Re: Why is AppleScript compiled?
Re: Why is AppleScript compiled?
- Subject: Re: Why is AppleScript compiled?
- From: John W Baxter <email@hidden>
- Date: Sun, 10 Aug 2003 12:33:19 -0700
- Envelope-to: email@hidden
At 21:33 -0700 8/9/2003, Christopher Nebel wrote:
>
On Friday, August 8, 2003, at 12:35 PM, Steve Roy wrote:
>
>
> Does anyone know why AppleScript is a compiled language? When
>
> introducing people to AppleScript, it has happened that someone was
>
> surprised that AppleScript is compiled instead of being simply
>
> interpreted like shell script or Perl. And when asked the question, I
>
> can't figure out why Apple made that choice. Does anyone know?
>
>
Everything Jon said is correct. Also, realize that Perl is just as
>
compiled as Applescript is. In fact, depending on your definition of
>
"compile", there's practically no such thing as a truly uncompiled
>
language. Most interpreters turn the original source into some sort of
>
internal form in order to execute it -- AppleScript is no different.
>
It does not, however, "compile" scripts in the sense of turning them
>
into actual machine instructions.
And, sometimes things turned into "actual machine instructions" are
nonetheless interpreted.
Consider compiled/assembled code for the Motorola MC-68000, running on a
68000 chip. This code is interpreted by microcode in the chip which runs
the "actual" machine instructions corresponding with the object code
instructions...often a one-to-many explosion of executed instructions.
This is by no means unique to the 68000.
While I was at NCR for my third time around in the late 1970s, a team of
engineers from NCR "collided" at Motorola with a team of engineers from
IBM. Our team was looking at whether different microcode could be written
which would make the chip replace our then-current processors; IBM's team
was looking at the same for the 370 instruction set. Each team concluded
that there wasn't quite enough silicon to do the job; each concluded that
with some more "room" the task could have been accomplished.
And both encountered the "annoyance" that Motorola's source code control
[sic] for the microcode made Woz's work on the Apple I and Apple II look
positively well organized (my "Woz notes" are somewhere in the
basement...they do have pretty much everything one needs). Motorola had
the additional little problem that the creator had left.
I limit my remarks to the 68000 deliberately, since I have no knowledge of
the 68020 and later chips.
--John
--
John Baxter email@hidden Port Ludlow, WA, USA
_______________________________________________
applescript-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/applescript-users
Do not post admin requests to the list. They will be ignored.