[Lancaster] s/w patents

[Andy Baxter]

--Boundary-00=_liXW/yRKn91cem2
Content-Type: text/plain;
  charset="us-ascii"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

Terry Wynne mep sent me the attached briefing he'd been given by the patents 
lobby a few days ago, asking for my comments. I replied last night, and I'm 
planning to write some more over the w/end. All I have done so far is check 
out for myself the argument the ffii are giving against it, and it looks like 
they're saying some important things that are not being listened to. Any 
comments welcome,

andy.

-- 
Please don't send me html mail or un-notified attachments. These will be 
automatically filed under 'probable spam' unless I'm expecting an email which 
hasn't come.
If you do need to send an attachment or html mail, put [attachment] or [html] 
in the subject line.
Thanks, andy.

--Boundary-00=_liXW/yRKn91cem2
Content-Type: text/plain;
  charset="us-ascii";
  name="wynne-briefing.txt"
Content-Transfer-Encoding: 8bit
Content-Disposition: attachment;
	filename="wynne-briefing.txt"

Supplementary briefing for UK MEPs – 1 September 2003

DIRECTIVE ON PATENTABILITY OF COMPUTER-IMPLEMENTED INVENTIONS (“SOFTWARE PATENTS” DIRECTIVE)

 

Background
In February 2002, the European Commission presented a proposal for a Directive on the patentability of computer-implemented inventions (the “software patent” directive). 
The United Kingdom and all other Member States supported the Commission’s proposal, designed to make clear the boundaries of existing patent law on software.  There was strong consensus in the Council on a draft agreed in November 2002, which aimed to improve the impact of the Directive. 

Why do we need a directive?
Patents exist to protect technical innovation. Increasingly, technically innovative manufactured products (ranging from machine tools to automobiles to household appliances) have computer programs embedded in them. The law needs updating to cope with the proliferation of these new inventions so that European industry is able to protect its investment and remain competitive with its US and Japanese rivals. As such, the directive has important implications for many sections of manufacturing industry, as well as the software, telecoms and broadcasting sectors.
In particular a directive is necessary to distinguish between inventions which use computer programs to make machines work better and “pure” software which merely manipulates abstract data within a computer. The former needs patent protection – the latter does not.  
At present the law is hazy, which is a disadvantage to EU companies operating in innovative sectors, especially SMEs.

What will the Directive achieve?
The fundamental aims of the proposed Directive are twofold:
	To clarify and confirm the current legal position, and so minimise the confusion as to what can be patented, and what cannot be patented.
	To put down a legislative marker to prevent any extension of the patent system towards a more permissive regime that would potentially undermine the principle that patents exist to protect technological innovation. 

What the Directive would NOT do - 
The proposed Directive would not make anything patentable that is not already patentable in the UK and elsewhere in Europe.  It would not extend patent rights to new areas, nor does it provide extra protection for patent holders. It emphatically does not follow the American model of patents for all software.

What the Directive would do -
The draft Directive would set out the current position on software and make clear – as far as is possible in a fast-changing environment – what is and is not acceptable for patenting.  
It makes clear that in order to be patentable a computer-implemented invention must make a technical contribution (in other words must make a machine work better). By providing greater certainty, it will allow software developers, owners and users a clearer legal basis on which to work.
Existing European law, while saying that software ‘as such’ is not an invention for patent purposes, is qualified by exemptions - upheld by UK and European courts – that mean that programs can be patented if they include innovation in a technical field .  The Directive would help to settle this situation in the light of 25 years’ development of software applications since the European Patent Convention came into force.

Difference between copyright and patents
Copyright is automatic intellectual property protection for the expression of a certain idea (such a song or lines of text). The text of computer programs benefits from copyright protection. The directive would not change this. The code for computer programs and other software is covered by copyright law and an earlier Copyright Directive (91/250/EEC): this also allows people to develop programs that will connect with other programs through an ‘interoperability’ article.
Patents are granted for finding technical solutions to particular technical problems – a new machine, drug or engine part. The conditions for granting a patent are much stricter than for copyright, to ensure that only truly innovative ideas are granted patent protection. 

Why are so many people worried about this directive?
Opponents of the patent system, and in particular those who consider that copyright provides sufficient protection for computer program-related inventions, maintain that the law as it stands excludes computer programs from the field of patentable inventions. Consequently they argue that by confirming that computer programs can be patented in some circumstances, the Directive would be extending the boundaries of patentability.  However, this view, which is widespread, is based on a misunderstanding of the law.
Opponents also argue that a lack of definition of the legal test of ‘technical contribution’ weakens the Directive.  However, it allows the Directive to be understood within the framework of current patent law and allows the decisions to move with changing technology: any over-specific definition would be out of date very quickly.



