|
|
| UK Workshops on "Technical Effect" Wording |
 |
|
Sunday, February 06 2005 @ 05:42 AM EST
|
As you probably have heard, the pro-software patent groups are making headway and intend apparently to continue to push for the software patents directive, despite the JURI vote to restart the process, and Poland seems to be wearing out:The Euro PAP web service of the Polish Press Agency (PAP) writes that the Draft of the Directive on Computer-Implemented Inventions is likely to be adopted on the 17 February meeting of the Council of Ministers of Finance as no country, including Poland, will block the vote. The directive, even if passed this way, may still be revised during the second reading in the EU Parliament. There are those working to define more stringently and precisely what software patents should and should not issue. Those located in the UK may wish to attend one of the series of workshops being hosted by the UK Patent Office:
Peter Hayward, divisional director at the Patent Office, commented: "We need to hear opinions from a broad range of interests - not just from patent attorneys and patent-owning software companies. The views of those software developers who work without patents are just as important to us."
The Patent Office says: "Participants will be shown a variety of definitions for 'technical contribution' and invited to work in groups to test these against a range of innovations. They will also be welcome to propose definitions of their own." The first workshop will be on March 15 in Coventry, and those who wish to attend must register by Febuary 18. Dr Stupid thought it would be valuable to bring this to everyone's attention, so that all of you who wish to can make suggestions that those planning to attend one of the workshops can then take with them to present there.
***********************
What Is a "Technical Contribution" Anyhow?
by Dr Stupid
This is not so much an article, as a call for suggestions. UK-based
Groklaw readers will doubtless have seen on this site and elsewhere
that the UK Patent Office will
be holding a number of "workshops" this spring to discuss the
meaning of the phrase "technical contribution" as it appears in the
draft EU directive on computer-implemented inventions.
[Note for non-EU readers: when a Directive is finally adopted, each
member state must enact it into their domestic law. The interpretation
of a directive's language is thus of more than academic interest.]
If the directive is passed in its current form, then the only obstacle
to pure software patents in the UK will be to persuade the UKPO to
adopt a sufficiently firm interpretation of "technical contribution."
This is not a hopeless cause, since current UK law contains the same
loopholes that the draft directive has, and yet the UKPO has rejected some software patent applications
on the precise grounds that they were not "technical." See here.
If the directive is reworked, either by the European parliament or
otherwise (as I personally hope it will), then the interpretation of
"technical contribution" will still be important in the marginal areas
between software and hardware; so it remains vital that we make our
opinions heard. It was the forceful, but polite, expression of opinion
at the previous
workshop on this topic that led in part to this consultation
exercise by the UKPO.
I know that some GL readers have already registered to attend. Others
may be unable to attend, either for logistic reasons or because places
are limited. Therefore this "story" has been created to enable all
interested readers to post their suggestions. Those of us who are able
to attend can then take your input with us.
The original thread started here,
and I suggest you read the posts there if you haven't already.
Obviously there is no need to repeat yourself (though I should really
take my own advice more often, having just read through my posts.. ;)
One theme that recurred strongly in the comments to date was that the
definition of "technical effect" should ensure that you cannot buy a
stock PC from the High Street, take it home, and infringe a patent just
by writing software for it. Such a definition would at a stroke block
algorithmic software patents (e.g. compression & encryption
algorithms), data format patents (e.g. network protocols, file formats)
and "look and feel" patents (e.g. todo lists, mouse gestures) There may
be other loopholes that need to be closed, though; and a definition
based purely on reference to a canonical "stock PC" has its own
problems of vagueness. So if you want to add your twopennyworth, please
do so.
|
|
|
|
| Authored by: warner on Sunday, February 06 2005 @ 06:34 AM EST |
| Drop tha links and make 'em hot. [ Reply to This | # ]
|
| |
| Authored by: lifewish on Sunday, February 06 2005 @ 06:39 AM EST |
My feeling is that the only good computer patent is one grounded very firmly in
physical reality. Given this, why not substitute "arbitrary Turing machine
with standard peripherals" for "stock pc"?
That would instantly block all the algorithmic and look-and-feel patents, whilst
leaving the door open for stuff like innovative use of handhelds (IIRC there
are, for example, cool plans to use these in education which probably deserve
patenting), or computers embedded in toasters, or whatever. A handheld does not
constitute an arbitrary Turing machine; a toaster does not constitute a standard
peripheral.
