[Legal] Fw: Re: [Osdem]software patents in europe

[Alexander Bokovoy]

"Краткая" сводка текущего состояния дел с борьбой против введения патентов
на ПО в Евросоюзе.

----- Forwarded message from Xavi Drudis Ferran <xdrudis на tinet.org> -----

Date: Thu, 19 Sep 2002 00:42:10 +0200
From: Xavi Drudis Ferran <xdrudis на tinet.org>
To: osdem на lists.raphinou.com
Subject: Re: [Osdem]software patents in europe
In-Reply-To: <20020918185543.06871948.brouits на free.fr>
User-Agent: Mutt/1.3.25i
Reply-To: osdem на lists.raphinou.com

El Wed, Sep 18, 2002 at 06:55:43PM +0200, Beno?t Rouits deia:
> 
> Is there a consise and synthetic resume of that somewhere ?
>
Errr... Kinda difficult to summarize, sorry. I've tried my best,
but you should use the URLs for important details.

The EU Comission presented a directive[3] to legalize 30000 software
patents[4] illegally granted by the European Patent Office[5] and make
software patentable while now it isn't. To do that they pretend
they're only allowing patents on software with a technical
contribution, but they say software is technical, therefore any
software is a technical contribution, and patentable.  They spread FUD
and contradictory information[6] to cover their actions.

The directive has to pass through EU Council and European Parliament.

EU Council is possibly dominated by patent offices who want to get 
more patents and more money at any price. The EU Comission itself 
has nominated the EPO as its representative[20], which is odd (EPO 
does not belong to the UE) and cleary partial. To influence the 
Council one has to convince member state governments to mandate
their representatives to oppose the directive. 

The European Parliament has 626 MEPs[7] divided in committes. The
Culture[8] and Industry[9] committes should give an opinion on the
directive, and the JURI committee[12] should consider that opinion,
listen to external experts in a hearing and make a report.  I'm
worried that maybe only pro-swpat experts will be heard.  We should
demand that anti swpat people be invited.  Then the plenary of the
EuroParl votes on the directive (ETA: early feb 2003). But things are
probably decided before, since MEPs don't have time to track all
issues, and they follow the committees work (maybe the report, or
maybe the opinion of the party's MEPs on the committee). The JURI
rapporteur in charge of the report looks very pro-swpats[31,30], and the
study they're using[32] is not good enough.
 
Then the council votes and the story goes on. There's a little
pseudocode explaining it [11], together with links to committees,
forums, etc.

> Well, I would really know what the heck i can do instead 
> of just thinking about to shout but not really arg while not 
> personnaly mastering the facts.
> 

Very constructive on your part. Extensive documentation is in the
FFII site[14]. And our little take[15] is at Caliu's if you can read Catalan
(if you can read a couple of Romance languages you can read them all ;) 

We need two types of help:

- people contacting MEPs, member state governments, etc. Getting 
  statements[16] and signatures[2] from people of prestige, labor unions, 
  consumer associations, politicians, academics, managers, etc. And 
  keeping the press informed. Preferably coordinating with Eurolinux
  so that we don't launch a DoS on politicians and knock them off and
  instead educate them, motivate them and put them on our side. And 
  of course one should be well informed before starting. I would
  specially wellcome
  news from Luxembourg, Ireland, Portugal and specially Greece (next
  council presidency, 1st half 2003). I don't know of any efforts 
  against swpats there. 

- People helping these people: translation, artwork, coding, 
  running sites, assembling FAQs, keeping the press informed, 
  etc. This is also a lot of work.

> I myself was in touch with a little 'high-tech' software company
> that
> put a patent on their work (about pseudo QoS on TCP/UDP/IPv4).
> What i can say is that the solution found by this 'high-tech'
> company had non-implemented equivalents in the univeristy (then
> public) 
> and has probably been implemented and improved by other people, 
> included linux kernel developpers (see fair queueing), so the
> patent 
> can there be seen rather just as a marketing action... that
> pays.
> 

You're falling in the common trap we've all fallen at the beginning.
The big problem is not triviality[17]. Triviality is a problem, but it 
is terribly hard to fix without revamping the patent system and 
breaking quite a lot of things. There is a more fundamental problem:
subject matter[18].

