Subject: Re: [FYI] Microsoft license spurns open source
From: Ben_Tilly@trepp.com
Date: Mon, 25 Jun 2001 10:00:24 -0400


"Stephen J. Turnbull" <turnbull@sk.tsukuba.ac.jp> wrote:
>>>>>> "Seth" == Seth David Schoen <schoen@loyalty.org> writes:
>
>    Seth> The point is that "shared source" is not even remotely
>    Seth> comparable to open source and provides virtually none of the
>    Seth> latter's benefits.
>
>Don't exaggerate.  Such exaggeration confuses the undecided, or
>convinces them that we are fanatics or have some sort of undisclosed
>ulterior motive.

How is this exaggeration?

The primary benefit of open source is that each user gets the best
legal guarantee that systems which rely on that software can continue
to be guaranteed - you have access to the source and are free to use
that.  I think we can call this benefit primary because it is the first
one recognized in the model and was the original motivation for RMS.
Plus according to ESR and others it is the argument that rings most
solidly with CEOs.

The primary benefit of open source is not addressed by Microsoft's
shared-source initiative.  In fact the need not to go back to any
particular vendor (specifically Microsoft) is exactly what they wish
to avoid.

The second benefit of open source is that debugging effort becomes
parallelized.  This was certainly noticed before CatB, but ESR happened
to say it very well.  This benefit can only be realized if there are
users contributing and ongoing maintainance that gathers important
bug-fixes together.  I think we can call this benefit secondary because
it was realized after the first but is key to further possible benefits.

We have at this point no guarantee that developers will jump on board
Microsoft's shared source initiative.  Certainly some other attempts at
making things somewhat but not really open source have gone over like
lead.  (*ahem* Sun's community source licensing anyone?)  Also from
what little I have seen of Microsoft's proposal, I see no signs that
they have given serious consideration to the maintainance need.

The third benefit of open source is (in time) improved reliability and
better features.  From the point of view of consumers this is the most
obvious reason to go with established open source software.  However I
am ranking it third because it derives from the first two.

Of course if I am dubious about the ones that this advantage derives
from, I am dubious that Microsoft's initiative will achieve it.  Plus
Microsoft's initiative will encourage development on top of current
Microsoft APIs and libraries.  Their trackrecord suggests that trying
to build solid software on that foundation is like trying to build
castles on quicksand.  (Hmmm...advertising line for Cygwin?  "Building
reliable software on Windows is like building castles on quicksand.
We help drain the marsh." :-)

So I just ticked off the biggest wins for me in open source, and I am
dubious that we will see any of them in shared source.  Why, then, is
it an exaggeration to say that Microsoft's proposal has virtually none
of the benefits of open source?

Cheers,
Ben