There is then the issue of what constitutes a standard peripheral (no toasters
need apply). I'd suggest:
*mouse
*keyboard
*scanner
*plotter
*screen
*arbitrary printer
as being "Turing complete" for input/output. It might be possible to
make these do something that could be considered worthy of patenting, but
certainly no more so than the algorithmic or look-and-feel patents were.
Any thoughts?
---
To err is human but, to really screw up, you need a computer.[ Reply to This | # ]
|
- Stock PC? - Authored by: YetAnotherSteve on Sunday, February 06 2005 @ 07:02 AM EST
- Sound I/O also - Authored by: macrorodent on Sunday, February 06 2005 @ 07:31 AM EST
- Stock PC? - Authored by: feldegast on Sunday, February 06 2005 @ 08:06 AM EST
- Simpler... - Authored by: Anonymous on Sunday, February 06 2005 @ 08:47 AM EST
- Stock PC? This varies over time - Authored by: PolR on Sunday, February 06 2005 @ 10:49 AM EST
- Stock PC? - Authored by: Sesostris III on Sunday, February 06 2005 @ 10:55 AM EST
- So, should Bauer and Samelson have had their 1957 patent? - Authored by: Anonymous on Sunday, February 06 2005 @ 11:29 AM EST
- Coffee pot? - Authored by: MadScientist on Sunday, February 06 2005 @ 02:35 PM EST
- Stock PC? - Authored by: eskild on Monday, February 07 2005 @ 02:15 PM EST
| |
| Authored by: Nick_UK on Sunday, February 06 2005 @ 07:13 AM EST |
I don't think I have even seen any legislation that has
been pushed, coerced and manipulated with so much
importance as what these people are doing as if their
lives are dependant on it.
I really wonder what or who is really driving this to make
it so important to these 'pro-members'.
It a shame we don't live in a democracy that is honest
with the voters.
Nick [ Reply to This | # ]
|
| |
| Authored by: Sri Lumpa on Sunday, February 06 2005 @ 07:33 AM EST |
To see if a patent using a computer implemented part should be granted (i.e. is
not a software patent) they should strike all the computer implemented part from
the patent and see if what is not stricken is still patentable (non-obvious, no
prior art,...).
If it is not then the patent should be rejected because it clearly doesn't have
a technical effect (or one hat isn't patentable ).
If it is then they don't need to include the software part to the patent.
And to make it even harder to create software patent the law should also say
that to determines if a product infringes on a patent any software (or other
nonpatentable part) portion of the product should be removed and the
infringement test be only on what's remaining.
That way if I sell some software over the net (or Free Software distributed over
the net) then if you remove the software part there isn't anything left to
infringe but if somebody uses that software as part of a greater whole that
infringes on a patent (to control a patented process for example) then the
software still doesn't infringe (because it is removed before considering
infringement) but the rest of the greater part might (as it would have if
controlled by a human or a mechanical device).
To implement that the law should say something like:
1. Software cannot be patented.
2. However an invention using a software part can be patented if the non
software part is patentable. In such a case the software part is tolerated for
explanation purposes but is not infringable.
3. When considering whether something infringes a patent, the court shall first
remove from the patent any nonpatentable part like the software part and should
remove from the product any non software part as well. If the non software part
of the product infringes on the non-software part of the patent then the
judge(s) can rule on these parts.
That way you allow companies to put all the software they want in their patent
but if it is:
"a software for doing what's been done for ages without a computer, on a
computer."
Then the only remaining part is "on a computer" which shouldn't be
patentable.
Of course a law implementing that idea would have a more developed vocabulary
but this is the gist of it.
Any thought as to why it wouldn't work, loopholes in it...
---
I do not suffer from insanity; I enjoy every minute of it.[ Reply to This | # ]
|
- Call me crazy... - Authored by: Anonymous on Sunday, February 06 2005 @ 09:00 AM EST
- Workable - Authored by: lifewish on Sunday, February 06 2005 @ 09:05 AM EST
| |
| Authored by: Sri Lumpa on Sunday, February 06 2005 @ 07:41 AM EST |
Let's say that the directive goes to and through the council and comes back for
a second reading at the parliement.
Let's further say that we can use the disdain shown by both the council and the
commission towards the parliement to get some amendments to the proposal during
the second reading (if I was part of the parliement I would want to vote on this
text even if I didn't care about it otherwise given how much both council and
commission seem to want to ignore me).
Let's say that they manage to reinstate all/most of the substantive amendments
that the council got rid of.
What would happen then?
