyleft,
              > > and are very mis-understood, over-hyped, and don't maintain compatibility
              > > with newer versions. Even the LGPL is reportedly problematic:
              > >
              > > http://discuss.joelonsoftware.com/default.asp?joel.3.642822.36
              > >
              > > The GPL and LGPL are of more political nature than other similar FOSS
              > > licences, and as such should be avoided. I recommend using the SleepyCat
              > > licence ( http://en.wikipedia.org/wiki/Sleepycat_License ), which is a
              > > strong-copyleft licence, that is compatible with GPLv2 and above, instead
              > > of
              > > the GPL and the Artistic 2.0 (or above) licence instead of the LGPL.
              > >
              > > I read the GPLv2 originally once and couldn't understand it. The LGPLv2
              > > or the
              > > GPLv3 would likely prove to be more problematic.
              >
              > True - simplicity is a valid consideration for choosing the license. If it
              > is more important to you that your *users* understand your license than the
              > lawyers of some software corps - don't choose GPL.
              > The FSF seems to take the position that it is more important to protect the
              > users and authors from IP lawyers finding loopholes to limit usage or
              > redistribution of the software.

              Yes.

              >
              > --------
              >
              > > Finally, not a statement of a bad idea in itself, but a call to use the
              > > X11
              >
              > (just dont confuse with XFree86 1.1 license ;-) )

              Yes.

              >
              > > Licence (or similar FOSS licences). That's because
              > >
              > > - It's simple.
              > > - It's easy to understand.
              > > - It's practically PD which is what people innocently expect.
              > > - Much fewer worries on who can violate your licence.
              > > - Still gives you protection against litigation.
              > > - It's GPL-compatible.
              > > - You can boast that your program is Public domain / BSD-style.
              >
              > Good choice in many cases.
              > However, note that some people who spent a lot of time developing free
              > software, want to ensure it is used on fair terms ( as Linus said: "I give
              > out code, I want you to do the same."
              > http://www.forbes.com/technology/2006/03/09/torvalds-linux-licensing-cz_dl_
              >0309torvalds1.html) and consider this more important than the "hacker ethics
              > rule" of problems not having to be solved twice, or the desire to increase
              > the user base. Copyleft is completely legitimate and there is nothing
              > morally wrong with it.

              Didn't say there was. What I did say was that GPLv2-only or GPLv3-and-above
              (or even LGPLv3-and-above due to its incompatibility with GPLv2) do often
              stand against the Hacker's Ethics because they pose unnecessary limits on
              which code can be linked to other code. This is while a strong copyleft
              licence such as the SleepyCat Licence, or simpler weak copyleft licences such
              as Artistic 2.0 - which are less political and loaded than the FSF licences -
              are much better in this regard. As a result, I find licensing under them
              preferable and in better accordance to the Hacker's Ethics.

              The reason I like to use the X11L, is not because of that reason, but rather
              because not only I don't mind people "abusing" my software by building upon
              it to make a proprietary/non-free/non-GPL-compatible product, but I
              encourage it in some specific contexts.

              But it's ultimately up to the originator to decide which licence to use. If
              you want to publish your software under the Microsoft Vista EULA, then all
              the power to you. I wouldn't touch it with a ten foot pole, but I don't
              consider it an immoral act per-ce.

              Regards,

              Shlomi Fish

              -----------------------------------------------------------------
              Shlomi Fish http://www.shlomifish.org/
              Parody on "The Fountainhead" - http://xrl.us/bjria

              Shlomi, so what are you working on? Working on a new wiki about unit testing
              fortunes in freecell? -- Ran Eilam
            • Omer Zak
              ... I understood that you are (not strongly) against all viral licenses, regardless of whether they provide us with protection against the unfair software
              Message 6 of 12 , Nov 8, 2008
              • 0 Attachment
                On Thu, 2008-11-06 at 20:33 +0200, Shlomi Fish wrote:
                > On Thursday 06 November 2008, Omer Zak wrote:
                > > In an ideal world, in which all software vendors act fairly and compete
                > > only over the merits of their products, politics would not be necessary.
                > >
                > > But in today's world, politically-charged licenses - on at least part of
                > > the software running the world - are essential to keep the unfair
                > > software vendors in check. For, without viral licenses, they can take
                > > over any Free Software and create closed source derivatives out of it.
                >
                > The SleepyCat licence is a viral licence, but is much less politically charged
                > than the GPL, and I didn't have a problem understanding it. As a result I
                > recommend using it instead of the GPL, should you decide to use a viral
                > licence.

                I understood that you are (not strongly) against all viral licenses,
                regardless of whether they provide us with protection against the unfair
                software vendors.

                > Similarly, the Artistic 2.0 licence is LGPL-like, and should be used instead
                > of the LGPL because it's less politically charged and problematic.

                See above.

                > > Unrelated to the above, consider the fragmentation of the Unix world in
                > > the 1980's vs. the natural tendency to converge in the Linux world. The
                > > GPL played an important role in making this convergence happen.
                >
                > Possibly - I'm not sure. It preceded my time. However, there's a strong
                > convergence in the BSD, X11, and Apache worlds today, and they are all based
                > on BSD-style licences. GPL doesn't positively prevent fragmentation, as it
                > explicitly allows forking.

