"open standards", RAND, RF

During the last decade, online systems have grown from limited
 services only available to certain categories of people (the
 military, academic institutions) or providing limited possibilities
 (BBS, Minitel) to an international network with ever expanding
 capabilities. One driving factor behind this phenomenon is that, by
 a large extent, clients and servers have been able to interoperate
 because of the existence of open standards.

Note that the current situation is not perfect: because of badly
 documented or proprietary extensions, users can experience
 frustration when accessing many sites, often because they do not use
 the latest version of the most mainstream operating system or
 browser. Studies point out that many people do not feel like using
 Internet E-commerce because the system is unreliable and perceived
 to be so - and incompatibilities that induce errors during
 transactions only contribute to those unreliabilities. Common sense
 dictates that we should avoid as much as possible those
 incompatibilities, which for the most part do not yield interesting
 features for the user.

Comprehensive public standards contribute to reducing
 incompatibilities and unreliability. Since today, software
 development, and most notably WWW-related software development, is
 undertaken by anything from groups of volunteers around the World to
 multinational corporations, it seems that the standards should be
 designed to be equally accessible to all those developers.
 Accessibility by groups of volunteers and small businesses require
 that the standards themselves remain free (some standard bodies
 charge hefty costs, which may be acceptable to medium- or
 large-sized corporations, but certainly not to volunteers) and that
 they should be unencumbered by patents (either containing no
 patented material, either requiring that the patent holders grant a
 royalty-free license to the patented material when used for
 implementing the standard). That last requirement applies more
 particularly to the so-called "free" or "open-source" software,
 since the developers are not able to redeem the cost of the patents
 on the users.

In short:
* Using patented technologies in standards will contribute to
 incompatibilities and thus will encumber efforts to develop reliable
 WWW services.

* Using patented technologies in standards will effectively prevent
 "free software" from being available for the services described in
 the standard, depriving the users from choice and the marketplace
 from healthy competition.

David Monniaux            http://www.di.ens.fr/~monniaux
Département d'informatique de l'École Normale Supérieure,
Paris, France

[This mail expresses my personal opinion as a researcher, not an 
official position of ENS.]

Received on Sunday, 30 September 2001 12:33:17 UTC