"Greg Comeau" <comeau@panix.com> wrote in message
news:f9ls9c$adu$1@panix1.panix.com...
> In article <%xUui.1425$3x.831@newssvr25.news.prodigy.net>,
> JohnQ <johnqREMOVETHISprogrammer@yahoo.com> wrote:
>>It would seem then that just saying "compiling" is indeed ambiguous if one
>>wants to know what's really going on behind the scenes...
>>like "compiling to proprietary intermediate form" is in order.
>
> If one really wants to know what's really going on behind the scenes,
> there is probably no term, because at that point exactly what
> is going on is up to the particular implementation. So, sure,
> then we're left with the terms we have.

I was suggesting that qualifying to _what_kind_ of compiling is in order. As
in "compiling to C" (which is of course incorrect, but an example).
"compiling to intermediate form" seems reasonable evolution of the
terminology, to me. Compared to the traditional usage: "compiling to .obj".
I'm wholly OK with your previous post and Bjarne's description of cfront as
"a compiler". As in, "compile to intermediate form". Moreso, I like it much
better than "compile to .obj" which completely ignores the in-between steps.
The way I see it, once you have the (proprietary) intermediate form (symbol
tables etc.), from there you can generate whatever you want to: object code,
C code, whatever. But I'd call going from intermediate form to something
else "translation" or something similar. At this juncture, I'm thinking that
"compiling" _should_ refer to "from source to intermediate form" and it has
sub phases such as preprocessing, syntax checking, semantic analysis, etc.

John