Would the council still be able to reject these amendments and then it comes
back to the parliement for a third reading or would the council have to vote on
the text, accept it or reject it, without being able to modify it?
does anybody knows what the remaining procedure is or where I could read about
it.
---
I do not suffer from insanity; I enjoy every minute of it.[ Reply to This | # ]
|
| |
| Authored by: stutchbury on Sunday, February 06 2005 @ 07:50 AM EST |
FWIW, this was my response on the application:
A
technical contribution should result in physical, measurable result. Software
per
se cannot do this without an interface to the physical world. Perhaps these
physical
interfaces constitute a technical contribution, but certainly not the
software that
drives them. The source code for all software is simply a
design document (or
blueprint) adequately protected by copyright law. Software
can *never* be pysically
implemented and can therefore never make a technical
contribution (or have a
technical effect). Even when the source code is
compiled, it is simply translated
into a different language (one that is
understood by the machine entity *reading*
it).
I don't
expect the UKPO office to agree... [ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, February 06 2005 @ 08:05 AM EST |
This is an important message to all those responding to
the request to register to attend the UK Patent Office
seminar. The patent office seems to be still bent of
obsfurcation in order to push through software as
patentable. Look at the website
http://www.patent.gov.uk/about/ippd/issues/cii-registerform.htm
and you will see what I mean.
In the bottom of the form we have a loaded question the
answer to which the patent office is going to be fed back
to the politicians as proof of what is acceptable as a
patent.
<quote>
Finally, you may submit an answer to the following
question and we will consider including it within the
workshop (response is not mandatory).
To be patentable, any invention must make a 'technical
contribution'. How would you define 'technical
contribution'?:
<unquote>
This is a loaded question because it implies that you have
already accepted the following:
1) That software can be patentable. In UK and most EU
member states, we have been expressly told by politicians
that software will not be patentable. Indeed this is
expressly what the European Union Members of the European
Parliament expressly voted 19 to 2 against just recently.
This question is loaded because it does not refer
specifically to software patents on which the comments are
invited, but to patents in general which are not up for
consideration here. It forces the assumption that patents
for software and other patents will be treated the same,
and that software will indeed be patentable which is
contrary to everything we have been decided by our
politicians and what we have been told by them.
2) That the only issue to be decided is what contributes a
'technical contribution'. The important issue here is what
is to be excluded from patentability (ie. software,
algorithms, data formats, protocols, standards for
interoperability, and scientific principles which the
politicians have told us are not to be patentable), the
definition of software etc. , and the scope of the
proposed patent changes (ie. for which laws changes are
proposed).
It is important all respondants make the loaded question
nature of this question crystal clear in their response
and also make it clear that SOFTWARE ie code of any kind,
ALGORITHMS, DATA STRUCTURES, PROTOCOLS, STANDARDS FOR
INTEROPERABILITY, and SCIENTIFIC PRINCIPLES should not be
considered to be technical innovations or should be
explicitly excluded from patentability.
Also it is worth stating that the patent system should
handle computer-implemented inventions under pre-existing
patent law that covered patents for hardware devices, by
limiting them to hardware devices where the functionality
of a hardware component is replaced by a software
implemented component. Pure software patents should not be
permitted. Nor should any computer implemented invention
patent where the inventive part is not implemented in the
combination of hardware and software. The basis for
determining whether a filing for a patent is to be
patentable will therefore be determined (and the claims in
the patent filing will be submitted) the same way as
hardware patents are dealt with now, but with the
contribution of the software implementation described
in terms of functionality as it is now for hardware
components. Anything other than the functionality in terms
of interaction with hardware that is included in the claim
will not be protected by the patent. In other words if a
mousetrap with a tripwire is patentable, then a mousetrap
with a laser beam which uses software to detect the mouse
should also patentable as a whole, but the software, it's
code, algorithms, data structures, protocols, and basic
scientific principles such as encryption methods will not
be patentable, and their use with computing devices
outside the inventive combination that is patentable for
the mousetrap will not be protected by patent law. Instead
software will be protected by copyright as it has always
been.
This last point is also important, because there is a
genuine need for patents for hardware devices with some
hardware components replaced by embedded programmable
components. What has happened is that some vested
interests have hijacked and expanded the issue of computer
implemented inventions to include software. If the
hardware manufacturer lobby is allowed to patent what they
are reasonable and fair, then the pro software patent
lobby will be isolated. If the patent office is to handle
device patents incorporating software implemented
components, then they must be given a fair, simple and
workable way of handling these. If respondents just say,
"no, we don't want patents" then that will simply make it
easier to push through patentability of software.
