From: Kenny Tilton
Subject: [ANNC] Lisp NYC RoboTrader SIG Meeting #1, Wed, Dec 17
Date:
Message-ID: <3FDCECC9.4070202@nyc.rr.com>
Who: Russ McManus and The LispNYKs and others
What: Introduction to the FIX Protocol: http://www.fixprotocol.org/
When: This coming Wednesday, December 17, 2003 at 7:00pm
Where: 171 West 73rd St, Apt 3 (btw Columbus & Amsterdam)
aka Kenny's studio/office (take 1-2-3-9 to 72nd & Broadway)
Why: Possible RoboTrader Project for Lisp-NYC (and others?)
Deets: Russ has already started work on a FIX System using Lisp. He has
suggested Lisp-NYC undertake a project to extend his work and develop it
into a trading exchange simulator suitable for trading competitions
between students of financial engineering, in which autonomous agents
would execute trading strategies in competition with each other, much
like RoboCup does with soccer.
Russ has a good contact at a local university where he thinks such a
project would be greeted with interest. (And I just remembered that a
good friend is the retired chair of the NYU School of Economics
(Finance?). Anyway...)
A quick survey of Lisp-NYC members revealed sufficient (substantial,
actually) interest to move to the next step, which will take place at
the time and place shown above and consist first of Russ's presentation
on FIX and his progress to date, followed by a discussion as to whether
to proceed with a Lisp-NYC RoboTrader (working name) SIG to develop a
trading exchange server suitable for student competitions.
Initial investigation reveals that other such competitions are hosted by
vendors of trading software, though at least one is run by a university
on top of a vendor's system.
A Lisp-NYC project would have the advantage of exposing students in the
financial sector to Lisp, and will let Lisp-NYKs bone up on FIX and
FIXML and enhance their resumes in ways attractive to the significant
NYC financial tech employers. (See the "Help Wanted" section of the FIX
site, URL above).
Folks are welcome to drop by Wednesday at 7 to kick things around. RSVP
if you can. And for those who cannot make it but are interested: drop me
a line as well so we can assess the interest level.
kenny
--
http://tilton-technology.com
Why Lisp? http://alu.cliki.net/RtL%20Highlight%20Film
Your Project Here! http://alu.cliki.net/Industry%20Application
I'd like to offer what help I can. I'm new to lisp, but not finance. My
job is creating models for exotic derivatives, and have written trading
strategies in the past.
It's good to see such a project, as in my experience it's rare, at least
in London, to find students of finance who are good coders, most seem to
prefer to avoid it, or at best can hack a little in VB and Excel.
I live in France so won't be there on Wednesday, but good luck.
Alex
Alex McGuire wrote:
> I'd like to offer what help I can. I'm new to lisp, but not finance. My
> job is creating models for exotic derivatives, and have written trading
> strategies in the past.
Super. The RoboCup project was packed with Lisp newbies. This is
probably a great low-pressure way to pick up a new language, especially
since you have the domain itself down cold.
>
> It's good to see such a project, as in my experience it's rare, at least
> in London, to find students of finance who are good coders, most seem to
> prefer to avoid it, or at best can hack a little in VB and Excel.
I bet Lisp would be much more approachable to non-programmers. What do
you think based on your experience? They don't have to worry about
types, there is not much syntax, dynamic/interactive for exploratory
learning, even clearer runtime and compiler errors, I dare say.
>
> I live in France so won't be there on Wednesday, but good luck.
We'll keep you posted. We'll certainly need sample TraderBots to test
the exchange with, and ideas for enhancing the exchange server. And you
certainly wouldn't be the first telecommuting team member I have worked
with.
kt
--
http://tilton-technology.com
Why Lisp? http://alu.cliki.net/RtL%20Highlight%20Film
Your Project Here! http://alu.cliki.net/Industry%20Application
Kenny Tilton wrote:
>>
>> It's good to see such a project, as in my experience it's rare, at
>> least in London, to find students of finance who are good coders, most
>> seem to prefer to avoid it, or at best can hack a little in VB and Excel.
>
>
> I bet Lisp would be much more approachable to non-programmers. What do
> you think based on your experience? They don't have to worry about
> types, there is not much syntax, dynamic/interactive for exploratory
> learning, even clearer runtime and compiler errors, I dare say.