                1. Research this! The above fragmentation can be considered to have
                preceded also my time, because I was not conscious of this world until
                1991 (unrelated to Linux).

                2. In practice, it turns out that overwhelming majority of GPLed
                software does not fragment.
                While the GPL allows forking, in practice it turns out that there is an
                incentive for people to stick together in a single project, because if
                the original GPLed project has forked into two projects, then they are
                allowed to use each other's improvements, and eventually one of them
                turns out to be better (code/documentation/management/politics/business)
                and wins over the minds and hearts of developers and the other project
                dies natural death.

                > In any case, I'm not too worried about trying to undermine "unfair
                > software vendors".

                You can afford the luxury of not being too worried about unfair software
                vendors, because other people do the heavy worrying in your behalf.
                This is like parents who get away with not vaccinating their children
                against childhood diseases due to ideological reasons, and it turns out
                that their children stay healthy because the other children in their
                kindergarten or classroom have been vaccinated.

                > If they pose a threat (which I don't think they do), then
                > no amount of keeping my code GPLed will prevent that threat.

                This is a case in which every little bit helps.

                > If they don't
                > pose a threat, then making my code viral will not help them and may likely
                > pose a problem to many benevolent developers (also of FOSS).

                I think that LGPL is the best compromise - LGPLed software can be part
                of proprietary software, yet any modifications to the LGPLed part must
                be available in source code to users of the proprietary software, along
                with means for linking modified LGPLed modules into the software.

                > To quote Alan
                > Kay, "Don't worry about what everybody else are doing. The best way to
                > predict the future is to invent it.".

                This quote can be invoked as a reason to use GPL/LGPL, not only as a
                reason against it. By using GPL/LGPL on your software, you invent a
                future for Free Software.

                > One thing I forgot to note is that the GPL often stands against the Hacker
                > Ethics as presented by ESR in http://www.catb.org/~esr/faqs/hacker-howto.html
                > (and which I agree with), that says that "No problem should ever have to be
                > solved twice.". For example, the FSF now started the GNU PDF project so it
                > will be a GPL version 3-licenced code because both ghostscript and xpdf are
                > GPLv2-only. Or the story here - http://www.osnews.com/story/7241 that said
                > that:
                >
                > {{{{{{{{
                > Once before, someone had contributed a patch to add boolean operations, but
                > that patch relied on a polygon clipping library provided under an
                > incompatible license. There's little more frustrating than having a solution
                > in hand, only to be hamstrung by legal problems. Even though it was an
                > important feature for us, we regretfully postponed development of it into the
                > distant future on our roadmap and proceeded with other work.
                > }}}}}}}}
                >
                > Or the fact OpenBSD is now re-implementing a lot of GNU GPL code from scratch
                > because they find the GPL too restrictive.

                The people, who slapped the GPL license on their code, expect to be paid
                in a way (not monetary), which OpenBSD is not willing to pay. This is
                not different from re-implementing a proprietary software application on
                your own because the one you can buy in shop is too expensive for you to
                buy.

                Your argument is not different in essence from the argument of someone,
                who wants to buy a Ferrari car but grumbles about its high price.

                > And I should also note that often one can build upon non-GPL-like-code and not
                > release the derived work as FOSS, without it causing any harm, and actually
                > causing a lot of good. Which is otherwise prevented if the code is GPL.

                RMS would disagree with you, and he could cite the printer driver
                example as an counterexample to your point.

                --- Omer
                --
                Did you shave a yak today?
                My own blog is at http://www.zak.co.il/tddpirate/

                My opinions, as expressed in this E-mail message, are mine alone.
                They do not represent the official policy of any organization with which
                I may be affiliated in any way.
                WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html
              • Shlomi Fish
                Hi Omer! Sorry for the late response. ... I am against them in the sense that I prefer not to use them for my code. But I accept the fact that some people want
                Message 7 of 12 , Nov 19, 2008
                • 0 Attachment
                  Hi Omer!

                  Sorry for the late response.

                  On Sunday 09 November 2008, Omer Zak wrote:
                  > On Thu, 2008-11-06 at 20:33 +0200, Shlomi Fish wrote:
                  > > On Thursday 06 November 2008, Omer Zak wrote:
                  > > > In an ideal world, in which all software vendors act fairly and compete
                  > > > only over the merits of their products, politics would not be
                  > > > necessary.
                  > > >
                  > > > But in today's world, politically-charged licenses - on at least part
                  > > > of the software running the world - are essential to keep the unfair
                  > > > software vendors in check. For, without viral licenses, they can take
                  > > > over any Free Software and create closed source derivatives out of it.
                  > >
                  > > The SleepyCat licence is a viral licence, but is much less politically
                  > > charged than the GPL, and I didn't have a problem understanding it. As a
                  > > result I recommend using it instead of the GPL, should you decide to use
                  > > a viral licence.
                  >
                  > I understood that you are (not strongly) against all viral licenses,
                  > regardless of whether they provide us with protection against the unfair
                  > software vendors.

