Eric Abrahamsen writes: > Rasmus <ras...@gmx.us> writes: > >> Eric Abrahamsen <e...@ericabrahamsen.net> writes: >> >>> Would this be eligible? >> >> Not that my .02€ are worth much, but I think the idea of inline notes is >> good, but I don't think it should be done using links. See e.g. the >> discussion on citation which introduced a [cite:⋯] command. A [comment:⋯] >> command would also IMO make much more sense than [[comment:X][Y]] as was >> allowed last time I read your patch (in the weekend, I think). > > Wow, I just went back and looked at the cite thread. That was > bewildering. I don't see a direct connection here, though -- cite was > needed for very specific academic purposes, with very clearly-defined > needs. Comment is much floppier: good for anything from notes-to-self, > to notes-to-editor, to notes-to-no-one. > > *None* of the complexity is in the format itself: if you unloaded > org-comment, the comment links would be perfectly human-readable. All of > the complexity is in helper functions for manipulating them. I suppose > it would be possible to define some non-link syntax for them, but why do > that when the link syntax works perfectly well?
The only reason I can see (coming from someone who uses links liberally for other purposes ;) is just to avoid the hacks required to get extra functionality, e.g. as you alluded to applying different faces, storing additional information (author, timestamp, etc...), avoiding a need to add a link type-checking for collecting comments (although, this is not a very difficult step). On the link side, they work perfectly well for the simplest kind of comment, and because of that, there is a working prototype already. But, I think extending it beyond this will require the hackery described above. I don't have a sense if it is more work than defining a new syntax, or the long term maintenance costs of that approach. For me, it is work I already know how to do. I admit though, that does not mean it is better than a new syntax ;) Maybe a study of the cite syntax code would clarify the differences. Can anyone point me to a code repository where we could read that code? > >> On inclusion in contrib I think you can put anything org-ish there. It's >> better if the copyright is cleared in case we want to make it part of >> core, but it's not necessary. There's little difference between core and >> contrib as neither are included in Emacs and thus are hard to rely on. >> >> Since you use cl-lib (last I checked) it could not be part of Org before >> 8.4. > > Ah, that's a good point. cl-lib isn't necessary, just convenient, and > could be removed. > > Thanks, > Eric -- Professor John Kitchin Doherty Hall A207F Department of Chemical Engineering Carnegie Mellon University Pittsburgh, PA 15213 412-268-7803 @johnkitchin http://kitchingroup.cheme.cmu.edu