One recent experience with non-programmers was with my daughter, who's
8. Together we knocked up something to play the board game Mastermind.
She typed while I explained what was going on. I found that after very
little time the syntax became intuitive to her, and so we could
concentrate on the algorithms involved, and not any secondary issues.
I suppose that in finance Excel is seductive as it requires _very_
little computing experience to knock up a simple app, of course you
quickly run into problems as your requirements become more complex, but
that doesn't seem to stop people. I've seen some huge, horrendously
complex spreadsheets used to manage positions worth tens of millions of
dollars.
Alex
From: Espen Vestre
Subject: Re: [ANNC] Lisp NYC RoboTrader SIG Meeting #1, Wed, Dec 17
Date:
Message-ID: <kwekv6b3qe.fsf@merced.netfonds.no>
Alex McGuire <····@alexmcguire.com> writes:
> One recent experience with non-programmers was with my daughter, who's
> 8. Together we knocked up something to play the board game
> Mastermind. She typed while I explained what was going on. I found
> that after very little time the syntax became intuitive to her, and so
> we could concentrate on the algorithms involved, and not any secondary
> issues.
Interesting. This was Common Lisp?
I've considered teaching a little programming to my kids, and was
wondering if I should have a look at Logo, or if I should use lisp
straight away.
--
(espen)
Espen Vestre <·····@*do-not-spam-me*.vestre.net> writes:
> I've considered teaching a little programming to my kids, and was
> wondering if I should have a look at Logo, or if I should use lisp
> straight away.
Squeak (Smalltalk) may be an option, too.
--
Raymond Wiker Mail: ·············@fast.no
Senior Software Engineer Web: http://www.fast.no/
Fast Search & Transfer ASA Phone: +47 23 01 11 60
P.O. Box 1677 Vika Fax: +47 35 54 87 99
NO-0120 Oslo, NORWAY Mob: +47 48 01 11 60
Try FAST Search: http://alltheweb.com/
> Espen Vestre <·····@*do-not-spam-me*.vestre.net> writes:
>
>
>>I've considered teaching a little programming to my kids, and was
>>wondering if I should have a look at Logo, or if I should use lisp
>>straight away.
That's something I always wonder about. I actually did quite a bit of
intense Logo using it as a serious development tool, not for the turtle
graphics. I guess one downside might be not having the iterators (tho
come to think of it, it has been twenty years since I played with Logo)
so you have to use recursion. And Logo has the turtle graphics and other
educational support (books aimed at younger learners).
But strictly from the language perspective, I don't know. Logo does not
have all the parentheses, but we also know that parentheses are not a
problem and in fact we prefer them. Maybe Alex could ask his daughter
what she thinks of Logo (and/or parentheses).
kt
--
http://tilton-technology.com
Why Lisp? http://alu.cliki.net/RtL%20Highlight%20Film
Your Project Here! http://alu.cliki.net/Industry%20Application
From: Joe Marshall
Subject: Re: [ANNC] Lisp NYC RoboTrader SIG Meeting #1, Wed, Dec 17
Date:
Message-ID: <d6apq1oz.fsf@ccs.neu.edu>
Kenny Tilton <·······@nyc.rr.com> writes:
> But strictly from the language perspective, I don't know. Logo does
> not have all the parentheses, but we also know that parentheses are
> not a problem and in fact we prefer them. Maybe Alex could ask his
> daughter what she thinks of Logo (and/or parentheses).
The reason I did *not* teach my kids Logo is that I think there are
some fundamental problems with the language.
First of all, dynamic scope (by default) is just wrong. It is true
that this makes very little difference to start with, but it is an
extra burden when you try to move to higher-order functions.
Second, the language is just too damn imperative: move here, draw
that, go there, draw it again. It severely reinforces the `recipe'
style of programming. Again, this leads to quick gratification, but
it adds a hurdle to overcome when you need to advance to more powerful
styles of programming.
Unfortunately, I have yet to find something that I really *like*.
In article <············@ccs.neu.edu>, Joe Marshall <···@ccs.neu.edu>
wrote:
> The reason I did *not* teach my kids Logo is that I think there are
> some fundamental problems with the language.
I explored introducing my six year old girl to Logo this
summer, but the books I found were kind of goofy and packed
full of example programs that I thought she would find too
mathematical and/or boring. (For example: Here's a routine
to draw a triange. Can you change it to draw a square?
How about a pentagon?)
> Unfortunately, I have yet to find something that I really *like*.
Then I stumbled upon the book that's the basis of this
website:
http://dbn.media.mit.edu/
He has written a stripped down Logo-like language that runs
as a Java applet. Also, the examples in his book produce
images that I think my daughter willl find (more) interesting.
I bought the book ($12 at powells.com), but then fall came,
school began, and now there's no time for programming lessons.
I plan to walk her through the book next summer.
Kenny Tilton wrote:
> But strictly from the language perspective, I don't know. Logo does not
> have all the parentheses, but we also know that parentheses are not a
> problem and in fact we prefer them. Maybe Alex could ask his daughter
> what she thinks of Logo (and/or parentheses).
This would indeed be an interesting data point. (I am serious.)
Pascal
--
Tyler: "How's that working out for you?"
Jack: "Great."
Tyler: "Keep it up, then."
Espen Vestre wrote:
> Alex McGuire <····@alexmcguire.com> writes:
>
>
>>One recent experience with non-programmers was with my daughter, who's
>>8. Together we knocked up something to play the board game
>>Mastermind. She typed while I explained what was going on. I found
>>that after very little time the syntax became intuitive to her, and so
>>we could concentrate on the algorithms involved, and not any secondary
>>issues.
>
>
> Interesting. This was Common Lisp?
It was MIT scheme, but the program was so simple you could have
translated it into Lisp trivially.
>
> I've considered teaching a little programming to my kids, and was
> wondering if I should have a look at Logo, or if I should use lisp
> straight away.
I've no experience of Logo, but I would recommend at least giving lisp a
try.
<pure conjecture>
People find lisp syntax a hurdle, and so imagine it would be even more
so for a child, but I'm not sure that's the case. I think that kids have
an advantage over adults in that their brains are highly geared to
language development, hence they have no problem associating arbitrary
symbols with things, for example, I've seen children learn emacs
key-bindings in much less time than it would take an adult.
</pure conjecture>
Alex McGuire <····@alexmcguire.com> writes:
> <pure conjecture>
> People find lisp syntax a hurdle, and so imagine it would be even more
^^^^^^
Programmers
Telling non-programmers, "Parentheses surround lists, and everything
is made up of lists" has worked damn well for me in explaining "the
weird syntax of that file". At least for people who haven't been
exposed to many for-programmers file formats besides HTML.
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
Thomas F. Burdick:
> Alex McGuire <····@alexmcguire.com> writes:
> > <pure conjecture>
> > People find lisp syntax a hurdle, and so imagine it would be even more
> ^^^^^^
> Programmers
>
> Telling non-programmers, "Parentheses surround lists, and everything
> is made up of lists" has worked damn well for me in explaining "the
> weird syntax of that file". At least for people who haven't been
> exposed to many for-programmers file formats besides HTML.
I second that. I have recently been teaching the basics of programming
to a person that has done little of it until now (not a child), and from
some experiments it seems that Lisp syntax works very well, much better
than Python syntax.
"Think mathematics. Everything is an expression. Evaluate this thing in
the middle (innermost list), and then proceed outwards..." -- this kind
of explanation makes a *lot* of sense to some people!
--J.
From: Arthur Lemmens
Subject: Re: [ANNC] Lisp NYC RoboTrader SIG Meeting #1, Wed, Dec 17
Date:
Message-ID: <oprz8ky7dfk6vmsw@news.xs4all.nl>
Alex McGuire <····@alexmcguire.com> wrote:
> Espen Vestre wrote:
>> I've considered teaching a little programming to my kids, and was
>> wondering if I should have a look at Logo, or if I should use lisp
>> straight away.
>
> I've no experience of Logo, but I would recommend at least giving lisp a try.
I gave my 11 year old son chapter 2 of Peter Seibel's new book; it shows
how you can build a simple CD database. My son likes it a lot so far.
All I've done is show him the editor and the listener, help him when he
gets error messages, and explain the modulo concept.
> People find lisp syntax a hurdle, and so imagine it would be even more so
> for a child, but I'm not sure that's the case.
I don't think so. Some grown-ups find Lisp syntax a hurdle because it's
different from what they're used to. But kids don't have this heavy mental
investment in infix syntax and are a lot more flexible.
Arthur Lemmens
From: Espen Vestre
Subject: Re: [ANNC] Lisp NYC RoboTrader SIG Meeting #1, Wed, Dec 17
Date:
Message-ID: <kwllpd9mjf.fsf@merced.netfonds.no>
Arthur Lemmens <········@xs4all.nl> writes:
>> People find lisp syntax a hurdle, and so imagine it would be even more so
>> for a child, but I'm not sure that's the case.