The UK patent office is looking to use this consultation
exercise as a whitewash, but if enough respondents
politely raise the above issues, then it they may not get
it.
[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, February 06 2005 @ 08:26 AM EST |
Shouldn't the workshops be held before the vote on the 17th. It seems that the
cart is being put before the horse!
Why won't the EU, instead, have hearing and meetings on this PRIOR to pushing it
thru the EU. Seems strange that you push the software patent law thru the EU
and then you get input on the wording? Go figure. Is that democracy or
feudalism in action? My guess is that it does not sound very democratic at all.
Special interests are the only ones being listened to. I wonder how much money
Bill Gates has spread around to get Poland to change it's mind. Any reason why
Poland will not block this, like before? What does a "fishery"
meeting have to do with "computers" anyway? Sounds fishy to me.
[ Reply to This | # ]
|
| |
| Authored by: geoff lane on Sunday, February 06 2005 @ 08:29 AM EST |
| I've seen programming seriously described as an art,
a craft and even a science
but rarely as a technology (at least by the people who
do it -- advertising copy writers are less restrained.)
It seems to me that
as an art, craft or science, computer programming should be as protected from
patents as any other activity described as an art, craft or science.
For
example, a new and novel tool used create a paint brush could have a legitimate
patent. A patent for the activity of "painting with a paintbrush" should not be
possible.
So, if I invented a novel way to program a computer (say perhaps
by plugging electronic lego bricks together) I might be able to get a patent on
the method. I would not expect someone else to be able to get a patent on
plugging the bricks together in a way I have not previously described as the
potential to do that has always existed.
I think what I'm suggesting is, the
person who invented the paper and pencil spreadsheet could have a legitimate
patent; Dan Bricklin and Bob Frankston who merely implemented that orginal idea as
software should not. The company that invented the "justifying typesetter"
should get a patent (Mark Twain
et al); the company that invented the "justifying typewriter" (IIRC it was
IBM) should not; the person who invented the software equivalent (again IIRC Electric Pencil written by Tom
Swift) should not get a patent.
What we need is a return to the absolute
requirement for "novelty". Some companies have managed to successfully argue to
the patent offices that they should dilute this requirement to the point where
painting a corkscrew a different colour is an innovation worthy of a patent.
As
my sig used to say, "Invention and innovation are not
synonyms".
---
Not using the GPL is not a character flaw.
[ Reply to This | # ]
|
| |
| Authored by: Chris Lingard on Sunday, February 06 2005 @ 08:30 AM EST |
I am not quite so pessimistic on Europe. Putting the patent Act on
the
Finance Council agenda is but an option. The European Council will
meet on
Wednesday and will then consider the motion from the JURA. They
do not release
any details of their meetings for at least a week.
Since the UK
delagate who sponsored the Act, Arlene McCarthy; has been
quoted as saying
"Everyone wanted this looked at again"; the Coucil
members may be loosing their
enthusiasm for the Act.
After the Council meeting this Wednesday they
have at least three options
1) Continue and try and force the Act
though some committee
2) Renegotiate the Act with the lower house
3)
Go very quiet for a month or two
Of these I consider 3) the most
likely.
Within the UK the situation is much more
interesting
The original UK Act was sponsored by Lord Salisbury; after
consultation
with industry and the UK Patent Office. The Patent office here
must
make a profit; so the more patents, the more work. It probably seemed a
good idea at the time.
After much lobbying and bad press, they have now
recognized the current Act
as a vote looser, and may do a U turn. A change of
emphasis on this, would also refects on the UK position within the Council; and
is an example of good things happening from lobbying.
Lord Sainsbury
has instructed the Patent Office to look again, and are holding these meetings.
It is therefore vitally important to get the right ideas across.
My
own simplistic thought are that software is a written algorithm based on logic.
Since this is discovered and not invented, then it is subject to copyright but
not to patent. But we need some good words to suggest
It is further
complicated by the drug industry. A new drug costs millions to develop; I am
not sure that the structural formulae can be patented, since once more it is a
discovery and not an invention.
[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, February 06 2005 @ 08:44 AM EST |
Anything which modifies the behaviour of a piece of hardware without physically
altering the hardware should not be patentable, period.
It doesnt matter that Nokia, Ericsson and Siemens want to patent the cell phone
equivalent of one-click shopping. As long as it's only software it just is not,
and should not be, patentable.[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, February 06 2005 @ 09:38 AM EST |
Someone needs to start a suit in the U.S. against the government to stop them
from issuing illegal patents, including all software patents. We should also
seek relief from all the software patents that have been issued. I guess that
IBM might not like this too much, and maybe not some other companies. Too bad.
Groklaw has marshaled support and stomped on SCO badly. Let's push it to the
next level and see how much freedom we can take back.[ Reply to This | # ]
|
| |
| Authored by: arch_dude on Sunday, February 06 2005 @ 10:46 AM EST |
We need to re-think our reasons for objecting to software patents.
There are two major and separate problems:
1) frivolous patents
The really infuriating problem is frivolous patents. These are patents on
"inventions" that are not novel in any sense or thta are glaringly
obvious to anyone who is "versed in the art" (except, apparently,
patent examiners.)These patents serve to impede and discourage innovation, and
they are a disproportinaly large percentage of software patents.
2) patents on Software.
The problem here is more subtle. There really are novel ideas that are
implementable as algorithms. The patent system is supposed to encourage the
inventor to publish such ideas rather than treating them as trade secrets.
Unfortunately, the software innovation lifecycle is much shorter than lifecycles
in other industries, so a valid software patent protects the invention for its
entire useful life in most cases. Thus, long software patents do not serve
society. In fact, the lifecycle is so short that a three-year patent would be
about right, but the patent system cannot operate that fast, the most
cost-effective solution is to disallow softwrae patents. Note however that
software firms wil revert to the trade secret approach.
My former employer applied for a patent on an invention of mine (Technically, I
applied for it and agreed to assign it.) It's novel and non-obvious to others
versed in the art. If patents on algorithms were disallowed, we could easily
have implemented it in hardware, and in fact, one piece of it was in hardware in
the initially-contemplated implementation. However, it can profitably be
implemented in software for slower-speed applications. We could also have kept
it as a trade secret.[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, February 06 2005 @ 11:47 AM EST |
The theory is that without the governments' grants of a temporary monopoly,
the incentive to innovate is reduced severly or disappears. That's the theory.
But look around, and everyone is announcing something new all the time in
the computing field. As long as Moore's Law (which, paraphrased, suggests
that performance doubles every 18 months) is in full force, there is no
requirement for the goverrnments to create artificial monopolies because
there is a natural obsolescence that means that companies that do not
innovate disappear in favor of those who have written to this cycle's hardware
capacity.[ Reply to This | # ]
|
| |
| Authored by: tyche on Sunday, February 06 2005 @ 11:56 AM EST |
| TECHNICAL CONTRIBUTION
TECHNICAL EFFECT
A software technical effect is
one which, when applied to a specific hardware design, is the only way to enable
that hardware design to function. This definition appears to be broad, so lets
see if I can knock it down.
Hardware: to fit this definition,
it must be sufficiently small and sufficiently specific to only apply to one
function AND must not be something that has already been done or is the ONLY way
to do it. Translation - if a piece of hardware is sufficiently small that it
only performs one specific function (a function which is new and not obvious) it
can be patented.
If the software which runs it is ALSO sufficiently
small that it only performs one specific function AND is also new and not
obvious BUT CAN ONLY RUN ON THAT SPECIFIC HARDWARE, then it can be patented WITH
that hardware, but not on its own. On its own, it would only be subject to
normal copyright laws and restrictions.
This does not preclude
OTHER hardware designs which could achieve the same effect using a different
design and different software.
What I am trying to eliminate, here,
is the use of overly broad definitions for patent purposes. Perhaps what I
should add to the above is that the technical design (hardware and software)
must be demonstrated to actually work. Or, perhaps not, since by restricting the
definition to specific instances rather than general principles, it would be
counter-productive to apply for a patent on a design that didn't
work.
There is the possibility that someone would notice
that my definition sounds like the definition of copyright as applied to
hardware. Congratulations! I am defining "Patent" as the physical expression of
an idea, rather than the idea itself. Just as a copyrighted work is the
expression of an idea (in specific form) rather than the idea itself. To my
mind, this would eliminate someone coming along and creating a patent which
would cover PC's AND THEIR SOFTWARE (stock or otherwise) since they perform more
than just one function. This would also eliminate a patent on just software
alone. The only way that the software could be patented is in conjunction with
the specific hardware. If the software could be used on other hardware, it could
not be a part of the patent. It would only fall under the protection of
copyright law. I would welcome others to knock this down, modify, enlarge, or
otherwise expand on the idea I'm trying to
propose.
