There are some similarities to this thread with one that happened last year.
Thankfully, this one is less acrimonious than that one :-) .
The thread is available here:
http://www.ae.iitm.ac.in/pipermail/ilugc/2005-September/020946.html
The similarity in issues raised comes further down in that thread. Am
quoting myself from that thread below - an arguement similar to what Prem
has made in this thread:
"A software company does not patent any of its innovations (cute ones
included) and releases all its code to the general community. If the
innovation is of value there will be dozens of other companies that seek to
provide support for the same s/w undercutting prices to a level that throws
the original company out of business (if it is not properly managed). After
a period the market will only have highly efficient players who can account
for the productivity of every penny.
Now, the question is: Can these super efficient players afford sinking
money in research projects that might (just might) give them the next killer
app. Since developing a killer app is unpredictable, can these companies
stay in business till their research team comes out with the 'killer'? Can
these companies do a Bell and employ the likes of Ken Thompson, Dennis
Ritchi, Kernighan, Pike etc and not care for 5 years or so about what value
they have added to their revenues? Would not the other efficient players
wipe such a company out of the market in a year? How will the free thinking
of great minds be funded? And even if such a company gives out such a
product will it have sufficient time to recoup investments before the others
catch up (the others can be super efficient because they did not spend a
penny on research and have got the source code for free)?"
"Physical product companies do not make that big a noise about reverse
engineering because the infrastructure cost of replicating the product is
itself a barrier to competition. Where the barrier is not strong enough
product patents come in. Example, a Ferrari model is not patented but the
suspension subsystem might be patented. So, if patents are OK in other
fields (where there are other barriers to copying) why is it a bad idea in
s/w (where the barrier to copying is weak - by copying I mean rewriting the
code... not pirating)?
If the source code of a s/w is to be given out and also there are to be no
s/w patents, how will a software company ensure that the product it has just
released in the market will earn them enough money to pay for more research,
give the programmers a pay hike (the biggest cost for a s/w company) and
offer better returns to its investors? Will not another company lift the
code make fancy modifications to it (just enough to fool the courts) and
sell it at a much lesser price?
Of course, it can be argued that it benefits the end user by being less
expensive! But then, the govt giving free food, free clothes and free cash
also benefits end users. But does it benefit society?"