                  I am against them in the sense that I prefer not to use them for my code. But
                  I accept the fact that some people want to use them. However, I do think that
                  the SleepyCat licence is preferable over the GPL for the reasons I cited
                  below.

                  >
                  > > Similarly, the Artistic 2.0 licence is LGPL-like, and should be used
                  > > instead of the LGPL because it's less politically charged and
                  > > problematic.
                  >
                  > See above.

                  The Artistic licence is not strong-copyleft. I am not against the Artistic
                  licence (at least not as much I am against the GPL or the SleepyCat
                  licences), but on the other hand, what I mean is that the GPL and LGPL, are
                  too over-complicated, incomprehensible, and heavily misunderstood to be a
                  wise choice as a licence.

                  >
                  > > > Unrelated to the above, consider the fragmentation of the Unix world in
                  > > > the 1980's vs. the natural tendency to converge in the Linux world.
                  > > > The GPL played an important role in making this convergence happen.
                  > >
                  > > Possibly - I'm not sure. It preceded my time. However, there's a strong
                  > > convergence in the BSD, X11, and Apache worlds today, and they are all
                  > > based on BSD-style licences. GPL doesn't positively prevent
                  > > fragmentation, as it explicitly allows forking.
                  >
                  > 1. Research this! The above fragmentation can be considered to have
                  > preceded also my time, because I was not conscious of this world until
                  > 1991 (unrelated to Linux).

                  I have researched it to some extent.

                  >
                  > 2. In practice, it turns out that overwhelming majority of GPLed
                  > software does not fragment.
                  > While the GPL allows forking, in practice it turns out that there is an
                  > incentive for people to stick together in a single project, because if
                  > the original GPLed project has forked into two projects, then they are
                  > allowed to use each other's improvements, and eventually one of them
                  > turns out to be better (code/documentation/management/politics/business)
                  > and wins over the minds and hearts of developers and the other project
                  > dies natural death.
                  >

                  Well, the same thing can be said about forks of BSD-style-licensed projects
                  that are both kept under the same licence. In regards to proprietary
                  spin-offs - given enough motivation on those who remained FOSS, they can
                  continue working on the codebase, and gain inspiration from the properietary
                  spin-offs.

                  > > In any case, I'm not too worried about trying to undermine "unfair
                  > > software vendors".
                  >
                  > You can afford the luxury of not being too worried about unfair software
                  > vendors, because other people do the heavy worrying in your behalf.
                  > This is like parents who get away with not vaccinating their children
                  > against childhood diseases due to ideological reasons, and it turns out
                  > that their children stay healthy because the other children in their
                  > kindergarten or classroom have been vaccinated.

                  Interesting analogy. Well, companies who wish to develop proprietary software
                  have enough BSD-licenced (or possibly also LGPL/Artistic/etc.) code or
                  alterantively proprietary code that they can build on (like the Microsoft
                  APIs) for them to not have to use any GPLed code.

                  Furthermore, I don't think the FOSS world is under any threat from proprietary
                  software vendors, that can be prevnted if I GPLed all my code. Can you cite
                  any possible substantial threat?

                  >
                  > > If they pose a threat (which I don't think they do), then
                  > > no amount of keeping my code GPLed will prevent that threat.
                  >
                  > This is a case in which every little bit helps.
                  >
                  > > If they don't
                  > > pose a threat, then making my code viral will not help them and may
                  > > likely pose a problem to many benevolent developers (also of FOSS).
                  >
                  > I think that LGPL is the best compromise - LGPLed software can be part
                  > of proprietary software, yet any modifications to the LGPLed part must
                  > be available in source code to users of the proprietary software, along
                  > with means for linking modified LGPLed modules into the software.

                  Like I said, I still prefer LGPL-like-licences that are not the LGPL, because
                  the LGPL is based on the GPL (only more complicated) and because I don't
                  understand it.

                  >
                  > > To quote Alan
                  > > Kay, "Don't worry about what everybody else are doing. The best way to
                  > > predict the future is to invent it.".
                  >
                  > This quote can be invoked as a reason to use GPL/LGPL, not only as a
                  > reason against it. By using GPL/LGPL on your software, you invent a
                  > future for Free Software.

                  "Invent a future"? If I release software under any licence, I invent a future.
                  The licence affects the kind of future there is going be, but not its
                  substance.

                  >
                  > > One thing I forgot to note is that the GPL often stands against the
                  > > Hacker Ethics as presented by ESR in
                  > > http://www.catb.org/~esr/faqs/hacker-howto.html (and which I agree with),
                  > > that says that "No problem should ever have to be solved twice.". For
                  > > example, the FSF now started the GNU PDF project so it will be a GPL
                  > > version 3-licenced code because both ghostscript and xpdf are GPLv2-only.
                  > > Or the story here - http://www.osnews.com/story/7241 that said that:
                  > >
                  > > {{{{{{{{
                  > > Once before, someone had contributed a patch to add boolean operations,
                  > > but that patch relied on a polygon clipping library provided under an
                  > > incompatible license. There's little more frustrating than having a
                  > > solution in hand, only to be hamstrung by legal problems. Even though it
                  > > was an important feature for us, we regretfully postponed development of
                  > > it into the distant future on our roadmap and proceeded with other work.
                  > > }}}}}}}}
                  > >
                  > > Or the fact OpenBSD is now re-implementing a lot of GNU GPL code from
                  > > scratch because they find the GPL too restrictive.
                  >
                  > The people, who slapped the GPL license on their code, expect to be paid
                  > in a way (not monetary), which OpenBSD is not willing to pay. This is
                  > not different from re-implementing a proprietary software application on
                  > your own because the one you can buy in shop is too expensive for you to
                  > buy.
                  >
                  > Your argument is not different in essence from the argument of someone,
                  > who wants to buy a Ferrari car but grumbles about its high price.
                  >