Contentious amendments
Art 5(2) allows programs to be protected when transferred by disk or over a network.  This is important to inventors, who need it to protect against piracy: without this article, a patented invention could be stolen by a third party and sold on to other users – and only the end-users could be subject to legal action with the third party safe from infringement action. Without Art 5(2) the Directive may be in breach of a WTO treaty since it discriminates against inventions in one field of technology.

Art 6(a) aims to extend the ‘interoperability’ provisions of earlier directives, but in effect it strips out patent protection from a swathe of industries (including telecommunications and broadcasting) by referring to “communication and exchange of data content”.  Previous directives do not attempt to say that such uses are “not considered to be a patent infringement” and the article itself appears to be designed to deal with competition and market-dominance issues rather than patent issues.  Art 6a would undermine the whole principle of the Directive and should not be supported. 


Examples of what can currently be patented and what cannot
Here are some examples of software-related innovations that would be patentable:

	Smart vacuum cleaner that adjusts suction automatically
	Faster calculations by linking several computer memories
	Improved control of airbags in cars
	Better automatic speech recognition
	Joystick for computer games with more controls

and here are some that would not:

	Program for managing litigation
	Faster calculations by using a new mathematical formula
	Improved automatic translation software
	Simpler financial management program for SMEs
	Database allowing improved the targeting of your constituents
	New characters for a computer game

However, something in the second list may be patentable if, unusually, it involves technical innovation (for example, if the new games character in example 11 were displayed more realistically by a new way of controlling a computer screen).  Conversely, something in the first list would not be patentable if it involves no technical innovation. 



--Boundary-00=_liXW/yRKn91cem2
Content-Type: text/plain;
  charset="us-ascii";
  name="mail-to-wynne2.txt"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
	filename="mail-to-wynne2.txt"

=46rom andy@earthsong.free-online.co.uk Fri Sep  5 09:18:51 2003
=46rom: Andy Baxter <andy@earthsong.free-online.co.uk>
Reply-To: andy@earthsong.free-online.co.uk
To: "Terence Wynn" <twynn@europarl.eu.int>
Subject: Re: Computer Software Patents
Date: Fri, 5 Sep 2003 09:18:51 +0100
User-Agent: KMail/1.5.3
References: <sf54f2c7.040@mailbrugt1.ep.ec>
In-Reply-To: <sf54f2c7.040@mailbrugt1.ep.ec>
X-KMail-Link-Message: 2639
X-KMail-Link-Type: reply
MIME-Version: 1.0
Content-Type: text/plain;
  charset=3D"iso-8859-1"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Message-Id: <200309050918.51086.andy@earthsong.free-online.co.uk>
Status: RO
X-Status: S
X-KMail-EncryptionState: =20
X-KMail-SignatureState: =20

Terry,

Thanks for sending me this. I have done a bit of research on the internet o=
ver=20
the last few days, and feel somewhat better qualified to give the comments=
=20
you asked for.
I should say a couple of things before I start. Firstly, I am no expert in=
=20
this area (though neither is Arlene McCarthy, not having I imagine spent mu=
ch=20
of her adult life working in the fields of patent law or information=20
technology). I have a good degree in natural sciences from some years back,=
=20
and computers have been a side-interest for me since I learnt to program as=
 a=20
teenager, but there are no doubt other people with more understanding of th=
e=20
issues here than I have. I have some programming experience, but not=20
professionally. I also spent some time a few years back studying for a PhD =
on=20
social aspects of science and technology, which I did not complete. Secondl=
y=20
my personal interest in this issue is that I am currently using and learnin=
g=20
to program with the linux operating system, which as you may know, is=20
released as free, or 'open source' software. I believe that if this directi=
ve=20
is allowed to go through unchallenged, it will have a bad effect on open=20
source development, which I do not wish to see happen. Thirdly, most of the=
=20
research I have done has been through the website produced by the ffii=20
(swpat.ffii.org), who are one of the groups campaigning against the draft=20
directive; however I have tried here to keep some seperation in my mind=20
between the information on their site which would be agreed by all as a=20
matter of fact (which I have tried to check against other sources), and tha=
t=20
which is their own opinion on the situation.
I have sent a copy of the briefing to the uk group of the ffii=20
(www.ffii.org.uk), as I think that they have a right to know how this issue=
=20
is being presented to meps. I hope that this does not cause trouble for you=
,=20
seeing as not all meps would be as open in sending something like this to a=
=20
stranger.

I have had a quick look through the briefing which appears mainly to set ou=
t=20
the claimed intentions of those who have drafted the directive, but goes in=
to=20
no detail about how those intentions will be carried through in law. If the=
=20
draft directive were to have the effects that they claim, then there could =
be=20
a reasonable argument for saying that it is a fair effort to update previou=
s=20
law in the light of technological changes since it was passed. However, it =
is=20
far from clear that this is the case, i.e. that the directive will indeed=20
have these effects.

