Subject: Re: {fsb} Re: Ximian's proprietary connector for open source Evolution
From: "M. Drew Streib" <dtype@dtype.org>
Date: Wed, 5 Dec 2001 23:40:06 +0000
Wed, 5 Dec 2001 23:40:06 +0000
On Wed, Dec 05, 2001 at 06:15:44PM -0500, Mark Eichin wrote:
> We didn't get that far, in practice.  Part of the justification was
> also that we weren't going to replace tivoli any time soon, so we'd
> have to coexist with it - I suspect there's similar logic for Ximian
> wrt Exchange as well...

History will tell, as this is just one step in the evolution of business
models and free software, and we've seen a few new entries to this model
recently.

Notables in this field:

* Ximian
  Proprietary modules for interaction with Exchange servers around the 
  free Evolution package.
* VA Software / SourceForge
  Proprietary modules for links to existing Oracle databases, and also
  Clearcase repositories and other development tools around the mostly
  free SourceForge platform.
* Collab.net
  Similar to SourceForge, selling proprietary modules for links to
  existing development tools around the mostly free Tigris platform.
* Sendmail, Inc.
  Sells packaged software and tools around a modification of the
  free sendmail package.

This model isn't completely different than the services and custom
software around free software model, currently in use by:

* NuSphere (MySQL)
* Mysql AB (MySQL)
* GreatBridge (out of business, PostgreSQL)
* Eazel (out of business, Nautilus, Gnome)
* Troll Tech (QT)
* Scyld Computing (Beowulf)
* Namesys (reiserfs)

...and of course distribution companies have flirted with attaching 
proprietary modules to otherwise free distributions as well.

On a not-so-good-for-free-software note, some companies tried the free
model and ended up changing:

* Lutris 
  Enydra (a j2ee server) appears to be non-free in its latest incarnation.
  Formerly Lutris had tried selling services around the free platform.
* Sistina
  GFS (Global File System) will be non-free in its new versions.

We'll have to wait to see how well these hybrid models fare in the long
term. If one takes a view of the group as a whole, it appears that the
model may work for those that are leaning towards a professional service
focus, but without the extreme growth that we saw in early attempts at
free software models. Of course, there are a lot of other factors
for each company as well that contribute to success or failure...

-drew

-- 
M. Drew Streib <dtype@dtype.org>, Free Standards Group (freestandards.org)
co-founder, SourceForge.net | core team, freedb | sysadmin, Linux Intl.
creator, keyanalyze report | maintnr, *.us.pgp.net | other, see freedom/law


["application/pgp-signature" not shown]