28 Nov
2003
28 Nov
'03
12:28 p.m.
On Friday 28 November 2003 08:07 am, Stonekeeper wrote:
It was much simpler than that: The Open source
philosophy (or lack of)
wouldn't be the primary concern for musicians "jumping ship". Most
(IMHO) would simply move out of cost (I hate to think how much all those
LADSPA plugins would cost if Steinberg had sold them).
Guilt could be another motive - I wonder how many of them are running
cracked copies of audio applications with cracked plugins on an illegal copy
of Windows? ;)
Hmm.. this makes me think: What if Ardour was so good,
it put steinberg
and cakewalk out of business? Would that be a good thing? For musicians,
perhaps? For music program innovation, maybe not. As much as i hate
microsoft, they do know the benefit of spending money on research. And
they reap the benefits of that. But so do we (especially in the areas of
HCI and usability). If that were translated into the music software
industry then perhaps a lack of research would be detrimental to future
features. At the end of the day research must be paid for one way or
another, because that's the economy we live in. Now I don't know the
answers to these questions. It's merely a thought.
I think the best research happens "in the field". It took Digi Design
years to figure out that more than one level of "Undo" would be useful - then
they gave us a whopping 16 levels - and we still couldn't undo simple things
like moving a fader. I think the best innovations come from observation of a
real working environment - finding the bumps in the workflow and addressing
them. I personally think we'd have a much more innovative software world
today if computer programs were still written only by the people who need
them. This is, again, purely my own fallible opinion.
<offtopic>
Which leads me onto another rant: with todays hard disk sizes, memory
capacities and internet speeds, why do we need to distribute things with
dependencies? Why not statically compile? My recent example of this was
with skaletracker (this is not a flame!!). But i unzipped the file, ran
it and it just worked (Im not sure if it was statically linked with SDL
or not). In it's current form it IS inferior to cheesetracker but people
are going to use what works for them. I think people should distribute
"developer" versions and "user" versions. developer is what we are
used
to, user is the statically linked (read no brainer, no options) version.
The amount of times I would have preferred to download a "non-bleeding
edge" version that was statically compiled and took 10 times as long to
download is countless. Anything but dependency hell!!! :)
I hope this isn't seen as a flame towards anyone, it's intended as
constructive criticism in a hope to better the linux experience as this
problem is rife community-wide.
</offtopic>
I agree. It is especially frustrating when I simply want to try out an
application to see whether it is useful to me, and I have to spend half an
hour tracking down dependencies (which the author of the program decided not
to provide links for on the web site) in order to find out that it doesn't do
what I need. Then again, I suppose it's good for me ;)
|)
|)enji