The following points seem to me to be fairly clear as a matter of fact from=
=20
the research I have done. I have included web references to where you can=20
look into some of these points yourself.

=2D Article 52 of the current European Patent Convention specifically exclu=
des=20
software, business practices, mathematical and scientific discoveries, and =
a=20
few other things from patentability.
http://swpat.ffii.org/analysis/epc52/index.en.html
http://www.european-patent-office.org/epc/pdf_e.htm

=2D However, computer software is currently protected under copyright law i=
n=20
Europe, which protects the particular code of a program against direct=20
copying. So there is already a barrier to the copying of creative ideas in=
=20
this field, namely the effort of reproducing the functionality of the progr=
am=20
in your own code which can be shown in a court of law not to derive from th=
e=20
copyrighted work.

=2D In spite of this, many pure software patents have been granted by the=20
European Patent Office over the last ten years or so, as well as some paten=
ts=20
on business methods. I have checked some examples of this from the EPO's ow=
n=20
archive. Those of these I have seen clearly (to my eyes) fall into the=20
category which the presentation says should not be patentable - "pure=20
software which merely manipulates abstract data within a computer" rather=20
than "inventions which use computer programs to make machines work better".=
=20
At least in the way that these words would be understood by most people wit=
h=20
a knowledge of computers. And as far as this distinction can be made at all=
 -=20
it could be argued that Microsoft Word (a program) makes a machine (your=20
computer) 'work better'. And a mobile phone (a machine) works the way it do=
es=20
because there is a program running on the computer inside it which=20
manipulates abstract data in its internal memory, as programs are wont to d=
o.=20
Many pieces of consumer technology nowadays contain a computer cpu chip whi=
ch=20
works in more or less the same way as the cpu of the machine you are readin=
g=20
this at, and the information processing algorithms developed for these are=
=20
equally likely to find applications in programs on home or business=20
computers. The point being that these are the sort of arguments a good pate=
nt=20
lawyer is likely to use to try to bend the rules, so any law trying to draw=
=20
boundaries of this sort needs to be well thought through and carefully=20
worded. Some of these patents are also absurdly broad or trivial (to my=20
eyes), for example the tabbed palettes one (trivial), or the network sales=
=20
system (broad).
See http://swpat.ffii.org/analysis/testsuite/index.en.html for the examples=
=20
the ffii have chosen to test the proposed law against, including the above=
=20
two.
See http://swpat.ffii.org/patents/samples/ep927945/index.en.html for the=20
Amazon gift-ordering patent, which is clearly for a 'business method', and=
=20
closely related to the 'US-style' one-click American patent which they used=
=20
in a suit against Barnes and Noble's online store. B&N won the case to have=
=20
Amazon's injunction lifted, but then settled out of court before it came to=
=20
the trial on the actual case, and neither side will reveal the terms, so it=
's=20
not clear who won in this. (Apart from the lawyers...)

=2D This being so, if the intention is in fact to prevent a move towards=20
american style patenting (which many would argue has already happened), one=
=20
would expect that the draft directive would contain language that firms up=
=20
the existing law excluding pure software patents, and gives it some teeth i=
n=20
the face of wilful misinterpretation.

=2D However this is not the case as far as I can see - the effect if anythi=
ng is=20
to make the law less clear and more dependent on the interpretation given t=
o=20
it by the patent establishment.
Comparison of BSA Draft, CEC Draft, and FFII counter-proposal:
http://swpat.ffii.org/papers/eubsa-swpat0202/kern/index.en.html
I have not yet found a copy of the original CEC draft directive on the eupa=
rl=20
site. There have also been several amendments since then I think, which I a=
m=20
intending to look into.

=2D The draft directive draws heavily on a proposal by the Business Softwar=
e=20
Alliance (BSA), an American organisation of which Microsoft is a principal=
=20
member. See:
http://swpat.ffii.org/players/bsa/index.en.html
and above comparison of bsa and cec draft documents.

These are also the main points made by the ffii in their case against the=20
directive. I have done my best to check these out for myself, and it seems =
to=20
be that they have good reason for being suspicious about the likely effect =
of=20
passing the directive. There are still things I would like to look into=20
further on this, and I am intending to make some more detailed comments on=
=20
the briefing you sent me; however this is unlikely to be before Monday=20
morning, so I've decided to send this now so that you have something to loo=
k=20
at before the weekend.

andy.

=2D-=20
Please don't send me html mail or un-notified attachments. These will be=20
automatically filed under 'probable spam' unless I'm expecting an email whi=
ch=20
hasn't come.
If you do need to send an attachment or html mail, put [attachment] or [htm=
l]=20
in the subject line.
Thanks, andy.



--Boundary-00=_liXW/yRKn91cem2--