I see. Duplication always raises the question
if one wants to keep up with changes and have
some people dedicated to mirroring changes to
be reflected from upstream original library.
It’s a maintenance issue.

If one doesn’t then why not develop independently
to start with, without constraints. If you want
it to always work exactly like something existing
you either call another process or need to do the
maintenance, and be sure there will always be
someone continuing on this chore.

Not that I don’t want to have it (would be great!),
but I wouldn’t want to keep people from doing more
interesting stuff by having them keep it up to date.

just my 2¢


Am 22.02.21 um 19:16 schrieb Emir U:
> Hey Hauke, I think writing to files and running processes is not an uncommon 
> way to interface with GNU Plot/GPLK because they have their own DSLs and 
> therefore support text intermediaries. Obs. not GNU Scientific. Point is 
> moreso that I'd want to be able to work with J data types, seamlessly 
> call/chain GSL functions, get J data types back, and make a plot entirely 
> unaware of differences between these core libraries and J itself. Emir
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
> 

-- 
----------------------
mail written using NEO
neo-layout.org

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to