Date: 2003/03/25 16:27
From: Gerry Gleason <gerry@geraldgleason.com>
To: Francois PELLEGRINI <pelegrin@labri.u-bordeaux.fr>
Francois PELLEGRINI wrote:
>
> Hello all,
>
> Just a precision: when I said from far away, I
> meant that I am furiously busy by fighting
> software patents in Europe (a software patent
> directive is being voted, which we try to amend
> so that it will prevent the patenting of software
> -- at this stage, it is easier to reverse the meaning
> of a legal text and make it say the contrary of what it
> meant, as a positive law, rather than trying to remove
> it). So, I can not read all what you send to me, and
> I will have trouble to help you.
Thanks for the clarification. I wasn't sure what you meant by that at
first, but it seemed like a good opportunity to encourage people to get
off the sidelines and make a contribution however substantial. From
what I read from Carl's recent link about the patent situation in
Europe, strong positive laws are needed to just to counter how the
current law has been stretched by creative lawyers driven by commercial
interests. To the extent that you are successful within your political
system, it can only help us hear in the US where the law has already
been changed to a large extent by the copyright and patent interests.
I think the only thing that can really turn the tide here is if a large
section of the technology industries in the US start to realize that
anti-competitive strategies harm entire markets and cause them to lose
competitive edge. I have no doubt that Europe will gain economically in
comparison with the US if you succeed in stopping and reversing this
trend, and that sound research already demonstrates the truth of this
assertion. The problem is that complicated arguments don't play well in
the political sphere. I'm thinking of the "If you are explaining, you
are losing" quote from Lessigs speach (referring to the environment in
Washington).
>
> Gerry Gleason wrote:
>
>> George Dafermos wrote:
>>
>>>> One thing to consider is whether the Organis design ideas can be
>>>> patented as unique business processes and
>>>> organizational structures. Obviously this has to be handled very
>>>> carefully to avoid major strategic mistakes, but
>>>> it could be the way to initiate the strategic moves as discussed
>>>> above.
>>>>
>>>
>>> Its an idea that has been in the background..... of my mind ...
>>> patenting business processes and organizational structures is pushing
>>> the domain of patents and supports it..... on the other hand is may
>>> allow investment in the idea. and by investment I mean not just VC, but
>>> more important code and other investments of seting up NPO's and daily
>>> operations.
>>>
>>>
>>> In my opinion, patenting the very organic structures and processes
>>> that make Organis what it is will be a strategic mistake which we
>>> may not be able to undo later on. Not only will it stiffle
>>> creativity in the area but it may also breed bitterness among
>>> enthusiasts. To a certain extent, I'm sure that 'embracing patents'
>>> in the the same way that GPL embraces copyright is a smart move,
>>> however, such a strategy may have unintended consequences as we can
>>> all imagine.
>>> I've started writing an email regarding the Osaka proposal and I
>>> hope I'm not too late to provide some feedback.
>>>
>>> George
>>>
>> I think you are right about the nature of the risk. In fact, when I
>> suggested it, I had a vague thought about how this could be used to
>> establish a clear and unbreakable connection to the GGPL licenses.
>> On second thought, I'm sure I wouldn't be happy if the FSF had
>> established a unique legal position for themselves with respect to
>> the GPL, so I am inclined to agree with your assessment of this
>> concept. At some level, there could be a need to defensively patent
>> a lot of stuff including this, but I guess it would surprise me if
>> anyone else wanted to patent something like Organis.
>
>
> Patents cannot protect innovation in the context of a
> rapidly moving innovation scheme.
>
> Because patents take more than 4 years to be granted,
> they are granted much too later compared to duration
> life of for instance software products (18 months), so
> that they can only be used as anti-competitive weapons
> of the largest range possible, i.e. broadest claims as
> possible.
>
> In the case of Organis, other people could well patent
> it just to make sure that you will not develop it. This
> happened in the past.
>
> The solution is to timestamp and/or publish its
> specifications ASAP, to make prior art. This will
> be as valid as prior art by patents, because you
> do not want to exclude people from using Organis,
> but just make sure that nobody prevents other people
> including you from developing it.
Agreed.
>
> Anyway, the PTO will grant patents to everybody asking
> for it (see below), so whether you have patents or other
> prior art will anyway require you to defend your position
> in court, at an expense of more than 1 M$ per trial, just
> to prove your good faith. This is why patents are so
> damaging for SMEs and are such a powerful anti-competitive
> weapon.
>
> Giving money to the PTOs, hoping that this will protect
> you, is completely wrong.
You're saying that an actual patent is no better protection as a
demonstration of prior art, and you have to defend it in court anyway?
It is really insidious, since you could end up with the same high legal
costs either way. If the PTO issues a bad patent that covers something
you need, you have to be ready to go to court, or not use the
technology. No wonder lawyers get such a bad rap (sorry Larry).
