Re: PSPS08-7832: Devilstick
Robert Niederreiter <
rnix@...>
2008-04-01 07:04:25 GMT
Am Montag, den 31.03.2008, 00:06 +0200 schrieb Malthe Borch:
> Martin Aspeli wrote:
> > I really don't want option 3 or 4. :) Note that Dexterity is only a
> > concept at this stage. Devilstick is code. That counts for a lot.
>
> Not if it's untested and undebated.
i'm a little bit surprised in your negative attitude.
even if it is untested (automated), there was a foregone debate about
the needs which took longer than a year before the first line of code
was written. and even if this debate was not public (through the mailing
list) the design result is reasonable, even if you're not going conform
with it.
beyond that, the current code came along by a bunch of refactorings of
proto types, there are some dirty interfaces left, but they are known,
and easily alterable by a better solution, keyword CA.
> If I may be so blunt, and partly recite some critical voices heard in
> Sorrento, the devilstick project has an air of developers gone off to a
> desert island.
please explain to me a 'developement desert island'.
> Pythonphobia: I think it's very unfortunate to shy away from development
> in the native code domain to the extend that "programming" in a
> declarative XML-dialect is the primary development strategy. Python is a
> great language, and to move such a large part of development away from
> Python is simply wrong;
(Continue reading)