[Date Prev] [Date Next] [Thread Prev] [Thread Next] Indexes: Main | Date | Thread | Author

Re: [ba-ohs-talk] open source + patents = puzzled

Johannes Ernst wrote:
> Am I off-base here, or is there quite a can of worms hidden here
> somewhere? Anyone ever looked at the patent portfolio of any of the
> (particularly larger) companies that often release some of their own
> code into the public domain?    (01)

I am not a lawyer, but I am a twenty+ year software developer with an
interest in copyright and patent legal issues, and it is my
understanding that:    (02)

1. You can release code under an open source license such as BSD (which
does not reference patents) and then sue users of the code for patent
infringement even if they are only using exactly the code you wrote.
This is clearly unfriendly, and users might fight in court saying you
gave them an implied license, but I think the author (patent holder)
might win. Look at it this way, if you own a patent on an algorithm,
like many "non-profits" do these days, 
why not just release an explicitly public domain implementation and then
say if someone wants to use the code, they must also pay to license the
patent?     (03)

The unfortunate thing about software patents is they imply having a
copyright license to copy code in no way ensures you have all the
licenses needed to run the code. Note that some especially greedy
non-profits (or for-profits) try to double dip by charging separately
both for patents and access to copyrighted implementations, but that's
another story (they often try to also claim trade secrets in the
copyrighted code as well to prevent discussing its shortcomings).    (04)

2. Some licenses like the GPL explicitly say any referred patent will be
licensed freely.     (05)

> Finally, any free program is threatened constantly by software patents.
> We wish to avoid the danger that redistributors of a free program will
> individually obtain patent licenses, in effect making the program
> proprietary. To prevent this, we have made it clear that any patent 
> must be licensed for everyone's free use or not licensed at all. 
[Note: IIRC Richard Stallman says this can be read to say a patent only
needs to cover free GPL implementations and can be otherwise enforced
against others.]    (06)

While one may or may not like the GPL's political stance on enforced
sharing, this is one gotcha it it confronts up front -- contributors
embedding patented algorithms to trap the unwary. Note that patents (and
copyrights) unlike trademarks can be enforced at any time -- even after
letting users slide by for years infringing the use of them (witness the
current British Telecom effort to enforce a hyperlinking patent).
  http://www.wired.com/news/print/0,1294,50398,00.html    (07)

3. Note though, that if GPL'd code is found to unknowingly reference a
patent that isn't freely licensed, the GPL says the code can't be
redistributed by anyone.     (08)

>From the GPL license text:
> If, as a consequence of a court judgment or allegation of patent infringement 
> or for any other reason (not limited to patent issues), conditions are 
> imposed on you (whether by court order, agreement or otherwise) that
> contradict the conditions of this License, they do not excuse you from 
> the conditions of this License. If you cannot distribute so as to satisfy
> simultaneously your obligations under this License and any other pertinent
> obligations, then as a consequence you may not distribute the Program at all.
> For example, if a patent license would not permit royalty-free redistribution 
> of the Program by all those who receive copies directly or
> indirectly through you, then the only way you could satisfy both it and 
> this License would be to refrain entirely
> from distribution of the Program.     (09)

Contrast this with "permission to use". 
  http://www.bootstrap.org/colloquium/permission.html    (010)

One reading of "permission to use" would imply if you contribute any
code to any Bootstrap effort you grant a license to Bootstrap and
Stanford to use any software patents somehow referenced by the code
however they see fit in any projects (including sublicensing to
Microsoft). However, this license does not extend to any end users of
your code or content (since "permission to use" just grants rights to
Stanford and BI, not the end users). Note that if you don't own all
patents referenced by your contributed code (say you put in hyperlinking
code but don't have a worldwide license to BT's hyperlinking patent)
then the indemnification clause of "permission to use" might be read to
imply you need to pay Stanford or BI whatever it takes to make the
patent holder happy (could be $billions for worldwide licensing rights,
including all legal fees etc.). This is why I harp on the revision of
"Permission to use" so that the Bootstrap OHS coding effort can move
forward as "open source" or "free software", where contributors in the
U.S.A. feel comfortable contributing given software patents. [Just
rattling around as another Thursday CPC meetign has presumably came and
went... ]    (011)

