Closed-source considered harmful (was: JavaScript considered harmful)

Clark C . Evans cce at clarkevans.com
Sat Jan 12 15:50:44 CET 2002


IMHO, Closed source software is bad for one reason, it
lacks "competition for the change of software".   That is,
given significant investment in a chunk of software, the
user wants to "upgrade" not buy a competitor.   With closed
source software, the user is limited to one vision for 
their upgrade... there is no competition of ideas.   This
is also called the "freedom to fork".   The freedom to fork
doesn't usually cause forks to happen, it is the constant
thread of a fork that keeps the software responsive to it's 
user community.   

That the end user can read/learn from the source code is 
largely irrelevant.  It's a nice side effect.  The primary 
problem is economic.  Software replaces business processes 
and becomes inter-twined with a companies culture and is hard 
to separate for not only data conversion issues, but also 
integration with other systems and with humans (we call the 
latter re-training).  This lock-in is huge.  Thus, you can't 
ask a customer to simply switch products.  This is an abuse 
of the foundations of copyright's justification in the 
constitution (since it limits innovation).  The only real goal 
is to have competition for the upgrade...  and competion of 
this sort doesn't happen with closed-source software.

Therefore, all said and done, closed software is bad beacuse 
after the initial purchase decision (in which there is vigorous
competition), closed software leads to non-competitive markets.
And non-competitive markets is the antithisis of our economic 
philosophy as described by Adam Smith.

Kind Regards,

Clark Evans

P.S.  Open source is not the only way to have "competition for 
upgrades".   Open source mixes free beer (gratis) with freedom 
(libre).  With a properly designed copyright mechanism, one
could have the latter without requiring the former.   See 
DistributedCopyright.Org for more details.




More information about the Python-list mailing list