I began working at Franz nearly 3 years ago. At the time,
I barely knew how to turn on a computer, let alone had any
idea what Lisp was and how it could be used.
Gradually, I began to discover the amazing power of Lisp and
the brilliance of the developers who use it. When asked to
describe who the "typical" Lisp programmer was, I would tell
people that they were the most experienced and innovative
programmers in any domain or industry. That with Lisp
programmers, it's not a matter of "can something be done"
but "when it will be done." With Lisp, developing the
"next big thing" is not a matter of possibility, but simply
a matter of time.
I wondered why Lisp wasn't more popular. I wondered why I
seemed to know more about Lisp than the Editors at the
various Tech Publications and websites I worked with.
I wondered why no one seemed to be talking about the exciting
things that Lisp could do.
From my perspective, I think it's partially because Lisp programmers
are a unique breed. Because Lisp allows developers to do so much
on their own, they don't need huge support groups of fellow
programmers. This independence is both a blessing and a curse.
Lisp programmers are significantly more productive, but they are
sometimes uncomfortable "playing with others."
The challenge becomes how to make Lisp more widely used and
respected -- even accounting for the diverse set of opinions
that permeate the community? How can we help other serious
programmers discover the power of Lisp?
In my opinion, the answer must begin with one simple task. Lisp
developers must talk about Lisp -- to other newsgroups, to websites,
to magazines, to students, to peers. Lispers must stop talking to
only each other if Lisp is to continue to grow.
I admit, a growing community yields both pros and cons. On the
"cons" side, one is faced with "newbies" and less astute
programmers. On the "pros" side, however, the community becomes less
inbred and benefits from new ideas and experiences. As Lisp becomes
more widely known and used, it also becomes more widely respected.
Instead of a battle and plea to use Lisp in an application, it can
become the standard and expectation. Imagine a world where one could
tell a manager that Lisp was the right tool for the job, and have
him/her agree!
So, help spread the word about Lisp!
Post to comp.lang.perl/python/C++ or on Slashdot. Of course, this
doesn't mean to make off-topic posts -- be sure to address a specific
question on the particular ng/website -- but I have seen numerous
comments on these locations that cry out for a lisp-type response.
In fact, when Lisp is mentioned, it tends to generate a significant
response. Paul Graham's article, "Beating the Averages" (whether you
agree with it or not) generated a tremendous discussion on Slashdot.
As did articles referencing AARON (Harold Cohen's cybernetic artist
that was built in Lisp) or the Final Fantasy Movie that used Lisp as
the foundation of its production system. When Duane Rettig posted
some lisp-centric articles on comp.arch he received quite a bit of
positive feedback. Further, when he posted an article in
comp.lang.dylan one day, a reader complimented him on the articles
he'd posted on comp.arch.
Speak about Lisp at your local university. You can also post the code
of your application on CLIKI or Sourceforge. Submit white papers to
relevant publications or conferences. Email me and volunteer to be
put on the list that I keep of speakers, writers and press contacts
(You don't have to be an Allegro CL user, just a Lisp user).
To the rest of the world, it doesn't matter which type of Lisp is used
-- only that Lisp is used at all. One voice won't do it. But many
voices, even if they're saying their own thing, can make one heck of a
roar!
Lisa Fettner
Marketing Director
Franz Inc.
····@franz.com (Lisa Fettner) writes:
> In my opinion, the answer must begin with one simple task. Lisp
> developers must talk about Lisp -- to other newsgroups, to websites,
> to magazines, to students, to peers. Lispers must stop talking to
> only each other if Lisp is to continue to grow.
I am a Lisp developer. However, I do not feel I _must_ talk about
Lisp. I will, if there's a reason for it. Getting paid for it would
be a good one. (Will travel, reasonable rates.)
It's a tool that I happen to use. I have chosen to use it because I
think it is the best choice, considering the alternatives. Now, if
the next guy uses it, good for him. If not, I don't have the time
nor the energy to save him from his misguided ways (unless he asks
me to). I didn't go to the university to become a missionary.
> So, help spread the word about Lisp!
I am trying to do that, though. How? By setting an example. The only
way those who do not already get it ever could is by seeing that
there is something worthwhile (=profitable) to this Lisp thing.
Faster time to market, fewer bugs, less brain hemorrhages within
developers, whatever. But the word alone is not enough. There has to
be something concrete to show for it.
And if we do not succeed, well, maybe we are/were lousy businessmen,
or the stars were misaligned, or something else. But in that case
using Lisp did not save us. Only the success stories will be
remotely interesting to anyone outside those who have seen the light
anyway.
> Post to comp.lang.perl/python/C++ or on Slashdot.