                  Have you noticed what I said about the Hacker Ethics? This is not about
                  complying to the originator's whims. This is about making sure everyone in
                  the world will build on each other's efforts instead of having to waste time
                  in duplicating them.

                  > > And I should also note that often one can build upon non-GPL-like-code
                  > > and not release the derived work as FOSS, without it causing any harm,
                  > > and actually causing a lot of good. Which is otherwise prevented if the
                  > > code is GPL.
                  >
                  > RMS would disagree with you, and he could cite the printer driver
                  > example as an counterexample to your point.
                  >

                  I said "often" not "always". People such as RMS would claim that all
                  proprietary software is Evil and should be avoided. My argument is that often
                  it doesn't cause any damage.

                  For example, someone integrated my Freecell Solver library in a shareware
                  (without source) game he wrote titled Freecell 3D:

                  http://www.ambermango.com/php/am/f3d/index.php

                  He sells ths game online. It did not do any damage, because there are plenty
                  of Freecell implementations around. I on my part am pretty happy with the
                  open-source PySolFC.

                  On the other hand, it has some innovative features that we can now duplicate,
                  or draw inspiration from.

                  Regards,

                  Shlomi Fish

                  -----------------------------------------------------------------
                  Shlomi Fish http://www.shlomifish.org/
                  Understand what Open Source is - http://xrl.us/bjn82

                  Shlomi, so what are you working on? Working on a new wiki about unit testing
                  fortunes in freecell? -- Ran Eilam
                • Nadav Har'El
                  ... Thinking of future threats is hard (in the banking business, this is a whole profession, and as you can see lately, the haven t been doing a very good job.
                  Message 8 of 12 , Nov 23, 2008
                  • 0 Attachment
                    On Wed, Nov 19, 2008, Shlomi Fish wrote about "Re: [hackers-il] "Licences Wars"":
                    > Furthermore, I don't think the FOSS world is under any threat from proprietary
                    > software vendors, that can be prevnted if I GPLed all my code. Can you cite
                    > any possible substantial threat?

                    Thinking of future threats is hard (in the banking business, this is a whole
                    profession, and as you can see lately, the haven't been doing a very good
                    job. But as they say, "Hindsight is always 20:20", and it's easier to look
                    at the past.

                    And the experience of the last 20 years indeed shown that proprietary
                    software vendor's use of BSD has been a problem to free software. Not a huge
                    problem, certainly nothing that could ever kill free software, but a problem
                    nontheless.

                    Let me give you just a few examples.

                    In the second half of the 90s, X-Windows was quite popular - a department
                    (in universities, companies, etc.) would have strong Unix workstations
                    (from Sun, DEC, SGI, HP and other vendors) and people would have cheaper
                    machines showing the output from the strong machine using X-Windows. As
                    MS-Windows grew popular, people wanted to use their PC running Windows to
                    display X-Windows sessions. But unfortunately, the only X server available
                    for Windows was commercial software (Exceed), which could happen because X
                    was BSD-licensed and not GPL. Users (at the time, most had corporate or
                    university funds - they weren't home users) bit the bullet and paid. It took
                    literarly years before a free X server for Windows became available.

                    Another example - Mosix. Mosix (if you're familiar with it) was written over
                    a *proprietary* BSD variant called BSDI. Because the BSD licence allowed BSDI
                    to make it proprietary, and Mosix made the error in judgement (or, more
                    accurately, at the time of the judgement there weren't many alternatives)
                    of using it, for many years Mosix itself could not be made free software,
                    even though the people in the Hebrew University who developed it wanted to
                    do so.

                    Another less clear-cut example is the massive Unix diversity of the 90s.
                    A dozen companies, like Sun, HP, DEC, SGI, AT&T, SCO, BSDI, all created
                    their own variants of Unix and spent thousands of man-years on their
                    development. On the positive side, the Unix world was thriving because of
                    this diversity, but on the negative side, none of these thousands of man
                    years went into free software, and Richard Stallman and his friends had
                    to spend nearly a decade to replicate all this work. Is it surprising, then,
                    that Richard Stallman wanted to invent the copyleft license?

                    > > > If they pose a threat (which I don't think they do), then
                    > > > no amount of keeping my code GPLed will prevent that threat.