Proposal: regarding patents, repeal "permission to use" and use the GPL
license for the OHS project. Insist (only) that all contributors say
their work is original, and that they have permission to contribute it
from their employer if such is needed, and that they contribute the code
under the GPL license.     (012)

The old approach over the past two years isn't working to produce code.
To paraprase an old TV commercial -- "Where the code?"     (013)

Why isn't there any significant OHS code released through this
colloquium? Perhaps, realistically, nobody who insists on a more
permissive license (like BSD or Apache) because they have proprietary
commercialization plans for OHS is likely to be going to do any work of
this magnitude gratis anyway, and there are too many smooth talking
(university) competitors for grant funds to pay people to do it
otherwise significant funds would have been here by now -- given the
effort Doug (and others) have expended in beating the bushes. [Feel free
to surprise me.] Yet, the "permission to use" policy has driven away the
very peopel who might build a libre OHS gratis. If someone finally gets
significant grants funds for an OHS, and it can be proprietary, the
effort will probably just vanish in the noise of all the other ongoing
proprietary KM efforts brought up by list participants.     (014)

With Doug's visibility coupled to a GPL'd OHS implementation, the
project might get somewhere faster. We could even have two competing
projects -- an Apache version and a GPL'd version. I'd work on the GPL
version. Sadly, "permission to use" directly blocks my own efforts in
any public OHS direction involving the community of this list. It is
very frustrating -- because I can't just say to this list "here is a
GPL'd OHS project" without also writing a blank check to Stanford and
BI. (I would if I could just to see the project go forward and Doug's
vision realized, but I can't - and even if I could, I doubt my wife
would approve. :-)     (015)

Besides, I could not in good conscience convince any other person to
contribute even one line of code under "permission to use" -- since even
if I were willing to risk my entire financial future myself, I could not
ask others to be so foolish given all the thousands of other open source
or free software projects out there (many I first saw here) that touch
on OHS issues and do not have such a problem.     (016)

That is mainly why I stopped posting on this list -- to disassociate my
efforts from any twisting of "permission to use". I post now in hopes
this "permission to use" and patent license etc. issues can finally be
resolved. Alternatively, can some lawyer or judge just declare
"permission to use" invalid because it is a very broad agreement to hold
if not signed (or at least the "extended" part) -- essentially becoming
in effect like an overly broad non-compete agreement for any prudent
person?     (017)

Anyway, better things to do than be a broken record on this topic. If
the "permission to use" issue as regards software patents and other
things gets resolved someday (and Mei Lin wrote 1 May 2002 a discusion
is in the works), fantastic -- and please let me know so I can
contribute then. If not, this is the last I'll post on this "permission
to use" indemnification issue, to avoid dragging down the other very
positive aspects of this discussion forum (and myself in negativity). I
learn much by following this list. I'm unfortunately still left in a
quandry about whether posting on other issues here or unrev creates too
much of an association of my other efforts (like the Pointrel Data
Repository System) and "permission to use" and so the prudent thing on
that basis is not to post (such as during most of the past year) until
the issue is resolved. Whether I can contain myself given all the great
discussions going on here remains to be seen.     (018)

It's also saddening to see great contributors like Chris Dent and
Kathryn La Barre 
left twisting in the wind so long regarding a firm answer from BI or BA
or Stanford on copyright issues for the email archives -- when the whole
point of "permission to use" should be to at the very least make work
such as theirs possible! So "permission to use" prevents my work here on
code, but yet has produced no real benefit when it could, thus making it
in practice the worst of both worlds. As I wrote before (April 30, 2002
on the unrev list), if the work can't be done pro-bono, I'll pledge
US$300 towards legal fees for a lawyer to unravel this mess of
"permission to use" to strike a better compromise for OHS coding
volunteers and get Stanford and BI (or whoever) to sign off on it.    (019)

-Paul Fernhout
Kurtz-Fernhout Software 
Developers of custom software and educational simulations
Creators of the Garden with Insight(TM) garden simulator
http://www.kurtz-fernhout.com    (020)