On Jun 25, 2010, at 7:07 PM, Mathieu Bouchard wrote:
On Fri, 25 Jun 2010, Bryan Jurish wrote:
I clearly and distinctly recall that you (Hans) mentioned to me off- list in 2008 that use of class_sethelpsymbol() was deprecated for
pd-extended ("it is best not to use class_sethelpsymbol() at
all" [for pd-extended]) and recommended that I remove those calls
from my codeThat's the usual mantra. Some kind of preaching without much
justification. Thou shalt not use class_sethelpsymbol, thou shalt
not use aliases, because it's ugly and wrong. it's wrong because
it's ugly, and it's ugly because it's wrong. It's wrong because if
you used class_sethelpsymbol, it would be ugly. (see also : begging
the question !). It sounds like a seminar about young sheep being
guided towards JAVA (Jesus Gosling was born in
com.sun.labs.Bethlehem at the beginning of our era...)Indeed. The same holds (these days) for 'moocow'. I still feel that this discussion is relevant, though.
I don't think it's Hans' business to impose a helpfile style through
deprecation of class_sethelpsymbol. Library writers deserve more
freedom than that... EVEN if they decide to misuse it, then Hans
should not seek to ban sethelpsymbol for example, and instead should
go against the content or the filename, such anything that Degoyon
decides to be teaching us about giraffe(s).
How about you guys all speak for yourselves and let me speak for
myself. This time around, all I said is that there is the only time
to use class_sethelpsymbol() is if the help patch is different than
the classname of the object. Its redudant in that case.
.hc
http://at.or.at/hans/