                    The GPL is not trying to prevent free software from being destroyed, but
                    rather to "force" people and companies into spending their work into free
                    software, rather than proprietary software. And I think you can't deny it's
                    succeeding. The Linux kernel is the most obvious example - after decades
                    where every company created their own kernel or took a BSD-licensed kernel
                    and "appropriated" it, in the last decade there is a huge push for all
                    comapnies to work together on the open Linux kernel, and all these companies
                    are "forced" to contribute their work back to free software.


                    > For example, someone integrated my Freecell Solver library in a shareware
                    > (without source) game he wrote titled Freecell 3D:
                    >
                    > http://www.ambermango.com/php/am/f3d/index.php
                    >
                    > He sells ths game online. It did not do any damage, because there are plenty
                    > of Freecell implementations around. I on my part am pretty happy with the
                    > open-source PySolFC.

                    Let's imagine that you wrote a great freecell solver library, but you suck at
                    UI and failed to create a graphical game based on that library, and wish that
                    someone created a UI on top of your library. If your library is BSD, someone
                    could create this UI but make it commercial and not even you can use it.
                    If your library is GPL, this person is forced to either make his code public
                    (so you and your friends can use it) or pay you for a new license.
                    Which of those scenarios will make you happier? I can't see how the GPL isn't
                    better in this scenario.


                    --
                    Nadav Har'El | Sunday, Nov 23 2008, 25 Heshvan 5769
                    nyh@... |-----------------------------------------
                    Phone +972-523-790466, ICQ 13349191 |The two rules for success are: 1. Never
                    http://nadav.harel.org.il |tell them everything you know.
                  • Shlomi Fish
                    ... I agree this is a problem. However, if X-Windows were GPLed, then the people who made Exceed and wanted to sell it, would not have made it in the first
                    Message 9 of 12 , Nov 23, 2008
                    • 0 Attachment
                      On Sunday 23 November 2008, Nadav Har'El wrote:
                      > On Wed, Nov 19, 2008, Shlomi Fish wrote about "Re: [hackers-il] "Licences
                      Wars"":
                      > > Furthermore, I don't think the FOSS world is under any threat from
                      > > proprietary software vendors, that can be prevnted if I GPLed all my
                      > > code. Can you cite any possible substantial threat?
                      >
                      > Thinking of future threats is hard (in the banking business, this is a
                      > whole profession, and as you can see lately, the haven't been doing a very
                      > good job. But as they say, "Hindsight is always 20:20", and it's easier to
                      > look at the past.
                      >
                      > And the experience of the last 20 years indeed shown that proprietary
                      > software vendor's use of BSD has been a problem to free software. Not a
                      > huge problem, certainly nothing that could ever kill free software, but a
                      > problem nontheless.
                      >
                      > Let me give you just a few examples.
                      >
                      > In the second half of the 90s, X-Windows was quite popular - a department
                      > (in universities, companies, etc.) would have strong Unix workstations
                      > (from Sun, DEC, SGI, HP and other vendors) and people would have cheaper
                      > machines showing the output from the strong machine using X-Windows. As
                      > MS-Windows grew popular, people wanted to use their PC running Windows to
                      > display X-Windows sessions. But unfortunately, the only X server available
                      > for Windows was commercial software (Exceed), which could happen because X
                      > was BSD-licensed and not GPL. Users (at the time, most had corporate or
                      > university funds - they weren't home users) bit the bullet and paid. It
                      > took literarly years before a free X server for Windows became available.

                      I agree this is a problem. However, if X-Windows were GPLed, then the people
                      who made Exceed and wanted to sell it, would not have made it in the first
                      place, because they had to make it GPLed. So either they would have
                      implemented it from scratch or not at all. It is possible that a different
                      group would have created a free X server for Windows, but it is possible that
                      no free X server would have been available at all. And if a different group
                      could have created a free X server for Windows without Exceed, they could
                      have certainly created it with it.

                      >
                      > Another example - Mosix. Mosix (if you're familiar with it) was written
                      > over a *proprietary* BSD variant called BSDI. Because the BSD licence
                      > allowed BSDI to make it proprietary, and Mosix made the error in judgement
                      > (or, more accurately, at the time of the judgement there weren't many
                      > alternatives) of using it, for many years Mosix itself could not be made
                      > free software, even though the people in the Hebrew University who
                      > developed it wanted to do so.

                      Well, the same problem could have been present if Mosix had chosen to use a
                      System V-based UNIX (or a different proprietary OS altogether) that never
                      originated from BSD-style code (although it did incorporate some code from
                      BSD). This is a problem on the judgement of the Mosix developers, not on the
                      availability of BSDI by itself.

                      >
                      > Another less clear-cut example is the massive Unix diversity of the 90s.
                      > A dozen companies, like Sun, HP, DEC, SGI, AT&T, SCO, BSDI, all created
                      > their own variants of Unix and spent thousands of man-years on their
                      > development. On the positive side, the Unix world was thriving because of
                      > this diversity, but on the negative side, none of these thousands of man
                      > years went into free software, and Richard Stallman and his friends had
                      > to spend nearly a decade to replicate all this work. Is it surprising,
                      > then, that Richard Stallman wanted to invent the copyleft license?



