[ILUG] ILUG sends s/w patents briefing document to Irish MEPs

[David Golden]

On Friday 18 March 2005 09:39, Paul Jakma wrote:
> NB: Please _include_ me in replies.
>
If you want...

> "Your honour, in my defense, I did not consider your esteemed
> colleague's ruling binding and hence seek that this patent action
> against me be dismissed" - good luck with that one.. ;)
>

It's more "you honour, here in $OTHER_COUNTRY, I do not
consider your esteemed colleague in Germany's ruling binding and neither 
should you, because XYZ, and not that prior rulings are necessarily 
binding in our legal system anyway".

> I've no idea I'm afraid how it works. I'd be sceptical that the
> judiciary in, eg, Germany has absolutely no interpretative role.
>

Well,  AFAIK it's much easier to argue against earlier decisions. 
According to that fount of reliable knowledge ;-) Wikipedia, indeed 
germans do tend to consider a form of precedent in a fashion similar to 
america, but they also _don't_ consider it binding over what laws 
actually say (AFAIK, here and in England,  laws ARE what the case law 
says, written "laws" are merely "statutes", or something like that.):

http://en.wikipedia.org/wiki/Civil_law_(legal_system)


> To be honest, it's not just software writers who have a stake in this
> question - it's those who provide the capital to allow the writers to
> write the software (who often will not be software writers, and even
> if they are, they possibly /do/ want to be able patent software, in
> principle).
>

Some of them probably do, anyway. At least one euro (small, only to the 
tune of ~ 3 million funding for ventures) venture capitalist has taken 
an anti-software-patent stance e.g. Laura Creighton, with some very 
eloquent arguments, including pointing out that US software patents may 
have worsened the .com bubble and subsequent bust in america:
http://www.vrijschrift.org/swpat/030508_1/

Deutsche Bank Research  has taken an anti-software-patent stance.
I don't know how much of an independent voice they are from Deutsche 
Bank Group proper, so I'm not sure if it's totally valid to extrapolate 
to the opinion of that part of the group that might be supplying 
venture capital, but in my experience euro small time software 
developers are more likely to approach a bank for a business loan than 
american-style venture-capitalists anyway as the costs just aren't that 
high....

http://www.dbresearch.com/PROD/DBR_INTERNET_EN-PROD/PROD0000000000175949.pdf


> There are no benefits to being a slave. 

Aside: Very much untrue, typically there was legally guaranteed bed and 
board and healthcare in western systems with legal slavery  - in some 
ways employees are worse off than most real slaves with 
not-particularly-evil-by-the-standards-of-the-time slave masters used 
to be. However, there was an element of hyperbole in my statement none 
the less for which I apologise. 

At the same time,  I'd rather be a lean self-employed (NOT unemployed!) 
and free to compute arbitrary recursive functions than a plump 
salaryman who needs corporate permission.  Now, you might argue it's
not right for me to decide re liberty vs. safety for other people, but 
it isn't any more right for them (particularly ones in the democratic 
minority) to decide for me either, so as I said, I'm willing to 
compromise: they can have patents so long as they don't restrict me,  
and in return I won't seek patents to restrict them. :-)

> This just wrong. You seem to hold the premise here that patent must
> always be primarily be for the good of the public. This is
> *fundamentally* wrong. 

I don't know where you got that from, surely the premise IS that the 
patent must primarily  be for the good of the public, just some 
nebulous and assumed long-term societal good, not the short term good.  
I'm fully aware of the compromises patent law is supposed to embody. 

I, as a member of society (maybe time to fork society), simply do not 
think it's worth it for software in the short or long term (or 
mechanical engineering for that matter, but I'm not in any vocal 
majority there, though my engineering economics lecturer would likely 
have agreed with me). 

> > states "software shall be protected as a literary work." (article
> > 10, paragraph one).  Patenting software pretty directly undermines
> > such copyrights, unless I guess one were to hold that books and
> > artworks should be patented or something (which some people do...)
>

> No it doesn't.
>

Hm. Note that historically, copyright lawyers have applied
the abstraction-filtration-comparison test to cover nonliteral copying

http://www.ip-wars.net/story/2004/12/15/94932/822

In conversations with specialised patent lawyers ... and many open 
source developers, I've found most to be woefully unaware of the huge 
scope (right or wrong) of existing copyright protection for software
over nonliteral copying.

If your software carries out substantially similar computational steps 
as another piece of software,  and it can be shown you didn't develop 
totally independently of that software but rather analysed it*, it's 
still been found to be copyright infringement in some jurisdictions, 
like rewriting a melody arranged for piano for the clarinet, or 
translating a book from english to irish (example where two languages 
are valid under one jurisdiction because translations across national 
boundaries are murkier).  

