Re: Why Emacs needs a modern bug tracker

[Eli Zaretskii]

> Date: Sat, 5 Jan 2008 16:48:10 -0500
> From: "Eric S. Raymond" <address@hidden>
> Cc: address@hidden, address@hidden
> 
> Eli Zaretskii <address@hidden>:
> > You are assuming that all parts of the code are being developed, it
> > seems.  That's not what happens in Emacs, probably due to its age and
> > the fact that core features are developed/refactored only very seldom,
> > or not at all.
> 
> OK, fair point.  In that case Emacs will behave, statistically, more
> like a project the size of the code's active region.  Not entirely,
> you could still have bad interactions with the "dark matter", but...
> do you have any idea what the approximate LOC of the active region is?

Sorry, no.  Someone will have to calculate this based on CVS history.
It's not trivial, because there's lot of development going on in
various Lisp packages, but many of those are fairly isolated from the
rest of Emacs.  So the first step towards estimating this would be to
identify Lisp files that are infrastructure used by many other
packages.  C sources are easier, because most of them are core
functionality, but one still needs to separate new features from
changes to old features (since new features cannot easily break old
code).

> All right.  I'll need administrator permissions on our Savannah site.  
> Can you set those?

If you mean Emacs project administrator privileges, then yes, I can,
but I'll need Richard's approval for that.  If you mean Savannah
administrator privileges, then I cannot do that; you will need to ask
Richard to ask Savannah hackers to do it.

Either way, it's probably a good idea to describe in some detail what
you want to do there, for Richard to consider.

Thanks.