Craig
Tyche ---
"The greatest enemy of knowledge is
not ignorance, it is the illusion of knowledge."
Stephen Hawking [ Reply to This | # ]
|
| |
| Authored by: John Hasler on Sunday, February 06 2005 @ 01:54 PM EST |
How about if, instead attempting to ban the patenting of inventions that can be
implemented entirely in software, we make pure software implementations of
inventions non-infringing? That would have no effect at all on most patents.
Many "software" patents would still be useful as long as their optimal
implementation is not pure software. "One click" patents and patents
on algorithms would be pointless. Most programmers would not need to concern
themselves about patents at all.[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Sunday, February 06 2005 @ 04:25 PM EST |
This is a bit of a trap. Sure, go to these workshops -
but question why there should be any patents anywhere near
software AT ALL, don't even give credence to any of their
forced-choice "definitions" of technical contribution.
Letting the enemy frame the debate by starting from "yes,
there will be softwa (cough) computer implemented
invention patents, but our 'technical contribution'
definition will prevent abuse" is dangerously wrong- as
our experience with the little words "as such" with the
EPO shows. You can bet they'll already have thought about
clever workarounds for their various 'technical
contribution' definitions, and/or plot to redefine it
quietly - probably be easy to get the law passed with one
definition, then redefine it to be more widely applicable
to pure software later.
Personally I believe ALL patents are wrong. But software
patents are especially wrong.
The only way I'd tolerate patents if there was a clause to
this effect on the books: "Only patent holders may be
held liable for patent infringement. Those who do not
seek to restrict others by holding patent monopolies shall
not be subject to them".
[ Reply to This | # ]
|
| |
| Authored by: KontinMonet on Sunday, February 06 2005 @ 05:32 PM EST |
I was directed to the following document: New Domestic
Law which apparently clarified the definition of 'technical XX' (where XX is
contribution/effect/matter/advance or whatever).
Unfortunately, I
find legalistic English is generally execrable and this document does little to
help.
But anyway, not patentable: [Section 1.2.(c)] a scheme,
rule or method for performing a mental act, playing a game or doing business, or
a program for a computer; with the caveat: "but the foregoing provision
shall prevent anything from being treated as an invention for the purposes of
this Act only to the extent that a patent or application for a patent relates to
that thing as such" and "In the main, the exclusions are directed to mental,
intellectual, aesthetic or abstract matters, though 'a program for a computer'
does not entirely fit into these characterisations."
I think that
sort of means "except as defined below". So, I read most of that document.
Nowhere, let me repeat that, nowhere are the phrases 'technical
contribution' or 'technical effect' or any (supposedly) equivalent phrases
actually defined in the document.
There are vague references to
German cases, EPO and UK court decisions. On that note, how is the general
developer community supposed to determine what is meant by 'technical effect' by
reading these cases (in German if you're Italian or whatever)?
On
the other hand, there are some lengthy paragraphs in very tortured English of
patents that were rejected because they had no 'technical effect'. One sentence
gives the idea: "Decisive is what technical contribution the invention as
defined in the claim when considered as a whole makes to the known art." Of
course, still no definition of 'technical contribution'.
Later
sentences give a hint perhaps: "...program causes the computer to operate in a
technically different way..." Aaargh! More rubbish! "...operate in a technically
different way..." What does that mean? How does a computer specifically built to
operate in a technically specific way alter that process? That vague definition
simply raises a ton more questions. It's not a useful definition.
Later on, we get the EPO (which, politically, tries its best to get pure s/w
patents and has had its wrists slapped as a result) saying that: "...allowing an
operator to ... feed information into a computer ... and that this is a
technical matter." Good grief. They'll define watching TV as a technical matter
next.
If this document implies the general process leading to a
definition then, in general, the UKPTO (and perhaps all Patent Offices) will
leave it to the courts to decide what is meant by 'technical effect' by deciding
what it is not. This sort of definition by negation means we will still
be navigating minefields for decades perhaps until a definition becomes clear.
This is unacceptable and it should be made clear in the workshops that it is
unacceptable.
In fact, if a defintion for 'technical effect' is
insisted upon, then it must be a clear definition such that the hundreds of
thousands of developers and managers that will be affected in Europe, will
clearly be able to determine whether the next module they write is patentable or
infringes on patents with a s/w component. Such a definition will, I venture to
suggest, be so tortuous that it would be better to drop the idea completely.[ Reply to This | # ]
|
| |
| Authored by: Winter on Monday, February 07 2005 @ 02:47 AM EST |
I have glanced over the contributions (no time to digest them all, really).
