change groups! 
Author Message
 change groups!

wow ... i subscribed to this list thinking it was a bob dylan ng only to
find its full of computer language stuff ... arghhhh!!! whoever is
subscribed to this list for real as a computer type person do yourself a
favour ... go and join rec.music.dylan and disover his bobness!!!! it will
change your life and its oh so much more stimulating than computer
languages!


Sun, 27 Jan 2002 03:00:00 GMT  
 change groups!

Quote:
> wow ... i subscribed to this list thinking it was a bob dylan ng only to
> find its full of computer language stuff ... arghhhh!!! whoever is
> subscribed to this list for real as a computer type person do yourself a
> favour ... go and join rec.music.dylan and disover his bobness!!!! it will
> change your life and its oh so much more stimulating than computer
> languages!

I can see how the "comp.lang" part of "comp.lang.dylan" could confuse
someone into thinking that this wasn't a newsgroup about a COMPuter LANGuage
called DYLAN.  I guess we should just rename the group to
"hey.stupid.types.this.is.not.about.bob.dylan.but.about.an.eponymous.program
ming.language" instead.

Or you could grab a clue.  Your choice, really.

--

          PGP Key: http://www.igs.net/~mtr/pgp-key.html
PGP Fingerprint: 40D1 33E0 F70B 6BB5 8353 4669 B4CC DD09 04ED 4FE8



Sun, 27 Jan 2002 03:00:00 GMT  
 change groups!

Quote:



> > wow ... i subscribed to this list thinking it was a bob dylan ng
> > only to find its full of computer language stuff ... arghhhh!!!
> > whoever is subscribed to this list for real as a computer type
> > person do yourself a favour ... go and join rec.music.dylan and
> > disover his bobness!!!! it will change your life and its oh so
> > much more stimulating than computer languages!

> I can see how the "comp.lang" part of "comp.lang.dylan" could confuse
> someone into thinking that this wasn't a newsgroup about a COMPuter
> LANGuage called DYLAN.  I guess we should just rename the group to
> [newgroup.name.which.is.also.an.insult]

> Or you could grab a clue.  Your choice, really.

Hey Michael, chill.  It could be worse, we could be programming in a
language called "Megadeath" or something ;-)  Anyway, for those not
as steeped in net lore as we hacker types, the abbreviations which
compose newsgroup names might be somewhat opaque, no?

OTOH, Suze's post is clearly off-topic and so somewhat inappropriate.
Suze, at the risk of sounding patronising, you might want to take a
look at the newsgroups "news.announce.newusers" and
"news.newusers.questions" for general nettiquette.  Oh, and you're
assuming we haven't already encountered Bob; but in any sizeable
group of hackers, there's bound to be someone to whom his wacky use
of words appeals (like me :-).

ObDylan:

-----------------------------------------------------------------------
// a la Occam
// (This allows more than fixed "i from 0 to 10" iteration,
// perhaps giving rise to some bizarre semantics.  And it's not
// strictly correct Dylan, but should work in Harlequin Dylan.)

define macro par
  { par ( ?for-clauses:* ) ?:pbody end }
 => { par-aux ( ?for-clauses:* ) ?pbody end }
pbody:
  { ?main:body } => { ?main, #f }
  { ?main:body finally ?val:body } => { ?main, ?val }
end macro par;

define macro par-aux
  { par-aux ( ?for-clauses:* ) ?:body, ?val end }
 => { let par-threads = #();
      for ( ?for-clauses )
        par-threads :=
          add!(par-threads,
               make(<thread>, function: method () ?body end));
      end
      while (~empty?(par-threads))
        let (joined-thread, #rest vals)
          = apply(join-thread, par-threads);
        par-threads := remove!(par-threads, joined-threads);
      end;
      ?val }
end macro par;

-----------------------------------------------------------------------

ObBob:

  All of the people can be part right some of the time,
  And some of the people can be alright part of the time,
  But *all* of the people can't be *all* right *all* of the time.

  I think Bob Dylan said he thought Abraham Lincoln said that.

Hugh G. Greene



Sun, 27 Jan 2002 03:00:00 GMT  
 change groups!

Things could be even worse. Imagine if  the newsgroup of our language of
choice was called
comp.lang.Britney.Spears ! Of course this is all just a little too silly now
. :)
-Russ
(a fan of both dylans)

Quote:
----- Original Message -----


Sent: Wednesday, August 11, 1999 8:56 AM
Subject: Re: change groups!