While it is an old tradition on Usenet to just post on groups and
not read them, I think I'll skip on the posting too, thanks.
--
Ola Rinta-Koski ···@cyberell.com
Cyberell Oy +358 41 467 2502
Rauhankatu 8 C, FIN-00170 Helsinki, FINLAND www.cyberell.com
Lisa Fettner wrote:
> I began working at Franz nearly 3 years ago. At the time,
> I barely knew how to turn on a computer, let alone had any
> idea what Lisp was and how it could be used.
>
> Gradually, I began to discover the amazing power of Lisp and
> the brilliance of the developers who use it. When asked to
> describe who the "typical" Lisp programmer was, I would tell
> people that they were the most experienced and innovative
> programmers in any domain or industry. That with Lisp
> programmers, it's not a matter of "can something be done"
> but "when it will be done." With Lisp, developing the
> "next big thing" is not a matter of possibility, but simply
> a matter of time.
Don't underestimate the value and importance of "time".
> I wondered why Lisp wasn't more popular. I wondered why I
> seemed to know more about Lisp than the Editors at the
> various Tech Publications and websites I worked with.
> I wondered why no one seemed to be talking about the exciting
> things that Lisp could do.
>
> From my perspective, I think it's partially because Lisp programmers
> are a unique breed. Because Lisp allows developers to do so much
> on their own, they don't need huge support groups of fellow
> programmers. This independence is both a blessing and a curse.
> Lisp programmers are significantly more productive, but they are
> sometimes uncomfortable "playing with others."
I'm not sure if this is true or at least a viable approximation of how
real life (tm) works. What I've realized so far is that _many_ Lispers
have experience with alot of other languages and have developed some
kind of "pragmatic" view. Common Lisp draws people on it's side that
already "hit" the wall of complexity using languages like C++ or Java.
It is _very_ difficult to fascinate people for a language like CL if
they had no real problems using their language so far.
> The challenge becomes how to make Lisp more widely used and
> respected -- even accounting for the diverse set of opinions
> that permeate the community? How can we help other serious
> programmers discover the power of Lisp?
by writing code... "running code counts"
> In my opinion, the answer must begin with one simple task. Lisp
> developers must talk about Lisp -- to other newsgroups, to websites,
> to magazines, to students, to peers. Lispers must stop talking to
> only each other if Lisp is to continue to grow.
I've written three articles on Common Lisp so far and have introduced
general interest in Lisp at the University I go to. (They offer a
"functional Programming" course using Scheme for the first time now...).
Next term I will hold a talk on "Common Lisp" in the course "Concepts
of higher Programming Languages".
> I admit, a growing community yields both pros and cons. On the
> "cons" side, one is faced with "newbies" and less astute
> programmers. On the "pros" side, however, the community becomes less
> inbred and benefits from new ideas and experiences. As Lisp becomes
> more widely known and used, it also becomes more widely respected.
> Instead of a battle and plea to use Lisp in an application, it can
> become the standard and expectation. Imagine a world where one could
> tell a manager that Lisp was the right tool for the job, and have
> him/her agree!
Many managers or even developers do not really have the insight or
education to see the benefit a language like Common Lisp provides over e.
g. Java.
> So, help spread the word about Lisp!
I try what I can...
> Post to comp.lang.perl/python/C++ or on Slashdot. Of course, this
> doesn't mean to make off-topic posts -- be sure to address a specific
> question on the particular ng/website -- but I have seen numerous
> comments on these locations that cry out for a lisp-type response.
I remember a post I replied to on de.comp.lang.java with "Computing
Permutations" as topic. I've shown some pseudocode (similar to Java)
to some algorithms I've known and understood and an additional algorithm
written in Lisp that I had not fully understood yet but had some benefits
over the others. I got criticized _heavily_ that my pseudocode was not
correct Java and that the Lispcode was somewhat an insult to the Java
programmers.
> In fact, when Lisp is mentioned, it tends to generate a significant
> response. Paul Graham's article, "Beating the Averages" (whether you
> agree with it or not) generated a tremendous discussion on Slashdot.
> As did articles referencing AARON (Harold Cohen's cybernetic artist
> that was built in Lisp) or the Final Fantasy Movie that used Lisp as
> the foundation of its production system. When Duane Rettig posted
> some lisp-centric articles on comp.arch he received quite a bit of
> positive feedback. Further, when he posted an article in
> comp.lang.dylan one day, a reader complimented him on the articles
> he'd posted on comp.arch.
It depends highly on what people you meet...
> Speak about Lisp at your local university. You can also post the code
> of your application on CLIKI or Sourceforge. Submit white papers to
> relevant publications or conferences. Email me and volunteer to be
> put on the list that I keep of speakers, writers and press contacts
> (You don't have to be an Allegro CL user, just a Lisp user).
