On Sunday 18 July 2004 03:42, Andrew Graupe wrote:
> There were binary drivers available from unichrome.sf.net, so I guess
> someone has done it. I bet if you offered the source to select people,
> in return for the binary, they would compile it for you on their
> distro/system/etc... It wouldn't even have to be open-source (get the
> people to sign an NDA).
let's apply some observation-based logic to this concept:
A. what you describe is the methodology we generally rely on today, inlcuding
NDA signing (a not-uncommon thing in writing kernel or X drivers)
B. what we have today doesn't work, otherwise your original lament wouldn't
have been made
C. ergo, you are basically suggesting that the status quo will solve the
problems inherent in the status quos, which is not only a circular argument
but has already demonstrated itself to be false. QED.
another way to look at the problem is that reworking N software packages for X
OS "flavours" is at _least_ X times more work than getting N software
packages to work on 1 OS "flavour" (even if that flavour is simply an OS
strategy standard, and not a single OS). in practice, it's much more than a
factor or X due to general inneficiences that occur.
given that we are dealing with limited resources (# of people * time in a
day), that sort of optimization should be desireable.
p.s. would it be possible to remove the parts of emails one is replying to
that are irrelevant to their reply? it's really annoying to scroll through
several pages or irrelevant, quoted material to find the additions, though
i've noticed this has become the norm rather than the exception on this list
in recent das. maybe we could have a mini-presentation on effective email
habits some month ;-)
--
Aaron J. Seigo
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA EE75 D6B7 2EB1 A7F1 DB43
_______________________________________________
clug-talk mailing list
[EMAIL PROTECTED]
http://clug.ca/mailman/listinfo/clug-talk_clug.ca