> > > wow ... i subscribed to this list thinking it was a bob dylan ng
> > > only to find its full of computer language stuff ... arghhhh!!!
> > > whoever is subscribed to this list for real as a computer type
> > > person do yourself a favour ... go and join rec.music.dylan and
> > > disover his bobness!!!! it will change your life and its oh so
> > > much more stimulating than computer languages!

> > I can see how the "comp.lang" part of "comp.lang.dylan" could confuse
> > someone into thinking that this wasn't a newsgroup about a COMPuter
> > LANGuage called DYLAN.  I guess we should just rename the group to
> > [newgroup.name.which.is.also.an.insult]

> > Or you could grab a clue.  Your choice, really.

> Hey Michael, chill.  It could be worse, we could be programming in a
> language called "Megadeath" or something ;-)  Anyway, for those not
> as steeped in net lore as we hacker types, the abbreviations which
> compose newsgroup names might be somewhat opaque, no?

> OTOH, Suze's post is clearly off-topic and so somewhat inappropriate.
> Suze, at the risk of sounding patronising, you might want to take a
> look at the newsgroups "news.announce.newusers" and
> "news.newusers.questions" for general nettiquette.  Oh, and you're
> assuming we haven't already encountered Bob; but in any sizeable
> group of hackers, there's bound to be someone to whom his wacky use
> of words appeals (like me :-).

> ObDylan:

> -----------------------------------------------------------------------
> // a la Occam
> // (This allows more than fixed "i from 0 to 10" iteration,
> // perhaps giving rise to some bizarre semantics.  And it's not
> // strictly correct Dylan, but should work in Harlequin Dylan.)

> define macro par
>   { par ( ?for-clauses:* ) ?:pbody end }
>  => { par-aux ( ?for-clauses:* ) ?pbody end }
> pbody:
>   { ?main:body } => { ?main, #f }
>   { ?main:body finally ?val:body } => { ?main, ?val }
> end macro par;

> define macro par-aux
>   { par-aux ( ?for-clauses:* ) ?:body, ?val end }
>  => { let par-threads = #();
>       for ( ?for-clauses )
>         par-threads :=
>           add!(par-threads,
>                make(<thread>, function: method () ?body end));
>       end
>       while (~empty?(par-threads))
>         let (joined-thread, #rest vals)
>           = apply(join-thread, par-threads);
>         par-threads := remove!(par-threads, joined-threads);
>       end;
>       ?val }
> end macro par;

> -----------------------------------------------------------------------

> ObBob:

>   All of the people can be part right some of the time,
>   And some of the people can be alright part of the time,
>   But *all* of the people can't be *all* right *all* of the time.

>   I think Bob Dylan said he thought Abraham Lincoln said that.

> Hugh G. Greene



Sun, 27 Jan 2002 03:00:00 GMT  
 change groups!

Quote:
> Hey Michael, chill.

I don't wanna I don't wanna I don't wanna!!!  :-)

Quote:
> It could be worse, we could be programming in a
> language called "Megadeath" or something ;-)

Please don't even JOKE like that!

Quote:
> Anyway, for those not
> as steeped in net lore as we hacker types, the abbreviations which
> compose newsgroup names might be somewhat opaque, no?

Agreed.  But this "Suze" character spent time here and found out it was
about people discussing a COMPuter programing LANGuage called DYLAN (by now
the penny would have dropped for anybody but the terminally clueless).  THEN
she posted an insulting message.

--

"get a life. its a plastic box with wires in it."



Sun, 27 Jan 2002 03:00:00 GMT  
 change groups!

Quote:

> Agreed.  But this "Suze" character spent time here and found out it was
> about people discussing a COMPuter programing LANGuage called DYLAN (by now
> the penny would have dropped for anybody but the terminally clueless).  THEN
> she posted an insulting message.

Well, she'd have really have been out of order if she'd posted an
insulting message before spending time here.

And to the person worried about comp.lang.britney.speares: Speares is a
well respected (if small) language used for defining CIE colour space
co-ordinates in a device-independent way, don't knock it. The newsgroup
is crammed out with enraged mathematicians and designers fending off an
ever increasing tide of autograph requests...

- Rob.



Sun, 27 Jan 2002 03:00:00 GMT  
 change groups!

Quote:

> -----------------------------------------------------------------------
> // a la Occam
> // (This allows more than fixed "i from 0 to 10" iteration,
> // perhaps giving rise to some bizarre semantics.  And it's not
> // strictly correct Dylan, but should work in Harlequin Dylan.)