>
>> The fact that it has been written about should establish it as prior
>> art, but at least the USPTO doesn't seem to be capable of rejecting a
>> patent even when the prior art is obvious to anyone.
>
>
> No, because they have an incentive in accepting
> patents and not rejecting them: more patents =
> more money, irrespective of the harm they provoke.
> For an analysis of the EPO, which is just a step
> beyond the USPTO, please look at:
>
> http://cip.umd.edu/Aigrain.htm
> (point 9)
>
>> That's really the motivation for 'embracing patents', because patents
>> often seem to be the only recognized prior art, making them necessary
>> if only defensively.
>
>
> No, as said above.
> See: http://www.forbes.com/asap/2002/0624/044.html
>
> Either you have 10000 patents, or you do
> not need to have any...
Good point. Haven't checked out those links yet, but I will.
>
>
>> It is also worth noting that the GGPL terms wouldn't restrict a
>> patent holder from licensing the patent under other terms while also
>> making it available for GGPL project inclusion under no-cost terms.
>> This has been discussed as a valid model for GPL release as well. In
>> this context, the restrictions of GPL and GGPL can be seen as an
>> advantage because the IP owner isn't giving away an essentially
>> unrestricted license, and thus the IP can retain a sale value to
>> people who want to create a derivative work under other licensing
>> terms including fully commercial terms.
>
>
> It is possible to mix licences, e.g. to have
> a GPL-like licence for free, and a proprietary
> licence for money.
> It is a bit more complex, but workable.
> This guarantee that other parties can contribute for
> free if they agree to put their contributions in the
> common pot, while allowing other bodies to derivate
> private work, if they choose to "play individual", but
> with some expense that will anyway benefit to the public
> stream by funding the organisation.
Although I think this arrangement is workable, your comment about the
possibility of losing it to a buyout is significant. After reading
Carls message about RTLinux, I'm wondering if the kind of free grant
under GPL (or GGPL) could be crafted to be protected under a buyout
situation or if it could be withdrawn at any time by the new owners.
The bottom line is that the unnatural extention of patents to ideas
(software and business processes), and the erosion of unregulated and
fair-use under copyright need to be confronted directly on their own
terms. In the context of Open/Free Hardware, I think this sort of
hibred approach to patents could be perfectly fine. A company could get
the benefit of follow-on designs and varification by releasing under
GGPL, and still extract revenues from many commercial uses.
>
>> This is more what I was referring to as having a connection with the
>> normal patent process.
>
>
> The economy of immaterial has nothing to do with
> the economy of material goods for which the patent
> system was designed.
>
>> I don't think RMS really anticipated that a large body of GPLed code
>> would develop with a large number of independent authors retaining
>> primary rights to their work while sharing under GPL terms.
>
>
>> The fact is that the universe of GPLed code is far larger than what a
>> single organization (FSF) could produce, and the GPL (and Free/Open
>> Source generally) is far stronger for it.
>
>
> Yes. But this must be encouraged, as it guarantees
> the flourishing and perpetuation of the model. This
> is a "network externalities" property: the more people
> use the model, the more it becomes strong and useful.
Absolutely! I'm pointing to this as a reason for not attempting to
control things more closely. For example, I know that Carl is very
insistent about the importance of the GGPL and its details, but in the
long-run, it is absolutely critical that it is able to evolve as an
ethical consensus of the entire community. The GPL example is very
good. Initially, RMS was pretty rigid about it, but eventually they
introduced the idea of GPL compatible licensing so that now there are a
number of variations on GPL and they all support each other. George has
made the point that a single clear license is important, but it also
works fine with a dominant one (GPL) that everyone (sort-of) understands
with variants for people who insist on some flexibility.
>
> f.p.
>
>
>
> P.S.: I have to hide my periscope as I have a heavy schedule
> this week.
Bringing your other message in for a comment or too ... I wasn't able
to get very clear about how the W3C policy is intended to work. If it
is more-or-less what we have been discussion about dual licensing, then
fine, but I suspect there are subtle differences. The bottom line is
that if the patent is in a standard, and a GPL implementation is
available, there should be no additional limitations placed on that GPL
implementation reguardless of who uses it, or how it is used. If it
doesn't do that, I can see why FSF is against it.
Of course, this would be moot if we didn't have software patents in the
first place ...
Gerry
F.P. also wrote:
Just another thing: the FSF is against
the current proposed policy of W3C, because
the ability to use the patents royalty-free
is limited to some uses, while the GPL
explicitely states that the covered software
could be used for * any* purpose.
The final word of the tale is then:
- if GPL once, GPL forever
- if patents once, patents forever
- if patents once, GPL never