I have come to the following summary:
- Software is everything that runs on a Universal Turing Machine. That should
NOT be patentable.
Executive Summary:
If it can, in principle, be done by pencil and paper, it is not technical.
Now assume for the moment that there are things that should be patentable.
- You cannot have copyrights, trade secrets, and patents on the same process or
invention. It is either or.
- Interfacing a Turing Machine (Computer) with the real world can be patented,
but only the Physical Effective part is technical and should be patentable.
Examples:
1 Ogg Vorbis/MP3 are Pure Software projects. However, they are only sensible in
as far as they describe a way to produce Audio that sounds exacly like the
original sounds. The part about human hearing and psychoacoustics is NOT
software. So I would suggest to allow patenting the AUDIO processing, but not
the algorithms themselves. Using these algorithms to construct sounds that are
indistinguishable from the original.
2 PSOLA is a scheme for manipulating the pitch and duration of speech units.
Again, the software is obvious, the Audio was highly non-obvious. It makes only
sense when used for Human speech or music. I think that apects should be
patentable.
3 Holography. Say someone invents a holographic computer display. The software
to run this device should not be patentable. However, the device and the way it
works (by software) should I think be patentable.
The original patent laws were much like this. A computer program was not
patentable, but is combination with some physical device (e.g., a robot arm), it
was.
Rob
---
If a theory explains all the facts, the theory is wrong because some of the
facts will be wrong.[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Monday, February 07 2005 @ 04:11 AM EST |
Please, explain me...
You seriously think that they will make a restricted interpretation of the
"Technical Effect" definition???
:-))
In the text must be clear that the "software implemented inventions"
are only related to custom developed hardware.
Let the "Technical Effect" definition and they will patent also the
software method to print "Hello World!".
[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Monday, February 07 2005 @ 07:04 AM EST |
I don't understand - the documents from the UK government say that one of the
prime reasons for the Directive is to harmonise law and practice across EU.
So how can the UK DTi consult on *it's* interpretation of technical effect?
Surely unless the same interpretation is applied EC wide this defeats
harmonisation - or is this just PR spin?
Confused of England
[ Reply to This | # ]
|
| |
| Authored by: TiddlyPom on Monday, February 07 2005 @ 07:19 AM EST |
I get the feeling that the Software Patents war is just beginning. IMHO patents
are being used by large companies such as Microsoft, Yahoo and AOL to squash any
potential competition from other smaller companies e.g.
Yahoo vs
XFire
Matsushita vs Justsystem (only real commercial
competitor to MS Word in Japan)
Just as worrying (IMHO) is the
re-emergence of 'Trusted Computing' - which is back door way to trying to rid of
any software competition (e.g. Linux) and small hardware vendor
competition
Chip
firms won't sell trusted computing to integrators
Trusted Computing Platform
(TCP) on Linux
I get the feeling that most people who use Windows on
PCs do not realise the dangers to freedom of choice and user rights that
software patents and TCP coupled with total lockdown DRM pose. Having said all
that, an older collegue (who is very pro-Windows and has been very scathing
about open source software and Linux) suddenly realised that he couldn't copy
tracks that he owned from a Sony mini-disk system into a PC because Windows and
the mini-disk prevented him from doing so...and asked me if this was possible
under Linux. So perhaps perceptions are starting to change (a little too late)
after all.
Is there anything that those of us who live in the UK can do
to try and stop all this (as the UK MPs do not seem interested and seem (to me)
to regard opposition to the patent laws as by being only by people who want to
break the law by pirating music, software or videos)
---
"There is
no spoon?"
"Then you will see that it is not the spoon that bends, it is only yourself." [ Reply to This | # ]
|
| |
| Authored by: modulus on Monday, February 07 2005 @ 07:42 AM EST |
First I'll build up a few things so that the definition is clearer and so you
can see where it comes from.
Computer: finite state machine which performs symbolic manipulation.
Computation: process by which input symbols are manipulated and output symbols
obtained. See UTM models, etc.
What is a technical contribution? For an effect to be technical it must:
1. Use specific forces of nature.
2. Have specific effects in the world.
So, some examples: mp3, let us say.
