Re: [DotGNU]Re: My Favorite soapbox : XML linkage

[Seth Johnson]

Peter Saint-Andre wrote:
> 
> This was (and is) true of HTTP/HTML too, no? The web is open
> but not free and there is no viral aspect to sending a
> receiving all sorts of data using Internet protocols (HTTP,
> SMTP, etc). I don't see this as a function of using XML.

I'm not advocating XML, just being more precise about the
problem for the GPL.  That's really an outcome of the
attempt to move to application server setups.  Just remember
all XML is is an agreed-upon convention for talking about
data structure.  All the hoopla is really just "Lookee what
we can do now that we agree on how to communicate data
structure."  There's nothing magic about XML.

> And James Michael DuPont wrote:
> 
> > > Please dont be offended, but
> > > I have to say that it is a little strange that there
> > > are so many commercial jabber implementation and very
> > > few good GPLed.
> 
> Don't worry, I'm not easily offended. However, the facts
> lead me to a different conclusion. The main server
> implementation of the Jabber protocol is the open and
> free jabberd code (dual-licensed under JOSL and GPL).
> The main libraries for Jabber are all GPL or LGPL, with
> one or two covered under JOSL. The two or three main
> clients (WinJab, Gabber, and Exodus) are all covered
> under the GPL. By "main" here I mean most popular and
> most widely deployed. Thus I must question your
> assertion and point out that the percentage of
> commercial code in the Jabber ecosystem is quite low
> compared to open and free code. However, the protocol is
> in the public domain in a non-viral fashion, so people
> definitely can write proprietary code that implements
> the Jabber protocol. Whether such proprietary code will
> win the greatest mindshare and marketshare in the end is
> quite open to doubt. Or so it seems to me.

The fact that this has occurred in the Jabber domain is a
function of the disjunction between the way the GPL works
and the general model of what web services are.  Teh web
services area focuses largely on the protocols, not the code
implementing the protocols.  Plus, the protocols are for the
purpose of interoperability, not execution -- so people can
use them in proprietary ways in a way very unlike the
circumstances for code.

The main reason for using the GPL is because from the
standpoint of software for local machine use, it assures
that proprietary motives regarding *software* don't overtake
the freedom of free software.  The same forces exist in the
protocols area, as shown clearly by the way Jabber has been
used -- regardless of the degree, there is still no
*assurance* of freedom in the use of protocols, that would
correlate with the assurance of freedom that the GPL affords
to software.

While it is within some measure of reason to question
whether proprietary code implementing the Jabber protocol
will "win the greatest mindshare and marketshare in the
end," the same forces are still in place against which the
GPL has necessarily and very effectively contended with so
far.  Only in this case, there is no GPL analogue to protect
freedom (for either the Jabber protocol, or for remotely
executing web services [application servers] in general, for
that matter).

Seth Johnson