                      >
                      > > > > If they pose a threat (which I don't think they do), then
                      > > > > no amount of keeping my code GPLed will prevent that threat.
                      >
                      > The GPL is not trying to prevent free software from being destroyed, but
                      > rather to "force" people and companies into spending their work into free
                      > software, rather than proprietary software. And I think you can't deny it's
                      > succeeding. The Linux kernel is the most obvious example - after decades
                      > where every company created their own kernel or took a BSD-licensed kernel
                      > and "appropriated" it, in the last decade there is a huge push for all
                      > comapnies to work together on the open Linux kernel, and all these
                      > companies are "forced" to contribute their work back to free software.

                      This is true.

                      >
                      > > For example, someone integrated my Freecell Solver library in a shareware
                      > > (without source) game he wrote titled Freecell 3D:
                      > >
                      > > http://www.ambermango.com/php/am/f3d/index.php
                      > >
                      > > He sells ths game online. It did not do any damage, because there are
                      > > plenty of Freecell implementations around. I on my part am pretty happy
                      > > with the open-source PySolFC.
                      >
                      > Let's imagine that you wrote a great freecell solver library, but you suck
                      > at UI and failed to create a graphical game based on that library, and wish
                      > that someone created a UI on top of your library.

                      That's a lot of if's. When I started working on Freecell Solver, there were
                      already plenty of graphical Freecell games, possibly with support for other
                      variants of Solitaire. And there were also several open-source ones,
                      including some very polished ones.

                      As a result, I could have easily integrated my solver into one of them. So
                      your thought experiment is not very realistic, although it may be valid in
                      the general case.

                      > If your library is BSD,
                      > someone could create this UI but make it commercial and not even you can
                      > use it. If your library is GPL, this person is forced to either make his
                      > code public (so you and your friends can use it) or pay you for a new
                      > license.

                      He could also:

                      1. Develop his own Freecell solving library from scratch. That will probably
                      be less work than developing a graphical Freecell game.

                      2. Write a command line executable that uses the library call it, and process
                      its output. This strategy was taken by the PySolFC developers out of
                      convenience using my existing "fc-solve" executable.

                      3. Write a GPLed server for solving Freecell that the graphical game would
                      communicate with using TCP/IP or a different IPC mechanism.

                      4. Decide that developing a graphical game only to later release it as free
                      software won't be worth his time, and as a result not develop it at all.

                      In all of these cases, we won't have a graphical game as FOSS.

                      > Which of those scenarios will make you happier? I can't see how
                      > the GPL isn't better in this scenario.

                      I'd rather have a proprietary derived work than no work at all, or instead
                      someone duplicating my effort in creating a BSD or a proprietary replacement
                      for my work.

                      Regards,

                      Shlomi Fish

                      --
                      -----------------------------------------------------------------
                      Shlomi Fish http://www.shlomifish.org/
                      Interview with Ben Collins-Sussman - http://xrl.us/bjn8s

                      Shlomi, so what are you working on? Working on a new wiki about unit testing
                      fortunes in freecell? -- Ran Eilam
                    • Shlomi Fish
                      ... Now that I think of it, I have another vital thought experiment to add to this. If X11 had been initiated under a non-BSD-style-licence, then it is
                      Message 10 of 12 , Jan 9, 2009
                      • 0 Attachment
                        On Sunday 23 November 2008 13:51:01 Shlomi Fish wrote:
                        > On Sunday 23 November 2008, Nadav Har'El wrote:
                        > > On Wed, Nov 19, 2008, Shlomi Fish wrote about "Re: [hackers-il] "Licences
                        >
                        > Wars"":
                        > > > Furthermore, I don't think the FOSS world is under any threat from
                        > > > proprietary software vendors, that can be prevnted if I GPLed all my
                        > > > code. Can you cite any possible substantial threat?
                        > >
                        > > Thinking of future threats is hard (in the banking business, this is a
                        > > whole profession, and as you can see lately, the haven't been doing a
                        > > very good job. But as they say, "Hindsight is always 20:20", and it's
                        > > easier to look at the past.
                        > >
                        > > And the experience of the last 20 years indeed shown that proprietary
                        > > software vendor's use of BSD has been a problem to free software. Not a
                        > > huge problem, certainly nothing that could ever kill free software, but a
                        > > problem nontheless.
                        > >
                        > > Let me give you just a few examples.
                        > >
                        > > In the second half of the 90s, X-Windows was quite popular - a department
                        > > (in universities, companies, etc.) would have strong Unix workstations
                        > > (from Sun, DEC, SGI, HP and other vendors) and people would have cheaper
                        > > machines showing the output from the strong machine using X-Windows. As
                        > > MS-Windows grew popular, people wanted to use their PC running Windows to
                        > > display X-Windows sessions. But unfortunately, the only X server
                        > > available for Windows was commercial software (Exceed), which could
                        > > happen because X was BSD-licensed and not GPL. Users (at the time, most
                        > > had corporate or university funds - they weren't home users) bit the
                        > > bullet and paid. It took literarly years before a free X server for
                        > > Windows became available.
                        >
                        > I agree this is a problem. However, if X-Windows were GPLed, then the
                        > people who made Exceed and wanted to sell it, would not have made it in the
                        > first place, because they had to make it GPLed. So either they would have
                        > implemented it from scratch or not at all. It is possible that a different
                        > group would have created a free X server for Windows, but it is possible
                        > that no free X server would have been available at all. And if a different
                        > group could have created a free X server for Windows without Exceed, they
                        > could have certainly created it with it.
                        >

                        Now that I think of it, I have another vital thought experiment to add to
                        this. If X11 had been initiated under a non-BSD-style-licence, then it is
                        possible it would not have become as ubiquitous as it is in the UNIX world,
                        thus making it irrelevant to port it to Windows in the first place. We can't
                        tell that for sure, but I think it is a possibility.