Only inventions can be patented, not logical creations or discoveries. 
This is what the directive is going to break (and what the EPO 
has ignored for one or two decades). The difference lies in where is 
the innovation: if the patents teaches new uses of forces of nature
to achieve previously unknown effects, then there might be a patent.
If the innovation is in calculation or organization rules (software,
business methods, math, literature ...) then it should be protected
at most by copyright, not patents. 

If you keep patents confined to the material world triviality would 
still be a problem but not so serious. If you bring patents to the 
sphere of mind then everything becomes patentable and you have deeper
problems than triviality.

> I'm just now not enough educated on the tricks&tips of patents.
> I'm "just a developper, not a lawyer" (quote from.. ?)
> 

We are citizens of a democratic (more or less) society, so we have
certain responsability on the political decisions of our
governments. That's why we should teach what we know as developers to
our representatives and learn (from lawyers) what we must to
understand those decisions.  I don't like politics, but that's like
saying I don't like washing dishes, you can't really choose. Fuster (I
think) said that politics which you don't do will be done against you.

>  Let me put on the table, quickly, what i would like to hear on
> patent's case, 
>  instead of just vaguely follow directions i don't understand,
> because they are
>  not well explained.
>

Sorry, what directions?. 
  
> 1.define cleanly on what exactly software patents can apply.

You mean by the directive proposal?. On anything you can do 
with a computer, so long as you have a competent patent attorney
that writes the patent application according to the ritual. 
(like saying you don't patent a program to add 2 + 2 but a system
comprising a general computer such that given inputs 2 and 2 
outputs 4 in some technical peripheral device - or suchlike bullshit).
There's a book somewhere (I think it's 200 sterling pounds) explaining
how to patent any software under the EPC.

Please note that there is no way to separate software from business
methods, math, or any other logical achievement, since once you 
eliminate the requirement of materiality (known as technicity), 
you are left with abstract entities that can always be encoded in 
software. 

If you mean what the law says, then it says you can't patent programs
for computers as such.

If you mean what the law should say. Well quite what it says, really, 
possibly without the much abused "as such". You might patent an 
industrial process that uses a programmed computer (is the innovation 
lies in the physics, chemical or other material charateristics of the 
industrial process and not only in calculations by the computer). But 
a computer program never infringes a patent, you could for instance
use the same program in a simulator. You would only be forbidden from 
performing the industrial process, not running the program. 

> 2.tell in which way it is supposed to promote sane consequences.

There are not such arguments from those who want software patents, 
or at least I haven't heard them. They only repeat myths long enough 
for people to believe them. Some of the myths:

Patents promote innovation. 
False: In mechanical engineering maybe (we should ask a mechanical engineer), 
but in software there's only one way to compete: innovate. You can't 
compete on production, distribution or raw material costs, and the 
only way not to compete is enjoying a monopoly, such as the one a patent gives.

Patents allow SMEs to get venture capital or compete with big fishes.
False: to the contrary, patents are expensive and distracting, big
companies have many and any piece of software infringes many patents,
so SMEs can't use them against big fishes (even if they can afford
such expensive litigation), because big fishes can always counter sue
with their patents.  Big fishes cross license their patents and thus
can ignore the patents effects (except for parasites with deep pockets
who do not sell software, only patent it [38]). But SMEs have no bargaining
power to cross license, they always lose. Btw, at least in Catalonia
SMEs (or slightly larger, up to 200 employees) employ 2/3 of the
workers in IT [19]. Possibly the same in all Europe?

Software is just another technology, you can't distinguish it from 
hardware so you can exclude software from patentability.
False: we don't want to exclude software exactly. Truely enough all
logical achievement are somewhat equivalent, and it's hard to tell 
whether a spreadsheet file, a postcript document, a VHDL dessigned
circuit or a shop in the web is software or not. But nothing of this
should be patentale so there's no problem. The question is wether 
the patent teaches anything new about the material world, not whether
it is implemented in software or hardware. And the material world 
can be separated easily from the world of mind. Anything that does 
not require experimentation (observation) is not worth a patent.

