Re: Volta enhancements tranche 1 (issue 6398055)

[David Kastrup]

Graham Percival <address@hidden> writes:

> On Wed, Jul 18, 2012 at 07:45:58PM +0100, Phil Holmes wrote:
>> ----- Original Message ----- From: "Ian Hulin" <address@hidden>
>> To: "Phil Holmes" <address@hidden>
>> Cc: <address@hidden>
>> Sent: Wednesday, July 18, 2012 6:48 PM
>> Subject: Re: Volta enhancements tranche 1 (issue 6398055)
>> 
>> >Implementation
>> >
>> ><tbs> by a grown-up who understands how the docs are compiled.
>
> That's the sticking point.
>
>> My understanding was that there was a desire to index /foo and foo
>> as the same, and have a single index.  You want to fix that, go
>> ahead.
>
> Yes.  Note that this would require adding functionality to
> texinfo, and since I don't want to require CVS versions of
> software as part of our doc build, that will require a texinfo
> release, and there hasn't been any texinfo release since 2008 so
> that would require a lot of work on that side of things.  Oh, and
> even if there _was_ a new texinfo release, we'd need to have a few
> rounds of bugfixing in texinfo and/or rewriting our docs so that
> they work on the new version of texinfo because probably a lot
> will have changed.
>
> Short answer: we're stuck with the current indices.  No point
> having a policy discussion for a policy that can't possibly be
> implemented in less than 12 months.
> (even if texinfo had a release today, it would take time to sort
> out the bugs and get that stable version into linux distributions)
>
>
> If and when texinfo _is_ updated, and our docs _do_ compile in
> that version, we can have a policy discussion about what we'd like
> to see at that time.

We can still decide on what we want to see in our source code.  Matching
that to the realities of an existing Texinfo implementation is then a
matter of changing macros.

-- 
David Kastrup