From marmalade@x-stream.co.uk Fri, 30 Jun 2000 20:31:58 +0100
Date: Fri, 30 Jun 2000 20:31:58 +0100
From: Nick Fleming marmalade@x-stream.co.uk
Subject: [Patents] technical effects in computer software.

Dear All,
    As a computer programmer by trade, I am a little amused by
the term 'technical effect' that has been talked about recently.
While I am certain that Steve Probert knows a lot about
patents, after reading that he thinks a curry receipe is NOT a
computer program, (even meant as   joke) I think I should point
a few things out before everyone drifts too far into neverland.

    A curry receipe can be a computer program. If it is written in a
way that the computer can understand, and yes, this could be a plain text file,
or even an email, then it can be regarded as a computer program.

For instance:

   title:    "Curry Receipe"
    ingredients:   (Curry Powder; water; sugar);
    Action:   group(all), mix, eat (optional)

    Now. Just because it is *not* compiled, and is human readable does not exclude
it from being a computer program. This is an important point - given suitable hardware,
*ANYTHING* that can be represented on the computer, in any format is potentially
a computer program. I suppose the technical effect as defined by Mr Probert would
be due to the inputs and outputs of the program, (depending on how much curry powder you used :-).

     However, as every computer science student knows, all programs have inputs and outputs.
My point, is that every computer program ever written produces a technical effect
of one sort or another, whether it is printing sideways or mixing ingredients.

    If you limit a technical effect to only the interface between man and machine, then that
is easy to get round : surround any program that contains (potentially) patented material
by calling it with one program, and outputing the results to the real world via another program.
Since the interface is then machine to machine, then If this sounds like unix - it is. 

    I am not against patents, nor would I advocate breaking the law. I would however like
to point out that computer software does not easily lend itself to being patented. There are no
nice dividing lines between what is and is not patentable with software. Software was excluded
from patentability for this reason  I believe. If anyone has a nice dividng line then please say
so, and I'll shut up (maybe..)

    Leaving the decisions to the law courts, without the assistance of a qualified software engineer is like getting me to
cook a decent meal.. (see receipe above..:-) You get the ingredients but not the expertise.

    Nor can you argue that the form of a program is  important, as pointed out in  a previous posting.
In fact, about the only thing you *CAN* do to prevent monopolisation on a worldwide scale, via
software, would be to shorten the time allowed for software-related patents to about five years. 
Given the rapid pace of change in the computer industry, usually anything five years old is out
of date anyway, and would also help prevent developing countries from being at a technical
disadvantage in the future, and still give businesses and investors the security they need when 
developing their market share.

    Regards,

Nick F.

P.s. For the record, I don't like curry!







