Re: Mathematica and Windows
Dear all,
[...] Yet the GNU world is filled with half done projects, with small
groups of people who will never come close to producing anything that
rivals the for $ stuff because they do not share the time it would take
to make something great with the entire Linux community.We have so many
brilliant people working on these programs and yet the outcome is almost
never the same quality as the propriatary stuff.
[...]
The code is open, so change it to fit your desires.
I am sorry to say that but this is the usual answer to this kind of
complain that Yigal has raised. And it is as naiv as well (as is my
following reply I guess).
Some people just want to use software either because they need a job to
be done or because they do not have the knowledge to code themselves.
IMHO there is little doubt that a larger piece of software can hardly be
brought to a usable state (from a "normal" user's perspective) if you do
not have a substantial knowledge on how to handle that kind of project
on many levels. In fact many projects die or do not advance any further
because of the complexity they reach after a while because there was no
clear roadmap and carefull planning being made at the beginning. Free
software is free and this implies that you can fork/restart if you do
not agree with a decision made during these intial steps or along the
way. This certainly has its positive implications but at the down side
it easily results in what Yigal has pointed out as "with half done
projects, with small groups of people who will never come close to
producing anything that rivals the for $ stuff".
Don't get me wrong, there are many projects that have execellent
products that can easily rival proprietary software but many others do
not (yet). Another basic problem is that many projects do not take the
end user into much consideration. They are feature orientated but not
task-orientated (which is generally the users perspective to the
problem). (Even though there are Interface and GUI-groups that evaluate
GUIs and assist developers by giving user input. They are still too
rarely used (My feeling is that the reason for this results from the
fact that if you create a program then this is your "baby" and it is
hard to accept that others may find it doesn't look/work nice. It is
very human to dislike criticism (may it be warranted or not))
Most of the problems Yigal (and I) metioned particularly cumulate in
areas that do not attract large numbers of actual developers (as
compared to e.g. KDE/Gnome).
Some conclusions IMHO:
- The "only user"'s perspective should be taken more seriously (which is
BTW the official Debian philosophy) - in particular task vs feature
orientated (GUI-)design
- Particularly "Small" Projects should try to collaborate more closely
to prevent unnecessary "double work" (this is the current trend in many
of the "bigger" projects now as well)
- "Free as in free speach not as in free beer": A successfull commercial
open source program may facilitate the development of more
user-orientated software particularly in niche areas
Just my thoughs
Regards
Stephan
PS.: I hope this will not end up in a flame war, as it was by no means
my intention to start one - but I fear that my posting may have the
potential to do so.
--
Stephan Gromer, MD. PhD.
Work: Biochemie-Zentrum Heidelberg / Im Neuenheimer Feld 504 / D-69120
Heidelberg / Tel.: +49 (6221) 544291 / Fax.: +49 (6221) 545586
Home: Sternallee 89 / D-68723 Schwetzingen / Tel.: +49 (6202) 855038
Mobil: +49 (172) 7694555 / URL: http://www.gromer-online.de
Reply to: