I think that the ALU ought to concern itself more
with local Lisp user groups in various cities, and
possibly publish (or webpublish) a regular journal.
I think "conference proceedings" tend to be both too
infrequent, and too specialized, to be very applicable
to the practice of creating programs with Lisp.
I would like to see more Lisps that aren't Common
Lisp discussed and considered; new good ideas come
from new designs.
Bear
Ray Dillinger <····@sonic.net> writes:
> I would like to see more Lisps that aren't Common
> Lisp discussed and considered; new good ideas come
> from new designs.
I agree with the second part, but not your implication that Common
Lisp is not a good platform for creating such designs. In particular,
I think revisiting the low-level design decisions that were made circa
CLtL1 is a pointless waste of time; that's not a space where there are
likely to be exciting new discoveries. A language innovation that can
coexist and intermix with CL is 1000x more interesting than some
crappy new Lisp dialect where you have to start everything all over
again. In my freetime at the moment, I'm experimenting with some
logical-programming-ish ideas, in particular trying to figure out how
to efficiently compile the system I have, or modify it to be amenable
to such. It's all freely interoperable with Common Lisp, so I can
actually use it and can concentrate on what I think are the
interesting language issues instead of having to reproduce the fine
work of many many CL designers and implementors; and the result, if
it's good, will be an addition to CL, it will extend an already nice
language, rather than make a better-in-some-ways, worse-in-others
competitor.
--
/|_ .-----------------------.
,' .\ / | Free Mumia Abu-Jamal! |
,--' _,' | Abolish the racist |
/ / | death penalty! |
( -. | `-----------------------'
| ) |
(`-. '--.)
`. )----'
Thomas F. Burdick wrote:
> Ray Dillinger <····@sonic.net> writes:
>
>
>>I would like to see more Lisps that aren't Common
>>Lisp discussed and considered; new good ideas come
>>from new designs.
>
>
> I agree with the second part, but not your implication that Common
> Lisp is not a good platform for creating such designs. In particular,
> I think revisiting the low-level design decisions that were made circa
> CLtL1 is a pointless waste of time; that's not a space where there are
> likely to be exciting new discoveries.
This is not true if other Lisps can implement something
that isn't easy to do in Common Lisp.
If some other Lisp comes along and it's easy to implement
something in it that you can't easily do in Common Lisp,
then it illustrates a limitation of Common Lisp, and that's
something that I think CL users ought to be interested in.
For example, Scheme's continuations (even if you hate 'em)
are something that's very hard to do in CL, and illuminate
a limitation of the language. If you never looked outside
CL, you'd never have realized there was a boundary of
expressiveness there that could be expanded.
I do not say that scheme-style continuations would be a
good thing to add to CL; please don't waste breath on that
old and pointless argument. I argue that looking at other
Lisps can make us aware if the Lisp we're currently using
has limitations and what those limitations are. More broadly,
it can make us aware of the design tradeoffs and alternatives.
Understand that I'm interested in Lisp. That is to say,
Lisp, the wide, branching tree of languages and dialects and
influences and design choices, from its roots in lambda
calculus to its current implementations in dozens of different
lisp dialects. Common Lisp is a single bright star in that
constellation, not the galaxy that contains it. Understanding
different lisps helps me to understand where Common Lisp (and
other lisps) fits in the Lisp Universe and what the design
choices and tradeoffs that differentiate them are and how they're
related to each other. I'm not looking for "exciting new
discoveries," although I do enjoy them and frequently benefit
from them in terms of understanding the semantics of Lisps as
they fit into a "pattern" or gestalt of design space. What
I'm looking for is that gestalt - the Lisp language design
space. Old knowledge is as valuable to me as new. Different
ways of understanding things are important because they make
parts of that gestalt easier to work with and relate to each
other.
Writing programs in Lisp is mainly a matter of designing
little languages - understanding the language design space
better is, I think, a big help no matter which Lisp you prefer.
We should not stop thinking about a language just because
we have learned to use it.
And the ALU, insofar as it limits its interest to *new*
discoveries, doesn't help to spread understanding of parts
of the design space discovered years ago. A good, clear,
readable essay about implementation design alternatives
or a detailed exposition of semantics for longstanding
widespread features has more pragmatic value me, and to
most Lispers, I think, than some esoteric paper on the
properties of type theorem provers as applied to a
strongly-typed variant of Common Lisp, even though the
one is not covering new ground and the other is.
Remember the essay about "Casting SPELs with Common Lisp"?
That's a beautiful, understandable introduction to Macros,
and I think there ought to be a Lisp Users organization
that takes an interest in publishing and promulgating such
materials, not because they're new academic results, but
because people can LEARN from them and gain new understanding
of basic features like macrology from reading them, and
become better Lispers and better language designers for
having read them -- even though they're not "new discoveries"
in the sense of formal knowledge we didn't have before.
Isn't a new discovery about a more effective way to teach
or learn or understand something as important as a new
discovery about the thing itself?
Bear
Okay, I think I can see where you're coming from more clearly now. In
general, I'm more interested in taking the best language I have
available and trying to push it forwards in terms of expressiveness
and power. At the moment that means trying to add
efficiently-compilable support for both one-way constraints and
LP-style backtracking. And I'm working on a couple tools to help
address the constant Smalltalk envy that most (all?) Lispers who've
touched a ST system suffer from. In the first case, that's a lot of
old and a little new; in the second case, it's purely a porting
effort. But neither of them involve revisiting old decisions in Lisp
design.
Not that those aren't interesting. And actually, learning about LISPs
in the LISP 1.5 and MACLISP eras has been useful in understanding
SKILL, the LISP dialect we mostly use at work. INTERLISP is always a
trip to read about, too.
> Thomas F. Burdick wrote:
> > Ray Dillinger <····@sonic.net> writes:
> >
> > > I would like to see more Lisps that aren't Common
> > > Lisp discussed and considered; new good ideas come
> > > from new designs.
Ray Dillinger <····@sonic.net> writes:
> I'm not looking for "exciting new
> discoveries," although I do enjoy them and frequently benefit
> from them in terms of understanding the semantics of Lisps as
> they fit into a "pattern" or gestalt of design space. What
> I'm looking for is that gestalt - the Lisp language design
> space.
(*cough*)
> Remember the essay about "Casting SPELs with Common Lisp"?
> That's a beautiful, understandable introduction to Macros,
Definately
> and I think there ought to be a Lisp Users organization
> that takes an interest in publishing and promulgating such
> materials, not because they're new academic results, but
> because people can LEARN from them and gain new understanding
> of basic features like macrology from reading them, and
> become better Lispers and better language designers for
> having read them -- even though they're not "new discoveries"
> in the sense of formal knowledge we didn't have before.
>
> Isn't a new discovery about a more effective way to teach
> or learn or understand something as important as a new
> discovery about the thing itself?
You'll get no argument from me on this.
--
/|_ .-----------------------.
,' .\ / | Free Mumia Abu-Jamal! |
,--' _,' | Abolish the racist |
/ / | death penalty! |
( -. | `-----------------------'
| ) |
(`-. '--.)
`. )----'
On 06 Aug 2005 01:18:19 -0700, <···@conquest.OCF.Berkeley.EDU> wrote:
>
>INTERLISP is always a trip to read about, too.
Have a recommended source? I haven't picked the right Google keywords
yet apparently.
--
For when I found the holy lisp
So pure the curves that are parens
With a report merely one short text
Of beauty's simplicity --r2q2
From: Kent M Pitman
Subject: Re: Comments, questions, gripes, or rants about the ALU?
Date:
Message-ID: <uvf2had0o.fsf@nhplace.com>
GP lisper <········@CloudDancer.com> writes:
> On 06 Aug 2005 01:18:19 -0700, <···@conquest.OCF.Berkeley.EDU> wrote:
> >
> >INTERLISP is always a trip to read about, too.
>
> Have a recommended source? I haven't picked the right Google keywords
> yet apparently.
I believe the keywords you want are (or will be):
RARE INTERLISP MANUAL FOR SALE
and you'll probably want to try them on ebay from time to time. ;)
There are probably a few other documents that obliquely treat the issue,
but the manual itself is quite a read and probably hard to obtain. :(
On Sun, 07 Aug 2005 17:09:05 GMT, <······@nhplace.com> wrote:
> GP lisper <········@CloudDancer.com> writes:
>
>> On 06 Aug 2005 01:18:19 -0700, <···@conquest.OCF.Berkeley.EDU> wrote:
>> >
>> >INTERLISP is always a trip to read about, too.
>>
>> Have a recommended source? I haven't picked the right Google keywords
>> yet apparently.
>
> I believe the keywords you want are (or will be):
>
> RARE INTERLISP MANUAL FOR SALE
Didn't seem to be that rare. Should be here in a week or so, thanks!
--
On a cloudy day,
You hear the cons cells whisper:
"We are lost and gone." -- Oliver
From: Matthias Buelow
Subject: Re: Comments, questions, gripes, or rants about the ALU?
Date:
Message-ID: <3m7u4eF14spujU2@news.dfncis.de>
Kent M Pitman <······@nhplace.com> wrote:
>I believe the keywords you want are (or will be):
> RARE INTERLISP MANUAL FOR SALE
>and you'll probably want to try them on ebay from time to time. ;)
Aren't they more or less floating around on the net? I have some
110mb of gzipped PDFs (scanned manuals) lying on my harddisk of
Interlisp-D, as of 1985, 1987, and 1988 which I just downloaded
somewhere a couple years ago out of curiosity (but haven't had the
time for a closer look so far), in particular:
1974_InterlispRefMan.pdf.gz 3101312_RelNotes_Dec1985.pdf.gz
3101272_InterlispD_1_Oct85.pdf.gz 3102464_CL_Impl_Jun1987.pdf.gz
3101273_InterlispD_2_Oct85.pdf.gz 3102465_1186HW_Jun1987.pdf.gz
3101274_InterlispD_3_Oct85.pdf.gz 400004_1108UsersGuide_Sep88.pdf.gz
If anyone is interested and cannot find them on the net, I could
put them somewhere for download, given a couple days. But I think
they should be easy enough to find on the net aswell.
mkb.
From: Kent M Pitman
Subject: Re: Comments, questions, gripes, or rants about the ALU?
Date:
Message-ID: <u3bpctga7.fsf@nhplace.com>
Matthias Buelow <···@incubus.de> writes:
> Kent M Pitman <······@nhplace.com> wrote:
>
> >I believe the keywords you want are (or will be):
> > RARE INTERLISP MANUAL FOR SALE
> >and you'll probably want to try them on ebay from time to time. ;)
>
> Aren't they more or less floating around on the net? I have some
> 110mb of gzipped PDFs (scanned manuals) lying on my harddisk of
> Interlisp-D, as of 1985, 1987, and 1988 which I just downloaded
> somewhere a couple years ago out of curiosity (but haven't had the
> time for a closer look so far), in particular:
>
> 1974_InterlispRefMan.pdf.gz 3101312_RelNotes_Dec1985.pdf.gz
> 3101272_InterlispD_1_Oct85.pdf.gz 3102464_CL_Impl_Jun1987.pdf.gz
> 3101273_InterlispD_2_Oct85.pdf.gz 3102465_1186HW_Jun1987.pdf.gz
> 3101274_InterlispD_3_Oct85.pdf.gz 400004_1108UsersGuide_Sep88.pdf.gz
>
> If anyone is interested and cannot find them on the net, I could
> put them somewhere for download, given a couple days. But I think
> they should be easy enough to find on the net aswell.
Cool. I didn't know this was electronically accessible.
What I assumed was rare to find, btw, was the original bound manuals
from Xerox. I don't know of a ready supply. But for most purposes
online copy is probably preferrable anyway. Once you collect a lot of
physical artifacts like this, it gets expensive and painful to house
and care for them unless you have extra rooms in your house or a huge
basement--something not all of us always do. And anyway, the hardcopy
will eventually fade. That's why I've been aggressively working on
webbing my papers and am always glad to see others doing the same.
Kent M Pitman wrote:
> Matthias Buelow <···@incubus.de> writes:
>
> > Kent M Pitman <······@nhplace.com> wrote:
> >
> > >I believe the keywords you want are (or will be):
> > > RARE INTERLISP MANUAL FOR SALE
> > >and you'll probably want to try them on ebay from time to time. ;)
> >
> > Aren't they more or less floating around on the net? I have some
> > 110mb of gzipped PDFs (scanned manuals) lying on my harddisk of
> > Interlisp-D, as of 1985, 1987, and 1988 which I just downloaded
> > somewhere a couple years ago out of curiosity (but haven't had the
> > time for a closer look so far), in particular:
> >
> > 1974_InterlispRefMan.pdf.gz 3101312_RelNotes_Dec1985.pdf.gz
> > 3101272_InterlispD_1_Oct85.pdf.gz 3102464_CL_Impl_Jun1987.pdf.gz
> > 3101273_InterlispD_2_Oct85.pdf.gz 3102465_1186HW_Jun1987.pdf.gz
> > 3101274_InterlispD_3_Oct85.pdf.gz 400004_1108UsersGuide_Sep88.pdf.gz
> >
> > If anyone is interested and cannot find them on the net, I could
> > put them somewhere for download, given a couple days. But I think
> > they should be easy enough to find on the net aswell.
>
> Cool. I didn't know this was electronically accessible.
>
> What I assumed was rare to find, btw, was the original bound manuals
> from Xerox. I don't know of a ready supply. But for most purposes
> online copy is probably preferrable anyway. Once you collect a lot of
> physical artifacts like this, it gets expensive and painful to house
> and care for them unless you have extra rooms in your house or a huge
> basement--something not all of us always do. And anyway, the hardcopy
> will eventually fade. That's why I've been aggressively working on
> webbing my papers and am always glad to see others doing the same.
All can be found here:
http://bitsavers.org/pdf/xerox/interlisp/
Matthias Buelow <···@incubus.de> writes:
> Aren't they more or less floating around on the net? I have some
> 110mb of gzipped PDFs (scanned manuals) lying on my harddisk of
> Interlisp-D, as of 1985, 1987, and 1988 which I just downloaded
> somewhere a couple years ago out of curiosity (but haven't had the
> time for a closer look so far), in particular:
Here is some Interlisp material:
http://bitsavers.org/pdf/xerox/interlisp/
Paolo
--
Why Lisp? http://lisp.tech.coop/RtL%20Highlight%20Film
Recommended Common Lisp libraries/tools:
- ASDF/ASDF-INSTALL: system building/installation
- CL-PPCRE: regular expressions
- UFFI: Foreign Function Interface
I don't know what program are you interested in
but international lisp conference 2005 has been a
great success .There's so much material that
draw my attention :
1SWCLOS: Semantic Web Processing in CLOS
2Performance Beyond Expectations
3A Model-Based Architecture for Entertainment Applications
and especially
4 Generating .NET Applications Using Lisp
http://www.international-lisp-conference.org/2005/schedule.html
If nothing on this list suits your problems than would be very
interesthing
to tell us what are you working on ?
Regarding second part i can not agree with you ,
Lisp community is not vast so further fragmentation
will do us more damage than good.
For my own experience i know that is very hard
to drove new faces to Common Lisp
beside it's standardization , number of quality implementations,
vast of code written , and quality of it's community .
Common Lisp is very well designed and you
have to be a hell of a hacker to make something
so much better than it would be worth the effort of
starting from scratch all of the above .