That's a nice one. What's not correct about it?

Andreas

--
"We show that all proposed quantum bit commitment schemes are insecure because
the sender, Alice, can almost always cheat successfully by using an
Einstein-Podolsky-Rosen type of attack and delaying her measurement until she
opens her commitment." ( http://xxx.lanl.gov/abs/quant-ph/9603004 )



Mon, 28 Jan 2002 03:00:00 GMT  
 change groups!

Quote:


> > ------------------------------------------------------------------
> > // a la Occam
> > // (This allows more than fixed "i from 0 to 10" iteration,
> > // perhaps giving rise to some bizarre semantics.  And it's not
> > // strictly correct Dylan, but should work in Harlequin Dylan.)

> That's a nice one. What's not correct about it?

The following call to apply:

Quote:
>         let (joined-thread, #rest vals)
>           = apply(join-thread, par-threads);

because apply in the DRM has the signature

  apply /function/ /argument/ #rest /more-arguments/
    => #rest /values/

i.e., it has *two* required arguments.  Don't ask me why.  HD lets you
away with this because its version of apply only has one, taking the
rest as #rest args.  In the implementation I gave, it's cheap to split
off the head of the list and pass that as the second arg, with the
tail as the #rest, but for other kinds of <sequence> that might involve
expensive copying.

Plus there were a couple of bugs, as Eric Gouriou pointed out in
private mail.

(1) I don't really need "vals" above, unless I plan to collect the
results.  It didn't occur to me at the time, but one sensible way to
do that would be to construct a <sequence> of <sequence>s of results
from all the spawned threads.

(2) the last name in

Quote:
>         par-threads := remove!(par-threads, joined-threads);

should be "joined-thread" (singular).

Hugh G. Greene



Mon, 28 Jan 2002 03:00:00 GMT  
 change groups!

Quote:

> because apply in the DRM has the signature

>   apply /function/ /argument/ #rest /more-arguments/
>     => #rest /values/

> i.e., it has *two* required arguments.  Don't ask me why.  HD lets you

Hm. Gwydion Dylan uses

define method apply (function :: <function>, #rest arguments)

as well, and it makes perfect sense to me. Of course calling apply
without any arguments is identical to calling the function directly,
but why should I be forced to treat that as a special case in my code
when it can be treated as a special case in apply and thus make user
code easier to read?

I think this should be flagged for change in a future revision of the
DRM.

Speaking of that, what's the deal with DRM updates? As far as I can
tell, Apple is still holding the copyright on the DRM. Wouldn't it be
more appropriate to hand over that copyright to a commitee, so the DRM
can be updated to reflect progress?

Andreas

--
"We show that all proposed quantum bit commitment schemes are insecure because
the sender, Alice, can almost always cheat successfully by using an
Einstein-Podolsky-Rosen type of attack and delaying her measurement until she
opens her commitment." ( http://xxx.lanl.gov/abs/quant-ph/9603004 )



Mon, 28 Jan 2002 03:00:00 GMT  
 change groups!


Quote:

>Speaking of that, what's the deal with DRM updates? As far as I can
>tell, Apple is still holding the copyright on the DRM. Wouldn't it be
>more appropriate to hand over that copyright to a commitee, so the DRM
>can be updated to reflect progress?

I own the print rights.  If there was an organized effort to do another
version of the language, we could do a new version of the DRM.  Right
now, I don't see that organized effort happening, just bits and pieces
of work.

As for the mentioned change to APPLY, I'm not sure I agree that the
change should be made.
--

                                            http://www.folly.org/~alms
This is our garden, although we have no grass.
-Bahrije Baftiu, Kosavar refugee in Macedonia



Mon, 28 Jan 2002 03:00:00 GMT  
 change groups!

[the DRM]

Quote:
> I own the print rights.  If there was an organized effort to do another

That's good. My copy says Copyright Apple, and I was afraid that we
had no chance in hell to ever see an updated version.

Quote:
> version of the language, we could do a new version of the DRM.  Right
> now, I don't see that organized effort happening, just bits and pieces
> of work.

It's about time to get things organized. Else we end up with three
dialects of Dylan: Harlequin, Gwydion and DRM.

Quote:
> As for the mentioned change to APPLY, I'm not sure I agree that the
> change should be made.

What's your rationale for that?

Andreas

--
"We show that all proposed quantum bit commitment schemes are insecure because
the sender, Alice, can almost always cheat successfully by using an
Einstein-Podolsky-Rosen type of attack and delaying her measurement until she
opens her commitment." ( http://xxx.lanl.gov/abs/quant-ph/9603004 )



Mon, 28 Jan 2002 03:00:00 GMT  
 change groups!


