Re: [O] Elisp programming style

[Eric Abrahamsen]

[...]


>> Would that be considered bad style from the original author, or is that
>> up to personal choice and not considered a problem?
>> 
>
> Not a problem - you can't predict the future. You do the best you can
> with your current knowledge. You can always refactor in the future:
> that's one reason that having the source code around is important.  When
> you refactor, you may be cursing the original author for the "bad"
> decicions, but c'est la vie: at least, you have chosen a system where
> *you* can do that and not depend on a third party: they might not even
> be alive, let alone willing to do what you want.

I think this is probably the key. As a non-professional programmer who
has gone on safari through the org-mode code (though this obviously
applies to a lot of elisp packages), I think incremental growth is
responsible for a lot of oddities. Refactoring is possible in theory,
but for fast-growing packages it never becomes a priority, and my guess
is by this point there are parts of org where even Carsten would tread
very lightly. It's not a disaster, but it does lead to regression bugs,
which I think we're seeing.

Entropy is contravened, however: witness the push towards an org test
suite, and the production of a generic exporter in org…

Slightly OT, sorry,

Eric

-- 
GNU Emacs 24.0.90.1 (i686-pc-linux-gnu, GTK+ Version 2.24.4)
 of 2011-10-06 on pellet
Org-mode version 7.7 (release_7.7.466.ga5cb)