>
> I don't think so. Some grown-ups find Lisp syntax a hurdle because it's
> different from what they're used to. But kids don't have this heavy mental
> investment in infix syntax and are a lot more flexible.
Exactly. My experience in teaching lisp to complete newbies and advising
programmers moving to lisp shows me that there's only _one_ group of
people that have real problems with sexp syntax: Programmers that are
not interested in learning lisp.
--
(espen)
Alex McGuire wrote:
> Kenny Tilton wrote:
>
>>>
>>> It's good to see such a project, as in my experience it's rare, at
>>> least in London, to find students of finance who are good coders,
>>> most seem to prefer to avoid it, or at best can hack a little in VB
>>> and Excel.
>>
>>
>>
>> I bet Lisp would be much more approachable to non-programmers. What do
>> you think based on your experience? They don't have to worry about
>> types, there is not much syntax, dynamic/interactive for exploratory
>> learning, even clearer runtime and compiler errors, I dare say.
>
>
> One recent experience with non-programmers was with my daughter, who's
> 8. Together we knocked up something to play the board game Mastermind.
> She typed while I explained what was going on. I found that after very
> little time the syntax became intuitive to her, and so we could
> concentrate on the algorithms involved, and not any secondary issues.
Cool. I was thinking the uniformity of (<operation> <args>) would be
more important than the familiarity of something like "3*x + 2".
Especially since most programming does not involve typing in math
formulas (mastermind being a good example).
btw, Mastermind was my first app on the Apple II. Actually had it
published and made a few hundred in royalties. I was going to admit that
I was 28 and give your daughter credit for a big win, but then I
remembered: I was using Apple's Integer Basic, she gets to use Lisp! :)
>
> I suppose that in finance Excel is seductive as it requires _very_
> little computing experience to knock up a simple app, of course you
> quickly run into problems as your requirements become more complex, but
> that doesn't seem to stop people. I've seen some huge, horrendously
> complex spreadsheets used to manage positions worth tens of millions of
> dollars.
I never thought about it, but I imagine it's a little hard working on
anything complex when the source is visible only one cell at a time.
kt
--
http://tilton-technology.com
Why Lisp? http://alu.cliki.net/RtL%20Highlight%20Film
Your Project Here! http://alu.cliki.net/Industry%20Application
>>>>> "Kenny" == Kenny Tilton <·······@nyc.rr.com> writes:
Kenny> Alex McGuire wrote:
>> Kenny Tilton wrote:
>>
>
> It's good to see such a project, as in my experience it's rare, at
> least in London, to find students of finance who are good coders,
> most seem to prefer to avoid it, or at best can hack a little in VB
> and Excel.
>
>
>
> I bet Lisp would be much more approachable to non-programmers. What
> do you think based on your experience? They don't have to worry about
> types, there is not much syntax, dynamic/interactive for exploratory
> learning, even clearer runtime and compiler errors, I dare say.
>> One recent experience with non-programmers was with my daughter,
>> who's
>> 8. Together we knocked up something to play the board game
>> Mastermind. She typed while I explained what was going on. I found
>> that after very little time the syntax became intuitive to her, and
>> so we could concentrate on the algorithms involved, and not any
>> secondary issues.
Kenny> Cool. I was thinking the uniformity of (<operation> <args>)
Kenny> would be more important than the familiarity of something like
Kenny> "3*x + 2". Especially since most programming does not involve
Kenny> typing in math formulas (mastermind being a good example).
Uniformity is nice also in other aspects. Compare teaching someone
Scheme and one of the infer^H^H^H^H^Himperative languages. How do I use
this function? Do I write "mutate(thing)", or do I write
"that=mutate(thing)"?
--J.
Alex McGuire <····@alexmcguire.com> writes:
>> I bet Lisp would be much more approachable to non-programmers. What
>> do you think based on your experience? They don't have to worry
[...]
> One recent experience with non-programmers was with my daughter, who's
> 8. Together we knocked up something to play the board game
> Mastermind. She typed while I explained what was going on. I found
Be sure to let her fill Kenny's Road to Lisp survey :)
Paolo
--
Why Lisp? http://alu.cliki.net/RtL%20Highlight%20Film