ML vs. Lisp - Programming Languages

In article <2007031315491173815371@k2.sage.att.com>, Garry Hodgson wrote:[color=blue]
> when i last checked this thread it was about stallman and cults.
> i go away for a few days and we're talking meth heads with nuclear weapons.
> long way from lambda calculus...[/color]

Got any questions about the lambda calculus?

I'll be happy to talk about anything you want.

--
Neel R. Krishnaswami
[email]neelk@cs.cmu.edu[/email]

Garry Hodgson schrieb:[color=blue]
> when i last checked this thread it was about stallman and cults.
> i go away for a few days and we're talking meth heads with nuclear weapons.[/color]

Well, I can say in defence that Thant provoked me with, er,
unconventional ideas :-)

I agree this has gone far, far off-topic.
I'm not sure we should stop - I'm pretty sure nobody who's disinterested
in political rights has stopped reading this thread long ago. (If
anybody is genuinely annoyed I'll stop right now.)

Regards,
Jo

Joachim Durchholz wrote:

[...ignoring the gun stuff. Sorry I mentioned it. But I can't resist...]
[color=blue]
> Well, at least it's not the world at large that's paying through dollar
> inflation.
> Not anymore at any rate - the dollar once was the only relevant global
> reserve currency, but according to Wikipedia, the Euro has partially
> replaced it in this role.
> If the U.S. continue with war expenses the way they did, the dollar will
> finally lose its reserve currency status.[/color]

There are a lot of people (and I'm one of them) that have become
convinced that defending the role of the dollar as a world reserve
currency is what the invasion of Iraq and the upcoming attack on Iran is
really about. (Saddam Hussein announced in August 2000 that he was
switching his oil-for-food program to euros. The first thing the U.S.
invasion force did was switch it back to dollars. The people in the
United States heard NOTHING about any of this. (Not that I think
"dollars: bad, euros: good." It's two (or more) sets of bad guys
fighting over spoils, and at least the New York/London faction is
getting desperate.)

-thant

Joachim Durchholz <jo@durchholz.org> wrote:
[color=blue]
> If anybody is genuinely annoyed I'll stop right now.)[/color]

i'm not annoyed. i just thought it was pretty funny.
i can't get that "meth heads with nuclear weapons"
image out of my head.

----
Garry Hodgson, Senior Software Geek, AT&T CSO

do for others with no desire of return.
we should all plant some trees
we will never sit under.

Garry Hodgson <garry@sage.att.com> writes:
[color=blue]
> Joachim Durchholz <jo@durchholz.org> wrote:
>[color=green]
>> If anybody is genuinely annoyed I'll stop right now.)[/color]
>
> i'm not annoyed. i just thought it was pretty funny.
> i can't get that "meth heads with nuclear weapons"
> image out of my head.[/color]

Have you ever read "Snow Crash" by Stephenson? It's a (cyber punkish)
world where everything has been privatized, including police, the
military and the courts (Judge Bob's judicial system) and exists in
multiple "franchises" (there is more than one prison system, more than
one judicial system etc).

And then there is this Rocker type guy "Raven" whom none of the
enforcers dare to touch when he comes into the city. Because: He
drives a large bike with a sidecar and actually in the sidecar there
is a huge hydrogen bomb which is somehow coupled with his heart beat
and will blow up when Raven dies. Therefore he is treated as a
"nuclear power" by the enforcers and untouchable to them.

(The books is IMHO a in many parts and really enjoyable to read, but I
think you must either like SF and/or surrealism)

Regards -- Markus

* Markus E. Leypold:
[color=blue]
> Florian Weimer <fw@deneb.enyo.de> writes:
>[color=green]
>> * Markus E. Leypold:
>>[color=darkred]
>>>> And GNAT provides implementation that has a large subset which
>>>> interoperates extremely well with C code. Unfortunately, using this
>>> ^^
>>> You probably wanted to write "c++" here :-).[/color]
>>
>> The C++ binding generator for GNAT has never been published, and work
>> on better C++ integration has begun only recently. Writing the
>> required annotations (for vtable offsets and the like) by hand would
>> be horrible.[/color]
>
> I'm not sure -- did you mean C++ or not?[/color]

I meant C. C++ interoperability is non-existent in practice. You
need to go the C route.
[color=blue][color=green][color=darkred]
>>> Maybe. But consider the track record AdaCore has concerning public
>>> releases (from day 1: hard to build,[/color]
>>
>> I didn't face many troubles when I went by the book.[/color]
>
> Not quite: I recently rebuilt Gnat 3.15p from sources[/color]

3.15p is ancient. I don't think you can even build the C part with
current GCC. 8-/
[color=blue][color=green]
>> But if they don't, you may have to wait ages until a new architecture
>> is supported (or work on it yourself).[/color]
>
> This is partly true, partly I don't see it relevant anymore. GNAT (do
> continue the comparison) these days only supports Windows, Linux and
> *BSD on 386 (and *BSD much worse than the other two).[/color]

Just AdaCore's public version. I'm sure they still support many other
platforms for their customers (OpenVMS, for instance).
[color=blue][color=green]
>> Debuggers are overrated (at least for some languages). I write quite[/color]
>
> ACK. Still, that is the very same reason why I want the compiler to
> "just work" when I'm in need of debugging: I is annoying to have to
> get to grips with a badly integrated debugger when there is pressure.[/color]

It's just that my experience with debuggers has been utterly negative.
The first time in recent years that I met something that just works,
despite my bizarre coding style, was Hotspot (via the Eclipse front
end). Partly, this is made possible by the simplicity of the JVM.
It's not that the end result is very useful, it's just that the basic
features work as expected.

Debuggers on other platforms can't reliably handle things like
exceptions, defunctorization, multi-threading or run-time code
loading. 8-(