There are lots of software patents applications, so people want swpats.
False: if the mayor sells guns you may want to get some even if you 
don't want your mayor to sell guns. In fact over 90% of the answers 
to the Comission consultation[20] are against swpats. 

Ok, patents are crap, but if the USA have them, Europe must have them 
too to be able to compete.
False: that pressuposes having patents is a competitive advantage, but
they're a cost and distraction, Europeans want innovation, not
litigation. Let the Americans litigate if they please (but they don't
like it, and will hopefully change it one day, so no use in copying them).
It also pressuposes that patents discriminate on nationality. They do not.
The convention of Paris from late 19th century is stil in place to assure
that foreign citizens are treated as nationals in order to patent. Since
patents are granted for specific states, European companies can and do patent
in the USA if they want to sell products there and think that's useful. 
In Europe neither Americans nor Europeans should be able to patent software. 
That's not discriminatory each market has its rules but all players in the 
market play by the same rules. If anything, it is favourable to Europe, because
we only have to care about patents if we sell to the USA. SMEs can grow 
in a more favourable home in Europe and when they're ready to set up an USA
presence, think of patenting. Furthermore, most illegally granted software 
patents in Europe belong to USA or Japan companies, so European developers
would be in disadvantage if those 30000 swpats are legalised.

> 3.define unsane implicit consequences of patents.

I'll recycle a text from another context (sorry I can't summarize more).

Software patents are unnecessary, counter productive, dangerous and
inconsitent:

1.- Unnecessary

Innovation is inherent in software, since it is the only way to
compete because "manufacturing" (producing copies), distribution and
raw materials are equally cheap for all players.

Free software[33] shows there is enough incentive to innovate and disclose
software creations without patents.

Patents would protect software ideas and concepts. But in software ideas
are easy to get, since the formal models of computers are prefectly
known. The costly part is implementing these ideas in software that
works, and that is already protected by copyright.

None of the actors in software innovation is asking for software
patents, they're either ignorant of any move to enact them or
already opposing them. Only some patent offices, some patent lawyers,
and some of those big corporations that delegate opinion to their
patent departments are asking for software patents.

2.- Counter productive

Software has a great natural tendency to monopolisation, due to network
effects (a program is more useful when many people use it) and other
reasons. Granting 20 year monopolies on software ideas can only shield
and strenghten present oligopolies, harming consumers, quality and
innovation.

Software development is very incremental, and combines many previous
ideas in any project. Patenting software would eliminate the availability
of raw materials and block creativity.

It is unfeasible for a programmer or SME to check their software for
patent infringement. So much so that many standards setting bodies[37] don't
require their participants to diclose what patents they have that may
impede implementation of a standard, because they feel companies would
not afford to search in their own patent potfolio. Let alone making
sure they're not infringinng anyone else's patents.

3.- Dangerous

Ability to patent algorithms and data structures would give private
companies control over too many assets in the information society. By
patenting data formats (or software to encode and decode those
formats), companies could prevent authors from creating or
distributing digital art. Already the 2 most used image formats in the
Web (GIF[34] and JPEG[35]) are patented (so that your photos in the
European Parliament web site might infringe on a patent). The same
goes for audio (MP3[36]) or video (MPEG-4[29]). These and other patent
holders can tax or censor much of the multimedia content in the world.

Since the EPO has already illegally granted more than 30000 patents on
software, mostly to USA or Japan companies, the directive while legalizing
them, would leave foreign companies at a competitive advantage over European
software houses

Since software is information, restricting trade in software is a
restriction in free speech too. Whereas a mechanical engineer will
always be able to publish blueprints of patented engines, a software
expert won't be able to publish blueprints of patented software, since
software is its own blueprint, and publishing it would infringe the
patents.

Similarly, mechanical patents affect (directly) only those having a
factory, while software patents affect anybody having a computer (it
only takes a command to "manufacture" a new copy of a program).

4.- Inconsistent

Patenting software is contradictory: A patent is a monopoly on a
device in exchange for diffusion of (previously unavailable)
information on the device. If you pretend the device to be software,
since software is information you end up with patents that are
a monopoly on information in exchange for diffusion of information.
You can't diffuse information and monopolise it at the same time,
so the deal can't work.