                        Regards,

                        Shlomi Fish (who is now trying to prepare a coherent document out of this
                        thread).

                        --
                        -----------------------------------------------------------------
                        Shlomi Fish http://www.shlomifish.org/
                        Original Riddles - http://www.shlomifish.org/puzzles/

                        <mauke> I'm not interested in what you're doing; what are you trying to
                        achieve?
                        <PerlJam> mauke: I'm trying to achieve world peace and this regex is
                        the last thing standing in my way! ;)
                      • Amit Aronovitch
                        ... In this case, your recommendations are also not free enough. Only license that would work is something like: You can do with this work whatever you like,
                        Message 11 of 12 , Jan 10, 2009
                        • 0 Attachment
                          On Sat, Nov 8, 2008 at 7:24 AM, Shlomi Fish <shlomif@...> wrote:
                          On Friday 07 November 2008, Amit Aronovitch wrote:
                          > On Thu, Nov 6, 2008 at 1:33 AM, Shlomi Fish <shlomif@...> wrote:
                          > > Now. The first thing I'll mention is that it is not entirely agreed upon
                          > > which
                          > > licences are free/open-source and which are not:
                          > >
                          > > The original Artistic Licence (authored by Larry Wall for dual-licensing
                          > > perl)
                          > > is:
                          > >
                          > > * considered non-free (and non-GPL-compatible) by the FSF
                          > >
                          > > * considered free by Debian.
                          > >
                          > > * considered non-free by RedHat.
                          > >
                          > > * considered free by Mandriva.
                          > >
                          > > * considered open-source by the OSI -
                          > > http://www.opensource.org/licenses/artistic-license-1.0.php
                          > >
                          > > * was considered a contract in a certain judgement:
                          > >
                          > > http://lwn.net/Articles/246695/
                          > >
                          > > (Who are you going to believe?)
                          >
                          > Possibly all. The reason is that "freeness" is not a boolean parameter. Not
                          > even a single parameter real-valued scale. There are multiple criteria,
                          > possibly contradicting each other. Much like human rights...
                          > It is much more practical to talk about specific definitions, such as
                          > OSI-free/DFSG-free/FSF-free etc.
                          > As your examples clearly demonstrate, these do not comply to a total
                          > ordering (in the mathematical sense:
                          > http://en.wikipedia.org/wiki/Total_order ).
                          >
                          > You should decide how you *want* your work to be distributed and choose the
                          > license accordingly (effectively decide which criteria should take
                          > priority). CC has a nice mechanism for it.

                          Well, generally speaking one should strive for endorsement as free (and
                          preferably GPL-compatibility) by everybody. Otherwise, you may be facing
                          problems.

                          In this case, your recommendations are also not free enough. Only license that would work is something like:
                          "You can do with this work whatever you like, even claim that you did it all by yourself, no need to mention me. However, you still have the right to sue me if it does not work".
                          The reason you don't find any license like that is because authors normally want to assure at least some of their own rights, and that comes at the expense of the "freedoms" of the end users. The question is how you prioritize these freedoms.

                          You actually want "endorsement as free" not really by *everybody*, just by a specific list of organizations which you consider relevant for your beliefs and for the projects you work on.

                          As for CC, the standard CC licences
                          (CC-by/CC-by-sa/CC-by-nc-sa/CC-by-nd/etc.) are not suitable for software due
                          to their attribution clause:

                          http://www.gnu.org/philosophy/bsd.html

                          In fact, the Creative Commons does not let you choose "software" as the type
                          of artwork, when filling their form.

                          I'm not sure what you mean, but it seems that the discussion in your link is irrelevant. It speaks about the specific case of attribution in "advertising materials" (you can't make effective ads if they must include 100 attributions).
                          The CC attribution clause says only that you must attribute "in the manner specified by the author" - this in itself places no restriction, all depends on what the author explicitly requires.
                           

                          >
                          > > The Creative Commons Attribution (CC-by) and Attribution-ShareAlike
                          > > Licences
                          > > (CC-by-sa) are considered free (but not GPL-compatible) by the FSF (see
                          > > http://www.gnu.org/philosophy/license-list.html#OtherLicenses ), while
                          > > the Debian legal team concluded that they were not free.[1]
                          >
                          > Notes:
                          > (1) Version 3.0 CC licenses *are* considered DFSG-free.

                          I see. That's god.

