Re: syntax-local-binding

[Ludovic Courtès]

Hello Happy Guilers!  :-)

Sorry for remaining silent in this heated thread (I spent my spare time
on other practical issues for 2.0.4, and felt I lacked the competence.)

My overall feeling is that providing a 1.8-compatible ‘local-eval’ in
2.0 is great, but that it’s essentially one bug to be fixed among many
others.

Thus, I appreciate all the work Andy and you have put into it, and I’m
glad it will be helpful to Guile users such as Lilypond.  However, I
think, that we should keep in mind that Guile is not just about
‘local-eval’, and that there are other great things to work on together.

As for the technical aspects:

Mark H Weaver <address@hidden> skribis:

> Andy Wingo <address@hidden> writes:
>> On Tue 24 Jan 2012 14:25, Mark H Weaver <address@hidden> writes:
>>
>>> Andy Wingo <address@hidden> writes:
>>>
>>>> None of the interfaces that I proposed leak internal psyntax
>>>> representations.
>>>
>>> `syntax-local-binding' leaks the internal representations used for
>>> bindings.
>>
>> You mean, whether something is a lexical, or a macro, or a global, or
>> whatever; OK.
>
> That's actually not what I meant, although I'm not convinced that we
> fully understand the implications of exposing even that much.  One thing
> that is already clear is that `identifier-syntax' and `local-eval' were
> previously capable of emulating variables perfectly before, whereas in
> the presence of `syntax-local-binding' they no longer are.
>
> This is a perfect example of how added flexibility in one aspect can
> lead to _reduced_ flexibility in other aspects.  I think we need more
> time to consider the implications of this.

I agree.  Yet, it’s the kind of API that is useful internally for
different purposes beyond ‘local-eval’, as this discussion showed,
right?

So, IIUC, the tension is:

  1. Such as API is desirable for Guile-internal consumption.

  2. Exposing it to users may restrict our ability to change the
     implementation in the future, just like ‘the-environment’ did.

I haven’t fully thought about it, but one possible trade-off would be to
mark ‘syntax-local-binding’ & co. as internal, somehow arrange to make
them visible only from some (system ...) module, and document them in
the “Implementation” chapter.

WDYT?

Thanks,
Ludo’.