Subject: Re: FW: Why would I pay for Ximian software?
From: "Perry E. Metzger" <>
Date: 03 Jan 2002 16:01:49 -0500

Harald Koch <> writes:
> User interface design and implemenation is *hard*, and unprestigious.
> One of the common incentives for OSS, community prestige, doesn't seem to
> come into play for user interface issues. Hackers don't congragulate
> each other on their cool user interfaces, IME.

So the "community appreciation" part of the compensation for working
on such things isn't as big for desktop apps as for, say, way cool
operating systems. Hackers don't use them, they don't appreciate them
much, and they don't give each other ego boosts for doing them.

(Actually I disagree with this in many ways -- look at what Evolution
did for Raster's reputation points -- but the overall point that good
GUIs are hard and good desktop apps are hard does apply.)

The question then is the same as is always here. Can one come up with
a way that you can get enough money from the 50M users of your
spreadsheet to pay people at least some staff to work on it full time?
If the answer is yes, we can get the benefits of open source even in
areas like the desktop by converting part of the effort from amateur
to professional in spite of the fact that desktop apps are
"different". (Things like GCC have large pro staffs behind them too
but we often neglect that.) If there is no reasonable model for doing
that, things are not good.

Ximian seems to be trying to make money off of providing the
equivalent of Microsoft's "Windows Update" for a fee. I'm not sure how
well that will work but time will tell.

I'd argue that part of the reason Apache and GCC win and AbiWord does
not is that Apache and GCC have people in big businesses who care
enough to throw money at them because they need them, but that so far
people haven't found their way to do that with desktop apps.

Perry E. Metzger
NetBSD Development, Support & CDs.