Legalizing software patents in the EU is also inconsistent with the
e-Europe strategy and the will to lead the world knowledge economy in
the next decade. Most current (but hardly enforceable for now)
software patents granted by the EPO belong to USA or Japan
corporations, that would be suddenly left in a dominant position if
the patents became legal. We can hardly win against our competitors by
copying their mistakes. We would be much better having our businesses
innovate while USA businesses litigate (like for instance the patent
suit against e-bay[38] that might stop or damage their activities in the
USA).

Last but not least software patents are incompatible with free
software, since patented software would not be free, and free software
authors could not pay patent licenses per copy since they don't even
try to know how many copies of their programs exist (so much for
shareware too). Free software such as linux[47] or samba[48] is
already suffering from this. Hence, legalizing software patents would
conflict with the recent trend to use and promote free software in
European public admnistration, as proposed in a report by Unisys[39],
for the European Program "Interchange of Data between Administrations"
(already proposed in 2001 by the European Parliament's Temporary
Committee on the Echelon Interception System, for security
reasons[40]). There are various other endorsements of free software in
public administration which would be incompatible with software
patents: UK[41], Germany[42], France[43], Spain[44], Extremadura [13],
Finland[45], Catalonia[46]...


> 3.tell about malicious use, especialy about non patent related
> competiton.
> (see the ashamed 'hyperlink' patent in U.K, if i remember well)

I don't know if I understand the question: 
Do you mean a survey of software patents effects [22]?


> 4.provide a solution that has the same sane goals, if any,
>   and cleaning unsane effects.
> Or, if not possible:
> 5.provide an alternative, which goal is not the same as fair
> patent, 
>   but rather fits the 'meta' of the patents goals.
> 
> Actually, (5) is very interesting and i hope (4) or (5) is in
> our minds.

I'm a bit lost. In fact the Commission goals is to cover the 
irregular behaviour of the EPO (for unknown reasons, maybe simply
for cowardy, because fixing the EPO is more work than giving it 
legal cover). Those goals are of course not sane, so you don't 
mean those, you mean the stated goals of the Commission. Those 
are harmonization of the interior market and clarification of 
a situation with big legal uncertainity. 

Harmonization is not needed because the laws are very very very
similar in all the UE, and there's even one law, the European
Patent Convention that is unique for all EU countries, you can't
make it more uniform than that. 

What might be diverging is the interpretation of the laws in the
courts, since the courts are faced by the EPC on one hand and the
practice of the EPO on the other, that does not follow the EPC.
The cause of this discrepancies is the unjustified practice of the EPO
and that should be fixed, not the laws. 

About bringing certainity, that's a joke. The way to bring certainity
is to have the EPO follow the EPC. The directive would increase uncertainity:

a- They say the only thing clear from years of deliberation is that
there is much confusion about patentability of software (sure, as
always when administrations break the law).

b- They say we cannot put up with this anymore, and a clarification is
needed, everyone seems to agree that the confusion must stop and the
EPO must follow clear and predictable rules.

c- They give only one example of things that would not be patentable,
the 1 click, amazon's patent.

d- (Impressive) They can't even be sure, under the new rules that their
own example is not patentable. It only passes from "who knows" to
"highly unlikely". And they don't think they should give opinion on
that because the EPO is considering the case.

So what the hell are they writing those new rules for? (sorry, it
wasn't they who wrote them, but someone in the BSA). If they don't
feel inclined to interfere with the EPO, that is, they don't hope for
any control of the EPO, and even themselves are not convinced of how
to apply the rules to an example, they've quit any clarification
attempt before starting it.

Fine new rules. The only news is there are no longer rules, and even
that is clear as mud.

If you mean what can we ask to be done to fulfill more sane goals, 
such as promoting innovation, competition and wellfare, then
easy, we need enforcement of the current law (EPC).
You may want to read an alternative to the proposed directive[1]
or a bigger action plan [2]. 

> I am sorry i couldn't hear RMS's Bruxelles speech (at fosdem).
> Would
> his speech have helped me this time rather that in 2001? 
>

I think his speech at Cambridge[21] was better (on swpats).
  
