Tony,
You wrote:
> On 5 October 2010 13:07, ron.hilton <ron.hilton-Wuw85uim5zDR7s880joybQ@public.gmane.org
> <mailto:ron.hilton%40comcast.net> > wrote:
>
> > You raise some good points, perhaps better addressed on Hercules-
> advocacy.
>
> Yup - I'll continue there. 'cause it's probably going to get messy... 
>
> > I doubt that this would be the first time that proprietary mainframe
> > software has been run on Hercules, and with the appropriate licenses it
> > should be quite legal to do so.
>
> Well, it's unclear. IBM seems to have claimed that Hercules violates a
> number of IBM patents on their hardware, so that would presumably have
> nothing to do with what guest software is running on it. But of
> course IBM has made a habit of obfuscation in this regard.
Of course you are right. I was referring to legality from a software
licensing standpoint only. Am I incorrect in assuming that at least some
ISVs (with the notable exception of IBM) have licensed their software to run
on Hercules? Or would PSC become the first to do so?
>
> > On the question of patents, let me first speak to the issue of software
> > patentability. Although I am a strong advocate of open standards and
> > open-source software, I think that there is no doubt that software,
> > including Hercules and the programs that run on it, represents
> patentable
> > subject matter in general.
>
> We can agree to disagree. Thought for the day: what arguments in
> favour of the patentability of software do not also argue in favour of
> the patentability of movies?
Copyrights protect the expression of ideas, whereas patents protect the
ideas themselves. Both types of protection are potentially applicable to
software. Only copyright protection is applicable to artistic expression
such as movies. One can circumvent a copyright through an original
re-expression of the idea, such as a "clean room" software rewrite. A patent
protects the underlying concept and cannot be circumvented in that way.
> > My advice to the open-source
> > community, including Hercules, would be to quit fighting a losing battle
> > against software patents. Instead, fight fire with fire. The best
> defense
> > against software patents is a good offense. Hercules should create or
> join a
> > "software patent commons" and actively start to build its own strong
> patent
> > portfolio. That is the way the patent system works - just like MAD
> > ("mutually assured destruction"). A patent is a right to exclude others
> from
> > practicing an invention. Only when both sides have a patent arsenal does
> > cooperation and progress occur.
>
> You know, this is rather like the advice given by Groklaw's Pamela
> Jones to "get a lawyer". It may well be good advice, but it is utterly
> infeasible for a small and unfunded group of developers to obtain
> patents on anything. It is virtually impossible to even think about
> patenting anything for under $10,000, and it's very rare to get in the
> door for that. It is also very difficult to maintain the open source
> development model while at the same time not jeopardizing
> patentability by early publication. Not impossible, certainly, but
> none of this is a good fit for the Hercules project.
It's true that patents cost money, but they also have value. My suggestion
would be to educate and equip the developers to timely file a provisional
patent application on their own, which only involves a nominal fee,
assigning the rights to the patent commons. Then the patent commons would
have a year in which to reassign the patent to a corporate or other investor
who wishes to commercialize it, with a license back to the patent commons
that is perpetual, worldwide, non-exclusive, and non-transferable. At a
minimum the assignee would have to agree to bear the cost of obtaining the
patent, which as you say probably represents a value in the $10K range.
Depending on its perceived value, they may also be willing to license other
patents from their own portfolio to the patent commons, and/or pay cash.
Thus the open-source project is protected, while the commercial entity has
the capability to use it in its own for-profit products. The patent commons
could also use any cash revenue thus obtained to purchase licenses to other
patents held by third parties that may be applicable to the open-source
project, or to fund the filing of low-cost provisionals on behalf of
developers who are unable or unwilling to do so on their own.
>
> You've worked in at least one patent-savvy company; you surely
> remember those ubiquitous page-numbered notebooks, and the corporate
> infrastructure that was needed to comb through ordinary software
> development to extract patentable matter. And you are perhaps aware
> that there is at least one US patent pending *by someone else* on
> material that was published and implemented in Hercules first. There's
> a good chance that patent will issue, because no one has the resources
> to stop it even at this early stage, let alone after it's been
> accepted.
>
The US patent office is actually borrowing a leaf from the open-source world
with their "peer-to-patent" pilot project. I think something like that will
be the key to improving the quality/validity of patents in general. There is
simply too much prior art for any one examiner to find. A community-based
approach is the only practical answer.
> > Regarding the PSC software, it is proprietary (e.g. copyrighted) and the
> > proxy-coupling technology upon which it is based is also patent-pending.
> But
> > as you correctly pointed out, that intellectual property protection
> applies
> > only to the PCT-based "software" and not to "hardware" functionality in
> > Hercules that exists now or may be developed in the future in response
> to my
> > proposal/invitation.
>
> Well I didn't really point that out; I don't know what PSC's pending
> patents purport to cover. So "for greater certainty" or "for the
> avoidance of doubt", as the contracts lawyers like to say, why not
> have your organization state clearly that it will not assert any
> patents related to this technology against Open Source hardware
> emulators?
I will state right here on the e-mail record that PSC will not assert any
patents against Hercules or any other open-source mainframe hardware
emulator. If and when Hercules creates a patent commons as described above,
PSC will license its PCT patents to it on the terms indicated.
Ron Hilton
RSS Feed