Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

A challenge... about Parrot

1 view
Skip to first unread message

al...@alfarrabio.di.uminho.pt

unread,
Mar 26, 2007, 2:08:09 PM3/26/07
to p6i, José Alves de Castro
Hi

There will be a conference about languages, compilers, interpreters and
so on, in Portugal. It is an "international" conference, or at least
they claim so. It's page is at http://corta.di.ubi.pt/ and the call for
papers is at http://corta.di.ubi.pt/cfp.html.

This is not a *known* conference, but I think that if someone has the
time to do it, it would be a good opportunity for a first "scientific"
publication about parrot.

What I really mean: Parrot is great (and I know it, it has some C and
Perl lines written by me), there is a lot of involvement from the
community, but it is lacking some involvement with universities and the
scientific community. I think both communities would gain if Parrot
start being used in Universities (to teach, to propose projects, etc,
etc). Leo had that same idea... The best way to start this wedding, is
to write anything about parrot internals. At least, that is my idea.

I can't offer to help writing it (unfortunately). I can help (if the
paper is accepted) with accommodation in Portugal for the speaker (I
think we can manage APPP (perl.pt) to help).

Ok, challenge done. Now it is your turn.

Alberto

--
Alberto Simões - Departamento de Informática - Universidade do Minho
Campus de Gualtar - 4710-057 Braga - Portugal

Jonathan Worthington

unread,
Mar 26, 2007, 7:29:35 PM3/26/07
to al...@alfarrabio.di.uminho.pt, p6i, José Alves de Castro
Alberto Simões wrote:
> There will be a conference about languages, compilers, interpreters
> and so on, in Portugal. It is an "international" conference, or at
> least they claim so. It's page is at http://corta.di.ubi.pt/ and the
> call for papers is at http://corta.di.ubi.pt/cfp.html.
Sounds cool; sadly I can't make it myself.

> This is not a *known* conference, but I think that if someone has the
> time to do it, it would be a good opportunity for a first "scientific"
> publication about parrot.

For sure!

> What I really mean: Parrot is great (and I know it, it has some C and
> Perl lines written by me), there is a lot of involvement from the
> community, but it is lacking some involvement with universities and
> the scientific community. I think both communities would gain if
> Parrot start being used in Universities (to teach, to propose
> projects, etc, etc). Leo had that same idea... The best way to start
> this wedding, is to write anything about parrot internals. At least,
> that is my idea.

I have presented Parrot in an academic setting before - namely, at the
Cambridge Programming Research Group (Cambridge uni's folks who do
research into programming languages stuff). It was hard work, though the
feedback I got was that it was somewhat interesting to them.

<rant>
The most stupid thing was the guy who made the "where are the formal
semantics" comment, and seemed willing to disregard or poke fun at
Parrot and Perl 6 as a result of it not having any. At the time I didn't
know how stupid his comment was. My assumption from this statement was
that we had formal semantics for a whole range of languages in common
use. Having been at POPL this year, I found the reality is more like,
bits of languages that are liked in academia have been formalized. Oh,
and lots of calculi, which are really more like assembly languages than
HLLs, though I use assembly in a pretty loose "it's something Turing
complete that I guess you could compile to" sense. In fact, folks were
there presenting their work on a mechanized semantics for Standard ML if
I remember correctly, and I got the impression this was the first work
towards giving any large language mechanized formal semantics (as in,
that you can check on a computer or do proofs with using a theorem
prover to assist). So while people are willing to say "where's the
formal work", it doesn't seem to exist for languages that people are
using to solve real problems. I'd love to be corrected on this, but it
seems like this is the case.
</rant>

I too would really love to see the gap bridged and I certainly don't
want to discourage anyone from giving a talk at this conference - had I
been available, I may have offered myself. But I hope this is a hint to
anyone who does, as to one thing they might get asked about. Guess it
depends how much of an idea the audience has about the way things work
in The Real World. I was talking to a pretty theory-centric group, after
all. I'm certainly happy to review whatever anyone produces for this.

Thanks,

Jonathan

0 new messages