                          Hmm... I though you were an atheist ;-)

                                 AA


                        • Shlomi Fish
                          ... This seems a bit contradictory. And who would be happy with such a licence and not with, say, the X11L? ... I see. ... Well, I have yet to hear of an
                          Message 12 of 12 , Mar 3, 2009
                          • 0 Attachment
                            On Saturday 10 January 2009 21:27:50 Amit Aronovitch wrote:
                            > On Sat, Nov 8, 2008 at 7:24 AM, Shlomi Fish <shlomif@...> wrote:
                            > > On Friday 07 November 2008, Amit Aronovitch wrote:
                            > > > On Thu, Nov 6, 2008 at 1:33 AM, Shlomi Fish <shlomif@...> wrote:
                            > > > > Now. The first thing I'll mention is that it is not entirely agreed
                            > >
                            > > upon
                            > >
                            > > > > which
                            > > > > licences are free/open-source and which are not:
                            > > > >
                            > > > > The original Artistic Licence (authored by Larry Wall for
                            > >
                            > > dual-licensing
                            > >
                            > > > > perl)
                            > > > > is:
                            > > > >
                            > > > > * considered non-free (and non-GPL-compatible) by the FSF
                            > > > >
                            > > > > * considered free by Debian.
                            > > > >
                            > > > > * considered non-free by RedHat.
                            > > > >
                            > > > > * considered free by Mandriva.
                            > > > >
                            > > > > * considered open-source by the OSI -
                            > > > > http://www.opensource.org/licenses/artistic-license-1.0.php
                            > > > >
                            > > > > * was considered a contract in a certain judgement:
                            > > > >
                            > > > > http://lwn.net/Articles/246695/
                            > > > >
                            > > > > (Who are you going to believe?)
                            > > >
                            > > > Possibly all. The reason is that "freeness" is not a boolean parameter.
                            > >
                            > > Not
                            > >
                            > > > even a single parameter real-valued scale. There are multiple criteria,
                            > > > possibly contradicting each other. Much like human rights...
                            > > > It is much more practical to talk about specific definitions, such as
                            > > > OSI-free/DFSG-free/FSF-free etc.
                            > > > As your examples clearly demonstrate, these do not comply to a total
                            > > > ordering (in the mathematical sense:
                            > > > http://en.wikipedia.org/wiki/Total_order ).
                            > > >
                            > > > You should decide how you *want* your work to be distributed and choose
                            > >
                            > > the
                            > >
                            > > > license accordingly (effectively decide which criteria should take
                            > > > priority). CC has a nice mechanism for it.
                            > >
                            > > Well, generally speaking one should strive for endorsement as free (and
                            > > preferably GPL-compatibility) by everybody. Otherwise, you may be facing
                            > > problems.
                            >
                            > In this case, your recommendations are also not free enough. Only license
                            > that would work is something like:
                            > "You can do with this work whatever you like, even claim that you did it
                            > all by yourself, no need to mention me. However, you still have the right
                            > to sue me if it does not work".

                            This seems a bit contradictory. And who would be happy with such a licence and
                            not with, say, the X11L?

                            > The reason you don't find any license like that is because authors normally
                            > want to assure at least some of their own rights, and that comes at the
                            > expense of the "freedoms" of the end users. The question is how you
                            > prioritize these freedoms.

                            I see.

                            >
                            > You actually want "endorsement as free" not really by *everybody*, just by
                            > a specific list of organizations which you consider relevant for your
                            > beliefs and for the projects you work on.
                            >

                            Well, I have yet to hear of an organisation that doesn't consider the X11L as
                            a usable licence. But who knows?

                            > As for CC, the standard CC licences
                            >
                            > > (CC-by/CC-by-sa/CC-by-nc-sa/CC-by-nd/etc.) are not suitable for software
                            > > due
                            > > to their attribution clause:
                            > >
                            > > http://www.gnu.org/philosophy/bsd.html
                            > >
                            > > In fact, the Creative Commons does not let you choose "software" as the
                            > > type
                            > > of artwork, when filling their form.
                            >
                            > I'm not sure what you mean, but it seems that the discussion in your link
                            > is irrelevant. It speaks about the specific case of attribution in
                            > "advertising materials" (you can't make effective ads if they must include
                            > 100
                            > attributions).

                            I see.

                            > The CC attribution clause says only that you must attribute "in the manner
                            > specified by the author" - this in itself places no restriction, all
                            > depends on what the author explicitly requires.

                            Hmmm... interesting. In any case, the FSF does not consider CC-by and CC-by-sa
                            as GPL-compatible. And I'm not sure the CC attribution clause (while taking
                            into consideration its difference from the original BSDL advertising clause)
                            is GPL-compatible. And the Creative Commons still recommends using one of the
                            more standard (and older) software licences for licensing programs rather than
                            one of the CC licences.

                            Regards,

                            Shlomi Fish

                            --
                            -----------------------------------------------------------------
                            Shlomi Fish http://www.shlomifish.org/
                            "Humanity" - Parody of Modern Life - http://xrl.us/bkeut

                            <mauke> I'm not interested in what you're doing; what are you trying to
                            achieve?
                            <PerlJam> mauke: I'm trying to achieve world peace and this regex is
                            the last thing standing in my way! ;)
                          Your message has been successfully submitted and would be delivered to recipients shortly.