Re: Hosting for Guile-PHP

[Jon Herron <jon.herron@yahoo.com>]

Ludo -

  Sure, that sounds fine to me.  I definitely understand wanting to promote Scheme over PHP from a GNU point of view, for me personally I have to use PHP at work and any potential chance to slide some Scheme in the mix would be welcome (to me at least).  I chose Guile as my Scheme implementation mainly due to the compiler tower, example language implementations and mention of a PHP translator under the cool ideas section.  I understand this was not a firm desire, nor guarantee of acceptance on the projects behalf.  I have not regretted my decision to use Guile one bit, and plan to continue Guile-PHP as time permits.  If a merge is beneficial in the future that's great, if not no hard feelings.

  With that being said, is there another area of the project that would be beneficial for me to look at?  I've seen various mentions of the ECMAScript front-end but not sure what is needed there.  Other bugs? Desired libraries?  I wouldn't mind having Guile-PHP be a side-side project since there's a strong chance I would be the only user for a long time anyway.  In short I'd enjoy helping out if there is something that needs looking at.

Thanks,

Jon




----- Original Message ----
From: Ludovic Courtès <ludo@gnu.org>
To: Jon Herron <jon.herron@yahoo.com>
Cc: guile-devel@gnu.org
Sent: Tue, April 13, 2010 5:12:00 PM
Subject: Re: Hosting for Guile-PHP

Hi Jon,

Jon Herron <jon.herron@yahoo.com> writes:

>   I'm starting to make some more slow progress, I have basic tree-il
> generated for simple cases, the below PHP script runs for instance:

Sounds cool!

> Although no where near finished yet, I'm getting to the point where I
> would like to host the code that I'm writing somewhere.  I was
> wondering if there's a chance of getting a branch that I could commit
> to (assuming there is project level interest in this implementation)?
> Or should I get a separate project hosted elsewhere (savannah
> perhaps?).  I'm not exactly sure the protocol for a question like
> this, but figured it was worth asking.

For now I would be in favor of setting up a separate project on
Savannah, rather than having a branch in the Guile repo.  That should be
technically doable as we’ll make sure to keep the compiler API
compatible during the 2.x release series.

To be honest, I personally don’t value the PHP language much and would
rather have GNU promote Scheme than PHP, though I can understand that
some people would find it useful to have interoperability between Guile
Scheme and PHP code.  Additionally, each language front-end adds a
maintenance burden and consumes hacker energy (the ECMAScript front-end
needs love, BTW! ;-)).

Anyhow, depending on how things evolve in the future, we can still
decide to merge it (which would itself be facilitated if all
contributors have assigned copyright to the FSF.)

What do you think?

Thanks,
Ludo’.