From: Eirikur Hallgrimsson (eh@mad.scientist.com)
Date: Sun Aug 27 2000 - 22:36:00 PDT
Adam Beberg wrote:
> I hear that over and over from open source advocates. Only one problem.
> It doesn't work that way. If I want to copy something in a piece of
> software, I really don't need the source.
Hubris? Adam, I don't want to write a replacement for that cool Mac
project manager (InfoDepot)
whose startup evaporated. I don't want to write a replacement for the
godawful copyprotected music editing software (whose company has
abandoned it). Both of those left me with files that can't be read on
contemporary versions of the appropriate operating system. This is what
happens to you if you play with bleeding-edge software in the
chained-software world.
I won't make that mistake again. I won't save any work into proprietary
formats, and I won't rely upon the operation of software for which I
only have binaries. It's just crazy.
And the continuing evolution of runtime environments (at ever-increasing
rates) means that eventually, AKA real soon, the only sane distribution
of software might just be source.
With all respect, I really don't think that you could clean-room rewrite
major applications, even granting you macho programmer status that I
don't claim. Yeah, you see a feature, I would expect that you could
copy it. A heuristic in a music or math processing package? I wonder.
For non-trivial software, I think you do need the source. And it needs
to be good source,
because uncommented,undocumented, unmaintainable source is not worth
having.
I'm not really such a hard-liner, but I really have been burned by music
software (proprietary project formats) and clever new applications from
startups that fold. I'd rather use and fix the free software than spend
money on such albatrosses.
Eirikur
This archive was generated by hypermail 2b29 : Sun Aug 27 2000 - 22:40:26 PDT