Subject: Re: India's Software Patents
From: Bernard Lang <>
Date: Tue, 28 Dec 2004 20:08:48 +0100

* Taran Rampersad <>, le 28-12-04, a écrit:
> Bernard Lang wrote:
> >This is interesting since India will not get any advantage from
> >this. Quite the contrary (I would better see China in that position).
> >  
> >
> Actually, I'm not sure. A lot of software is being outsourced to India,
> as we all know, but it's a matter of time before they decide to cut out
> the middle man (as China is doing with hardware). All it would take
> would be some motivation for entrepeneurs to assure that code developed
> in India belongs to Indian companies. Suddenly, this does make sense on
> a large scale.
> I wonder how much of this decision has to do with the rest of Asia?
> >The major question here is that they seem to restrict software patents
> >to embedded software.  This is _extremely_significant_.  I wrote a
> >paper that addresses (among other things) this issue (unfortunately,
> >the english version is not ready yet).
> >  
> >
> What language is it in? French? I can muddle through with an automatic
> translation for now...

The general idea of this section of the paper is the following.

One of the major problems with software patents is that they introduce
heavy costs and risks in an activity (software creation) that need not
be commercial.

It need not be commercial because the goods created are non-rival and
can be duplicated and distributed at no cost - provided that some
other mean is found for the initial investment ... and exprience shows
that it is sometimes possible ... see free software or wikipedia.

When duplication has a cost, then no one can afford to make the goods
free, because someone has to pay the duplication, which limits anyone
power to distribute freely to all without going commercial.

Patents introduce a cost and risk that no longer allows free
distribution of non rival goods without going commercial, killing in
effect a new mode of creation and distribution of wealth.
[this should actually be detailed a bit ... ]

But if you consider as embedded software only software that is
embedded in a non programmable way (for example requiring changing a
non commoditized piece of harware to be modified), then you are in
effect in a rival economy, and patents have a less significant effect
on this economy.

Of course, if someone makes the same equipment with included software
that can be changed by anyone without touching the hardware (changing
a CD is not changing the hardware, because it is a programmable
commodity), then the software would no longer be considered as
embedded and the patents would not apply.

This is to some extent a trade-off.  The industry guys can choose to
have programmable devices, but they can no longer have effective
patents on the software.  They can choose to have non programmable
devices (possibly modifiable through the change of a non-commoditized
piece of hardware, i.e. non programmable piece of harware), but this
will mean greater cost to them (and make the goods somewhat less
attractive, though not much).

They cannot enforce the patent if anyone can build a competitive
hardware device, and use the same software approach (beware
copyright) on a programmable device.

However, if they hold a patent on the hardware, or on some way of
using that hardware, then they can enforce that patent and prevent
competition ... and prevent also reprogramming of the hardware without
their permission (since they have to provide the support).

(comments or references to similar ideas are welcome)


Now, it seems that the India law has nothing to do with this.  It is
actually very unclear, and very open to interpretation (see Arnoud
Engelfriet post).

* Arnoud Engelfriet <>, le 28-12-04, wrote:
> I found the text as PDF online at
> which was linked on the frontpage of
> The relevant article in Indian patent law is article 3, which
> defines excluded matter. Clause k of this article will be amended
> to read:
>    "(k) a computer programme per se other than its technical application
>    to industry or a combination with hardware;
>    (ka) a mathematical method or a business method or algorithms;".
> The "computer program per se" sounds very much like the European
> "computer program as such". I wouldn't be surprised if someone
> took the European wording, translated it into Hindi and then
> back into English. :-)
> But here they explicitly state that a "technical application"
> of a computer program to industry, or a combination of a computer
> program with hardware is not covered by the exclusion. 

I would not interpret it thus.  But I do not have the full text of the
original act that is modified.  The only reference I found with Google
is not freely accessible.
   This is possibly the text, but I am not sure :
  and the k and ka do not fit too well.

Basically, they leave the interpretation very open, as usual,
depending on the interpretation of what technical means.

What could be a technical application to industry that is not in
combination with hardware ?

Why do they say it applies only to embedded software ?

seems like the IP lawyers have tried to hide the issues by
deliberately lying to the public and probably to the politicians
... as is customary with them (sorry for the few honest guys in this
business ... your colleagues are crooks).  We see the same in France.


> >Depending on the definition of "embedded" that they chose, this may
> >affect free software in very different ways.
> >
> >Does anyone have more details about the actual text, and the chosen
> >definitition.
> >  
> >
> I wish I did. With the tsunami, it's hard to get anything out of India
> other than things about the tsunami right now. A white out... :(
> >I am also concerned as to how the manage to limit the effect of
> >patents to embedded software, and remain compatible with the TRIPS
> >constraints.
> >  
> >
> I'd love to know this as well.
> -- 
> Taran Rampersad
> "The wave of the future is not the conquest of the world by a single dogmatic creed
but the liberation of the diverse energies of free nations and free men." ? John F.