If you want, you can put me on that list.
ciao,
Jochen Schmidt
--
http://www.dataheaven.de
Hi. I'm just a lurker :)
> > The challenge becomes how to make Lisp more widely used and
> > respected -- even accounting for the diverse set of opinions
> > that permeate the community? How can we help other serious
> > programmers discover the power of Lisp?
>
> by writing code... "running code counts"
>
I post this because, I find, it has some anedoctal flavor:
the last release of Netscape 4.7, that my sysadmin
has installed, precisely the news reader crashes
without explanation when I try to access comp.lang.lisp
(under Solaris).
A CL www browser would probably not have
such sudden crashes... (up to low-level code but
I imagine the trick is to keep that minimal and heavily
tested ;)
[Is just an hypothesis, I'm not asking someone
to write such a thing]
It seems that there is no "commodity" software
around written in CL ... but people happily
live with crashes, even programmers...
I stop here, I don't comment further,
don't know if this was funny or interesting
but in any case not against CL.
regards, Samuele Pedroni.
"Samuele Pedroni" <·······@inf.ethz.ch> writes:
> I post this because, I find, it has some anedoctal flavor:
> the last release of Netscape 4.7, that my sysadmin
> has installed, precisely the news reader crashes
> without explanation when I try to access comp.lang.lisp
> (under Solaris).
>
> A CL www browser would probably not have
> such sudden crashes... (up to low-level code but
> I imagine the trick is to keep that minimal and heavily
> tested ;)
> [Is just an hypothesis, I'm not asking someone
> to write such a thing]
>
> It seems that there is no "commodity" software
> around written in CL ... but people happily
> live with crashes, even programmers...
There actually was a web browser written in CL, but I think it stopped
being maintained a long time ago. I started to write a web browser in
CL once, but there's not that much of the problem that's very
interesting, and I decided I had better things to do with my time. It
was kind of fun writing an html-parsing framework that could be
incrementally extended to deal with all sorts of cases of broken
input. Because a browser that can only parse valid or semi-valid html
isn't very useful. Actually figuring out and writing up the various
rules for finding and correcting mistakes was really boring, though.
I'm surprised how many modern web browsers there are, actually (IE,
NS, Mozilla, Konqueror, Opera) considering what a mind-numbing
exercise I found it to be. Maybe it gets good again later on...
···@famine.OCF.Berkeley.EDU (Thomas F. Burdick) writes:
> There actually was a web browser written in CL, but I think it
> stopped being maintained a long time ago.
Closure, the web browser written in Lisp:
http://www.uni-karlsruhe.de/~unk6/closure/
Chris.
--
http://www.double.co.nz/cl
Jochen Schmidt <···@dataheaven.de> wrote:
> Many managers or even developers do not really have the insight or
> education to see the benefit a language like Common Lisp provides over e.
> g. Java.
>
> > So, help spread the word about Lisp!
>
> I try what I can...
Emacs can be used to introduce Lisp and its benefits as an
environment and language:
- teach the basics of Emacs and how to learn Emacs alone
(Tutorial, Info, etc.)
- go beyond the basics and show stuff such as ange-ftp, dired,
jka-compr, VC, shell, macros, etc. so that they realize that
Emacs is much much more (useful) than what they knew/thought and
that actually, most editing tasks that would require some Perl
or Python could be done in Emacs without writting a single line
of programming and messing the file system
- talk about configuring Emacs, by using *scratch* to set some
simple variables, a good time to talk about Lisp and show the
interactive nature of Lisp with *scratch* and C-j, like:
- define a function and use it immediately, oups, it is
bugged, well, edit and re-evaluate
- talk about load and load-path to introduce lists
- talk about hooks to mix functions and lists
- create some useful functions and use them
- don't forget to tell them that Emacs Lisp is an old Lisp,
that's why there's the CL package, which is a subset of
Common Lisp (good moment to introduce it)
- etc.
- let's say that some programs (written or used by colleagues,
manager, etc.) generate useful files (trace, log, data, ...): a
major mode can be written to navigate in these trace files and
provide extra functions (like data mining, data conversion,
user-friendly views of data, ...) and show them
- use Emacs to simulate some of your company's applications
functions to show how you can customize the stuff on the fly
(for example, by using gnudoit to simulate client-server) that
wouldn't be possible on the current applications without big
efforts, especially when it involves output format, calculation
formula, functions, etc. and make them (colleagues, manager,
...) wonder what kinds of leverage they could get if they had
the same capabilities in the applications...
- etc.