*The main and ultra-important exception in the EU is reverse engineering  
for interoperability, thankfully, or many open source projects would 
already have been litigated into the ground.

In some ways, therefore, patent law might be preferable to copyright 
because its term is shorter. However, it's not "patent law instead of 
copyright law", they want both, belt-and-braces.

But the major thing would be that with patents, even carefully 
independent (re)development is restricted.

> Read the comission's proposal again.
 
I've read it several times at this stage. It is unconvincing.

> (Even if you think the commision 
> are biased and their reasoning is wrong - its /very/ useful as a
> document to gain insight on the only arguments which /matter/ and

That allows them to set the terms of the debate.  They may be the only 
arguments that matter to _them_, but they are our servants, not 
vice-versa, at least notionally :-).  Too many people in authority 
these days seem to think they are ruling subjects rather than governing 
in our name.    They would simply love everyone to be arguing about 
what constitutes a "technical contribution" rather than whether the 
patents should exist in the first place.

After all, purely economically, the reason for doing in software what 
was once done in purpose-designed hardware and was patentable is to 
save substantial development and updating costs.  Sure software is 
expensive and manual to write, for implementing a complex behaviour 
it's still vastly cheaper and easier than laboriously designing a 
mechanism (n.b. I'm trained in doing both...), and vastly cheaper to 
release a firmware patch than to do a product recall.   So, even for 
embedded devices and cases where equivalent hardware would be 
patentable, I do not accept that software patents including those 
described as "computer implemented invention" patents are a good idea
or that such equivalence arguments are valid!


> I really don't think it's at all constructive to approach everything
> commision related with an assumption that they act deceptively or
> with malice.

Perhaps not, but it's not constructive for them to have (in my opinion,
allegedly, yadda yadda) acted deceptively and perhaps with malice in the 
past.

> (note that 'Free Software' is mentioned in the proposal several
> times)

Yes, "we'll set up a monitoring board to observe the effects upon it".
Not so much clear mention of what they could or would actually do if it 
is harmed.  They present "it's okay, we'll monitor it" as if it were a 
solution in itself.

Given that some of the people involved and with the commission's ear 
have a stated goal to derail free software (Microsoft), or a tendency 
to try to steer free software for their own ends (IBM), they may well 
monitor it, and cheer each time it gets another paper cut from a 
patent.

> and concluded the majority of interests were in favour of 
> harmonising on current EPO practice.
>

They concluded the "economic majority" was in favour, while 
acknowledging that the vast numeric majority of respondents were 
opposed. If you don't think that's wrong, well, that's just a 
fundamental difference of opinion, plutocratic vs. democratic.

> This is hardly some big evil commision + EPO + patent lawyer
> conspiracy..
>

Heh. You're right, it's not _just_ them: after all, it turned out the 
BSA had "helped" with the original proposal, and various multinationals 
have been busily telling people they'll punitively pull out of
europe if they don't get the law they want.  :-)

> Actually, my dictionary suggests patent derives from "to be open".
>

Uh yeah, "patently obvious" -> "openly obvious". Which was my point,
thought I was illustrating that  point there, sorry if it wasn't 
sufficiently clear: the letters patent were open letters so that other
people could see the king's orders - not much point if the letters
were sealed - " 'ere, I have this letter from the king, I can't show
it to you, but it says you can't do that".  Patent as in open as in
"patent" means the patent letter was an open letter. 

"""
Letters patent are a type of legal document which is an open letter 
issued by a monarch or government granting a right, monopoly, title, or 
status to someone or some entity such as a corporation. The opposite of 
letters patent is letters close which are personal in nature and sealed 
so that only the recipient can read the contents of the letter.
"""

Perhaps you thought it was to do with the disclosure patents were later
supposed to embody as a retrofitted rationale for continuing with
the monopoly grants? 


> - that the granted term be shorter than traditional patents
>    in recognition of the fact that software usually does not require
>    massive capital investment (eg factories, tooling) to develop
>

Now that _is_ forbidden by TRIPs.  If you grant a patent in one field,
has to be the same length as all other fields.  I estimate that it would
be easier for someone to that software patents aren't necessary
under TRIPs than argue against that relatively clearly stated principle 
in TRIPs.  

> - that they include a working implementation
>

Well, the patent people would probably say that would reverse historic 
decisions otherwise in other fields where you now need just to be able 
to instruct someone to build an implementation.

> - that they be worded very very specifically and in detail
>

Heh, well that'd be nice (and not just for software patents). But
patent lawyers acting for patent applicants seem to pride themselves
on making patents both as obscure and as general as possible, and do it
as a full-time job.  Dunno how you could stop that.