Re: Trunk still not open

[Glenn Morris <rgm@gnu.org>]

Juanma Barranquero wrote:

> Hmm. I'm not so sure, because I don't agree with the "doing nothing at
> all". I mean, surely there are some metrics in which it is "much
> better", but these metrics are not necessarily related to efficiency,
> but peers' goodwill (or project satisfaction, if you will), etc.

Documenting someone else's code is a good way to learn it, but it wears
you down when you spend months and months doing nothing but that.
So yes, it eventually burns through my goodwill, which is basically what
caused me to start this discussion.

>  If developer A requires X hours to document something, and developers
> B, C or D could do it in a tenth of that time, it is really efficient
> for A to do it, assuming that in general A will spend these X hours in
> another Emacs-related task anyway?

I don't think A and B are who you think they are in this scenario. I
think you underestimate the amount of time it would take for me (say) to
document frameset (say). I first have to get familiar with the code, and
only then can I start to try and document it. The first bit takes me a
long time. So to repeat the point I made to start with:

   Some people said they want more frequent releases.
   Well, you can't have that unless more people start doing some of the
   less-interesting work that goes along with releases.

> I suppose a counterargument could be given against developers who only
> want to do "shiny new" things and leave the grunt work to others, but
> I do believe that's not the case in the emacs devel community: most of
> us do our share of grunt work (squashing bugs, fixing typos, testing,
> revising and commenting proposed patches, commiting code from
> non-committers, building snapshots, etc.).

I think I disagree with your assessment here.

>> To get specific: please try and document frameset.el in the lispref.
>
> I didn't even know that it had to be documented in the lisp reference.

Well it does. This just seems obvious to me, if you want people to
actually use it.

> There are pretty fundamental libraries, like uniquify, which are not
> (uniquify is briefly described in the Emacs manual). 

I'm sure there are older features that are not well documented.
All I can say is, patches welcome.

> I will be very surprised if the non-doc status of framesets has kept
> the freeze from unfreezing ;-)

I'm surprised you're surprised. It's been sitting in NEWS without
---/+++ for months, during the time in which Stefan asked for people to
document remaining NEWS items.