Quote:
>> version of the language, we could do a new version of the DRM.  Right
>> now, I don't see that organized effort happening, just bits and pieces
>> of work.

>It's about time to get things organized. Else we end up with three
>dialects of Dylan: Harlequin, Gwydion and DRM.

Harlequin and Gwydion are already working the Common Dylan effort to
ensure compatibility.  Common Dylan mostly covers extensions, so far,
but it could also cover clarifications or changes to the existing
language.

I don't think the DRM should be changed incrementally.  That is,
changes should be collected and documented separately .  When a the
next round of language design has been completed, that's the time
to revise the DRM, and come out with the DRM II.

Quote:
>> As for the mentioned change to APPLY, I'm not sure I agree that the
>> change should be made.

>What's your rationale for that?

My rationale is that APPLY is a basic part of the language, that its
behavior is based on the Common Lisp and Scheme APPLY, that I wouldn't
consider changing it without giving the matter a lot more thought and
soliciting a lot more opinions from Scheme, CL, and Dylan implementors,
and that there are much more important things to spend time on, if
we want to engage in language design.

But don't let me be a party pooper.  If Harlequin and Gwydion want to
organize a language change effort, you could certainly put APPLY on the
list.
--

                                            http://www.folly.org/~alms
This is our garden, although we have no grass.
-Bahrije Baftiu, Kosavar refugee in Macedonia



Mon, 28 Jan 2002 03:00:00 GMT  
 change groups!

Quote:



> I don't think the DRM should be changed incrementally.  ...

Not speaking in any official Harlequin capacity, that sounds like a
good idea, with Common-Dylan to tide us over meanwhile.

Quote:
> >> As for the mentioned change to APPLY, I'm not sure I agree that the
> >> change should be made.

> >What's your rationale for that?

> My rationale is [several good points ...]

With a bit more thought, and some prompting from another Harlequin
hacker, I realised another reason to leave it as-is.  The DRM
definition, confusingly-explained though it be, prevents the following

  apply(fn);

at the level of function signature checking.  The HD and GD
implementations can only discover something is wrong when they come to
evaluate that call to apply (which might be at compile-time or runtime).

(Why is it "wrong", by which I mean nonsensical?  Consider:
  apply(fn, 1, 2, #(3, 4))  <=>  fn(1, 2, 3, 4)
  apply(fn, 1, #(2, 3))     <=>  fn(1, 2, 3)
  apply(fn, 1, #())         <=>  fn(1)
  apply(fn, #())            <=>  fn()
  apply(fn)                 <=>  ???
)

If I were to table a change to the DRM, it would be to provide a less
confusing explanation of the existing apply, possibly with the above
example to explain why it is the way it is (assuming I'm right).

Hugh G. Greene



Mon, 28 Jan 2002 03:00:00 GMT  
 change groups!

Quote:
> I don't think the DRM should be changed incrementally.  That is,
> changes should be collected and documented separately .  When a the
> next round of language design has been completed, that's the time
> to revise the DRM, and come out with the DRM II.

Sounds like a plan.

Andreas

--
"We show that all proposed quantum bit commitment schemes are insecure because
the sender, Alice, can almost always cheat successfully by using an
Einstein-Podolsky-Rosen type of attack and delaying her measurement until she
opens her commitment." ( http://xxx.lanl.gov/abs/quant-ph/9603004 )



Mon, 28 Jan 2002 03:00:00 GMT  
 
 [ 15 post ] 

 Relevant Pages 

1. ?list{proplist:width+proplist:group,SomeNr} does change field width, not group width

2. Rexx/PalmOS group on Yahoo (http://groups.yahoo.com/group/r-e-x-x_palmos/)

3. click the link for the best google group http://groups-beta.google.com/group/boysworld

4. Cleveland, OH, USA: Clipper/VO User Group changes

5. Starter for 10: Changing owner/group of a symlink

6. Group changes to stacking order (Tk)

7. TPF Group: COMP.OS.TPF - The Group is up and running

8. Bay area (Northern CA) Smalltalk Users group/ Special Interest Group

9. How to obtain a reference to a group within a group

10. don't make smalltalk discussion group into consultant hunter's group

11. Adding a Group Footer to an existing Group Break

12. Group level move vs sub-group level move.

 

 
Powered by phpBB® Forum Software