--
Georges Ko (Taipei, Taiwan) 2001-09-08 ···@gko.net / ICQ: 8719684
Cycle 78, year 18 (Xin-Si), month 7 (Bing-Shen), day 21 (Jia-Xu)
On 7 Sep 2001 17:13:20 -0700, ····@franz.com (Lisa Fettner) wrote:
> I began working at Franz nearly 3 years ago. At the time,
> I barely knew how to turn on a computer, let alone had any
> idea what Lisp was and how it could be used.
I discovered Lisp more than 20 years ago, at a time when graphic screens
were in _very_ limited use, usually vector screens in CAD systems. And
then Xerox, and Symbolics and LMI and Texas put out Lisp systems with
graphics capabilities, bitmapped screens, windowing and integrated
development support, at a time when almost all programmers were using a
simple ascii screen. Whoa
> .. With Lisp, developing the "next big thing" is not a matter of
> possibility, but simply a matter of time.
That's the thing. Some Lispers get annoyed that so much of the
functionality in Lisp systems has disappeared. That means taking time to
set up what should have been there in the first place. Unless they buy
ACL of course? :-)
> I wondered why Lisp wasn't more popular. I wondered why I
> seemed to know more about Lisp than the Editors at the
> various Tech Publications and websites I worked with.
> I wondered why no one seemed to be talking about the exciting
> things that Lisp could do.
Lisp, being a programming language, does nothing. The programs created
by using Lisp _may_ do a lot. If the programmer has used a good Lisp
implementation (some say Lisp system), the program can do lots and lots.
But it depends on the people using Lisp, and people is your main theme
here.
> From my perspective, I think it's partially because Lisp programmers
> are a unique breed.
A Lisper is more of a problem solver than a programmer. Efficient use of
Lisp requires a certain affinity to abstract thinking, and lots of
serious programmers do not have that. One may learn techniques to some
degree, but I think that Lispers have this affinity in the personality
from the beginning. Perhaps this is what causes you to say "breed"?
> Because Lisp allows developers to do so much
> on their own, they don't need huge support groups of fellow
> programmers.
See above why there will probably never be huge groups of Lisp
programmers. This is what scares most managers, whether rightly or
wrongly. They view people in terms of "thinkers" and "doers", and are
not overly fond of having too many "thinkers". "Doers" are, they think,
the salt of the earth and people they can depend upon.
> Lisp programmers are significantly more productive, but they are
> sometimes uncomfortable "playing with others."
What kind of productivity are you thinking of? Creating powerful
software which is largely used only by yourself and with no observable
results may not seem very productive to the outside world ...
Maybe the term here is "effective"? That may be so, and partly because
they use effective support tools and usually cover larger ground when
the program. But this needs training and it would erronous to say that a
good Lisp system is easy to learn. You can't win new Lispers that way
..
> In my opinion, the answer must begin with one simple task. Lisp
> developers must talk about Lisp -- to other newsgroups, to websites,
> to magazines, to students, to peers. Lispers must stop talking to
> only each other if Lisp is to continue to grow.
I like that word "continue".
> I admit, a growing community yields both pros and cons. On the
> "cons" side, one is faced with "newbies" and less astute
> programmers. On the "pros" side, however, the community becomes less
> inbred and benefits from new ideas and experiences.
Such as having old fogies like me harping about the Lisp systems that
once were?
> more widely known and used, it also becomes more widely respected.
> Instead of a battle and plea to use Lisp in an application, it can
> become the standard and expectation. Imagine a world where one could
> tell a manager that Lisp was the right tool for the job, and have
> him/her agree!
I assume you mean "Lisp _is_ the right tool.."? Then there is no
problem. Just give him a sound business case with all reasonable short-
and long-term aspects covered.
> To the rest of the world, it doesn't matter which type of Lisp is used
> -- only that Lisp is used at all.
It is true that bulk advertizing will attract the uninformed.
What might attract the people holding the money bag is what kind of
applications that were created with Lisp, and the circumstances around
the project. An experienced manager is usually wary when the arguments
are mostly dealing with technical elegance. He has seen too many
non-technical problems in software projects. Show him that an
implementation in Lisp could have solved those other problems, and you
are on your way, I think.
Lars
----- Original Message -----
From: "Lisa Fettner" <····@franz.com>
> To the rest of the world, it doesn't matter which type of Lisp is used
> -- only that Lisp is used at all. One voice won't do it. But many
> voices, even if they're saying their own thing, can make one heck of a
You can do it. Actually,there is a discussion on slashdot:
http://slashdot.org/developers/01/09/08/0113203.shtml
about Lisp as an alternative to Java. This theme is not new and has emerged
on slashdot a few month ago. Now it has re-appeared with a new thread.
As a dealer you should get wrapped before start reading...
S. Gonzi