>  I promess i won't drink as much in 2003.
> 
> I hope i don't flood, hope this won't launch a flame war, since
> fosdem list is really
> not made for that. True thanks for fosdem, for sure i'm comming
> back in 2003.
> 
>  Ben

Err.... ok, if the fosdem list was not for that, I guess I'll be 
kicked off the list quite quickly. But I thought ongoing discussion
of issues that were talked in the conference was a purpose of the
list. Specially those that generated so much interest that there 
was an unscheduled meeting with people from FSFE and others.


>  
> Michael De Nil:Michael wrote:
> > last year, after the speech of RMS, I had something like:
> > "let's get on the street. they can't take away our freedom!"
> > 
> > now, half a year later, the issue is still not solved, so
> > maybe we could tell the world what we think about those
> > software patents during fosdem... what about a demonstration
> > against those patents ? ...
> > 
> > there is a little thread about it on be.comp.os.linux, but
> > it's in dutch.
> > 
> > please let me know what you think about this
>

Why not?. Sometimes you tell them and they don't listen, or they
understand and still do nothing. Then you must show them there's a lot
of people who know what they're worth. I can picture hundreds of
hackers marching from the campus to the EU quartier, holding keyboards
(the tool of our craft) with the cover page of an European software
patent wrapped with adhesive tape around the keyboard, to show the EPO
is blocking our work.

But first we have to talk. The demostration will be little use unless
the Parliament and governments know what we propose first.

 
Hello!. Anybody read until here?. Thanks, that's patience. I hope 
you don't mind a few references :


[1] FFII Alternative to the directive text
http://swpat.ffii.org/papers/eubsa-swpat0202/prop/

[2] Call for action
http://swpat.ffii.org/papers/eubsa-swpat0202/demands/index.en.html

[3] Analysis of the Commission directive proposal
http://swpat.ffii.org/papers/eubsa-swpat0202/

[4] Software patents horror gallery
http://swpat.ffii.org/vreji/pikta/index.en.html 

[5]  European Patent Office against public interest
http://swpat.ffii.org/players/epo/index.en.html
http://www.caliu.org/Caliu/Patents/porc.html

[6] FAQ on the Commission directive proposal and press release
http://swpat.ffii.org/papers/eubsa-swpat0202/faq/index.en.html

[7] Members of the European Parliament
http://wwwdb.europarl.eu.int/ep5/owa/p_meps2.repartition?ipid=0&ilg=ES&iorig=home&imsg=

[8] Culture committee in Europarl
http://www.europarl.eu.int/committees/cult_home.htm

[9] Industry committee in Europarl
http://www.europarl.eu.int/committees/itre_home.htm

[10] Juridical committee in Europarl
http://www.europarl.eu.int/committees/juri_home.htm

[11] Codecision procedure for the directive
http://www.caliu.org/Caliu/Patents/codecisio.html

[12] EPO as the Commission speaker in the Council Working Group on
software patents
http://swpat.ffii.org/papers/eubsa-swpat0202/epo020621/index.en.html

[13] Debian distribution of the Extremadura government
http://www.linex.org/

[14] Foundation for a free informationa infraestructure on swpats
http://swpat.ffii.org

[15] Caliu capaign against software patents
http://www.caliu.org/Caliu/Patents/

[16] Statements on swpats
http://swpat.ffii.org/archive/quotes/index.en.html
http://petition.eurolinux.org/statements/

[17] Why are Software Patents so Trivial?
http://swpat.ffii.org/analysis/trivial/index.en.html

[18] Patent Jurisprudence on a Slippery Slope -- the price for dismantling the concept of technical invention
http://swpat.ffii.org/analysis/invention/index.en.html

[19] Telecommunications and Information technologies statistics from the Catalan Government.
http://www.gencat.es/csi/pdf/cat/estadistiques/Oferta_TIC_2001/2_Situacio_actual.pdf

[20] Analysis of the answers to the Commission consultation
http://swpat.ffii.org/vreji/papri/eukonsult00/ 

[21] RMS at cambridge, march 2002
http://swpat.ffii.org/papers/rms-cam020325/index.en.html

