[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Dylan FFI
- To: firstname.lastname@example.org, "pierce" <pierce@at-ms-mail-server>
- Subject: Re: Dylan FFI
- From: Robert A. Cassels <email@example.com>
- Date: Sat, 10 Oct 1992 12:56:20 -0400
- Full-name: Bob Cassels
- Sender: firstname.lastname@example.org
>David Moon writes:
>>>Dylan programs need good access to the facilities that are out there in the
>>>system, and that it must be possible, efficient, convenient, and natural to
>>>write applications in a mixture of Dylan and other languages.
>I realize this is an implementation issue, but can someone comment on how
>Apple's Dylan implementation handles this issue of hybrid programming.
These would be product issues, which is why no one can comment on them
beyond Dave's motherhood-and-apple-pie statements. I would expect the
answers to your questions to change over time, as implementations mature
and as computing platforms change. Certainly all the alternatives you
mention are possible, and none seem mutually exclusive (ignoring resource
constraints). It wasn't immediately obvious from your questions whether
you had an opinion about what the right options would be for your intended
>Will a future Apple Dylan implementation be directly linkable in the MPW
>environment, or will Dylan require that foreign code be loaded into it like
>current Lisp implementations do?
>Does having good access to facilities that are out there in the system mean the
>ability to call traps, or does it mean that Dylan programs should be able to
>take advantage of foreign coded libraries without building code resources out
>More Specifically, Will some future desktop Dylan implementation support access
>to MacApp or Bedrock?