The mp3 patent is about a way of manipulating symbols. It does not use specific
forces of nature, as you can do it with pen and paper if you're so inclined, and
with several different mechanisms (the FFT might be done in software on a
general purpose computer or hardware).
So, it fails test one.
One-click shopping: does it use specific forces of nature? No. The click event
detection can be embodied in a plurality of ways. Fails test 1.
So, what do you think? Perhaps test #2 is not even necessary, I put it there
since it seemed a good idea.
[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Monday, February 07 2005 @ 10:08 AM EST |
| This coming from an American (you know, we have more lawyers per capita that
anywhere else in the world?).
Scare the b******* off.
Go ahead and allow
patents on anything and everything related to computers. But throw in a hook.
Strengthen the prior art part of the patent process.
If any patent claim is
found in court to duplicate any functionality of any product available before
the patent was filed, and if that product was not listed and accurately
described in the patent application, or if any other relevant prior art
is found that is not listed in the patent application,
then:
Overturn the entire patent
The
patent application in its entirety then becomes prior
art
The patent holder is held liable for all costs,
legal and otherwise, incurred by the other party in the suit
See if that wouldn't deter the kind of patents Microsoft has been claiming
in the last few years. No more, "Look at this neat feature X has had for 20
years!"
No, it doesn't really solve all the problems with patenting software,
but it would certainly clean up the worst offenders.
[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Monday, February 07 2005 @ 12:53 PM EST |
Programmers, software authors, think in code. Consider
them like musicians or story authors. One can copyright
what one puts to media. Can a musician patent a particular
refrain or measure such that no one else can ever produce
a similar set of sound with out a licence? What about a
story plot? Think about it. How many unrelated books have
similar plots? What about movies? That is the where
software patents fall. Software patents completely stifle
a software author's freedom. That is so wrong.
Software is just an expression of an idea that can be
realized on a computer. Nothing more, nothing less.
Everybody has ideas. Programmers have the ability to
express their ideas in software just a musicians have
the ability to express theirs in music.
Copyrights are just fine. Software patents are just plain
wrong. My guess is that they will eventually fail
regardless because of the creative nature of humans.
[ Reply to This | # ]
|
| |
| Authored by: Anonymous on Wednesday, February 09 2005 @ 09:31 AM EST |
Discalaimer: IANAL I am a programmer however!
Software unfortunatley blurres lines in peoples perceptions between arts,
crafts and science (unpatentable) and technology and engineering which are
patentable.
My view of software is that it is a recipe for how to solve a problem.
This analogy is easily extended: If software is a cooking recipe then hardware
is the the kitchen utensils (I'm trying not to use 'hardware' here again) such
as cookers, pans, toasters and knives. I'm including in the recipe both the
ingredients and the technique for preparation - what would be printed in a
cookery book.
In both situations innovations and inventions for hardware (or kitchen utensils)
should be patenetable - you've invented a better toaster, fine!
However recipes and software shouldn't be patentable - they should however be
copyrightable. Could FastBurger patent their techniques and ingredients for
their "SuperFastBurger"? - I hope not. If they were to print a recipe
book then their recepe would be copyrighted however. Could they patent their
"SuperFastBurger Grill" - I can't see why not if it stood the usual
test for a patent.
I make gravy to go with a roast meat in the roasting pan from the meat rather
than a saucepan. I'm not the only one to do so but if I were would I be able to
patent that technique of using the hardware differently to cook a meal - no.
Or if I came up with a way to make puff pastry in 30 seconds? The same should
apply to software patents.
The analogy extends even further - if we view the ingredients to be the same as
data to be processed then we see that the interfaces to that data (so to know
how to core an apple or for how long to boil a particular type of rice) should
also be unpatentable because again it is essentially a recipe for one particular
subset of the whole. This would mean that header files and the like aren't
patentable but a specific implementation of a structure that interfaces with
hardware is elegable for copyrights. There is nothing to stop someone else
coming along and implementing a different functionally similar structure and
that beeing free from any ownership by the first party.
So in short the whole notion of software patents baffles me :-) It's simple,
it's something that in many other walks of life most people can intuit.
The problem is many people use the term software without fully understanding
what it actually is or wanting too. Unfortunaltey some greedy capitalists have
decided to try to exploit this. IMHO anyone high up in a company organisation
has by defnintion to be a greedy capitalist because they have a duty to the
shareholders to maximise profits. I believe this is why the FUD can continue
and those of us enlightened enough to undestand software properly can't really
see why it's not a no-brainer.
[ Reply to This | # ]
|
|
|
|
|