[22] Software patents effects
http://swpat.ffii.org/patents/effects/ 

[...] sorry pasted some references and I'm too lazy to renumber

[29] MPEG-4 patent license costs 0.02 EUR per hour of video
http://petition.eurolinux.org/pr/pr18.html

[30] Analysis of the working document on facts
http://www.caliu.org/Caliu/Patents/mcCarthy-fact.html

[31] Analysis of the working document on principles
http://www.caliu.org/Caliu/Patents/mcCarthy-opinion.html

[32] Analysis of the Bakels & HugenHoltz study
http://www.caliu.org/Caliu/Patents/mcCarthy-opinion.html

[33] Free Software Foundation
http://www.fsf.org

[34] Patent on GIF image format
 http://swpat.ffii.org/pikta/xrani/gif-lzw/

[35] Patent on JPEG suddenly appears and forces companies that have
been using it for ages to discontinue it (like Sun's Java) or pay
tribute (like Sony)
News article at The Register
http://www.theregister.co.uk/content/4/26272.html
equivalent European Patent from the EPO public database
http://l2.espacenet.com/espacenet/bnsviewer?CY=ep&LG=en&DB=EPD&PN=EP0266049&ID=EP+++0266049B1+I+

[36] Patent on MP3 audio format. The patentee is now charging
for playing as well as recording MP3 files.
http://swpat.ffii.org/patents/effects/mpeg/index.en.html

[37] Patent policies of the Institute for Electric and Electronic Engineering
and the Internet Engineering Task Force.
http://www.ietf.org/html.charters/ipr-charter.html
http://www.ftc.gov/os/comments/intelpropertycomments/ieee.pdf


[38] e-bay declares losses due to patent suit by patent experts
that do not produce software, only patent it.
 http://news.com.com/2100-1017-956638.html

[39] IDA study: Pooling Open Source Software in Public Administration
http://europa.eu.int/rapid/start/cgi/guesten.ksh?p_action.gettxt=gt&doc=IP/02/1010|0|RAPID&lg=EN
One of the documents talks about software patents:   
Pag. 54-55
http://europa.eu.int/ISPO/ida/jsps/index.jsp?fuseAction=showDocument&parent=news&documentID=550

[40] The UE proposes free software to improve security
Referenced in page 56 of the IDA study OSS market document [39]

[41] Free software in UK administration
http://zdnet.com.com/2100-1104-945784.html

[42] Free software in German administration
http://www.gnupg.org/aegypten/ 
http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2869075,00.html
http://www.computerworld.com/softwaretopics/os/linux/story/0,10801,71659,00.html
http://www.nwfusion.com/newsletters/linux/2002/01383301.html
http://www.pcmag.com/article2/0,4149,16097,00.asp

[43] Free software in French administration
See reference [39], OSS Usage document, pages 29-37


[44] Free software in Spanish administration
Rhodas project of the Ministery for Public Administration, also recommending
avoidance of software patents
 http://lucas.hispalinux.es/Articulos/0000otras/doc-proyecto-rhodas/doc-proyecto-rhodas/

[45] Free software in Finnish administration
 http://linuxtoday.com/news_story.php3?ltsn=2002-06-17-011-26-NW-DP-PB&tbovrmode=1#talkback_area

[46] Free software in Catalan administration
 pag. 37 of the  Butllet? Oficial
del Parlament de Catalunya n. 318
http://www.parlament-cat.es/pdf/06b318.pdf

[47] Linux development so impeded by patents that Linus Torvalds does
not want to know
http://lwn.net/Articles/7001/

[48] Microsoft attacks network free software Samba with their patents
http://www.advogato.org/article/453.html
http://swpat.ffii.org/patents/effects/cifs/
http://www.caliu.org/Caliu/Patents/nota08042002.html


-- 
Xavi Drudis Ferran
xdrudis на tinet.org



_______________________________________________
Osdem mailing list
Osdem на lists.raphinou.com
http://www.raphinou.com/cgi-bin/mailman/listinfo/osdem

----- End forwarded message -----

-- 
/ Alexander Bokovoy
---
Housework can kill you if done right.
		-- Erma Bombeck