Lisp - Re: Yet another Lisp Myth debunked

Author

Re: Yet another Lisp Myth debunked

Cameron MacKinnon

2004-04-23, 12:50 pm

Duane Rettig wrote:
>
> It seems as though this quip to what even I thought was a generally
> true statement has caused a flurry of anecdotal evidence to the
> contrary. Of course, it is only anecdotal and self-selecting, but
> I am surprised by the numbers and especially by individuals whom I
> would have thought had started Lisp much earlier in life. Very
> interesting.

To be pedantic, PG only said that *he* couldn't convince programmers of
a certain age of the value of Lisp.

That said, I myself am a counterexample, as his writings were part of my
motivation.

I'd be inclined to think that the over 25 market would be more likely to
be receptive to radical new :-) technologies. They've worked long enough
with the industry standards not to believe the hype, and are receptive
to solutions to the problems they've come to understand.

Younger programmers would be more susceptible to expensive marketing
campaigns by the usual vendors, and more focused on the skills listed in
the employment ads.

--
Cameron MacKinnon
Toronto, Canada

Duane Rettig

2004-04-23, 1:38 pm

Cameron MacKinnon <cmackin+nn@clearspot.net> writes:

> Duane Rettig wrote:
>
>
> To be pedantic, PG only said that *he* couldn't convince programmers
> of a certain age of the value of Lisp.

Pedantry is only effective (in annoying people) if it is correct, and
you've gone and refuted your own pedantry.

But to practice a little one-upsmanship on your pedantry, that is not
actually what PG said.

Consider the four phrases:

"I don't expect for myself to convince ..."

"I don't expect us to convince ..."

"I don't expect this article to convince ..."

"I don't expect anyone to convince ..."

All four interpretations are equally correct, and might be applicable
depending on the context. If you look at the article from which the
statement came (http://www.paulgraham.com/avg.html) and especially the
surrounding paragraph, it is clear that he intended the third meaning,
not the first. However, even with that interpretation the statement
is false, since many I have talked to have attested to coming to
Common Lisp based on his books _and_ his articles, including this one,
and I wouldn't doubt that some are over 25.

Note also that in the article the parenthesized phrase (over 25) is a
link to an article by a high school graduate. Interestinly, this article
was not about learning Common Lisp, but about learning Scheme. Perhaps
there is some correlation there.

> That said, I myself am a counterexample, as his writings were part of
> my motivation.

And indeed, you are not the only one who has mentioned being a
counterexample even to the first interpretation of PG's statement.

> I'd be inclined to think that the over 25 market would be more likely
> to be receptive to radical new :-) technologies. They've worked long
> enough with the industry standards not to believe the hype, and are
> receptive to solutions to the problems they've come to understand.

I agree with this argument, but it leads me to the opposite conclusion -
that the older one gets the more one wants to go with the tried and
true, or at least with the grounded, not with the radical new...

> Younger programmers would be more susceptible to expensive marketing
> campaigns by the usual vendors, and more focused on the skills listed
> in the employment ads.

Agreed.

--
Duane Rettig duane@franz.com Franz Inc. http://www.franz.com/
555 12th St., Suite 1450 http://www.555citycenter.com/
Oakland, Ca. 94607 Phone: (510) 452-2000; Fax: (510) 452-0182

William Bland

2004-04-23, 2:36 pm

On Fri, 23 Apr 2004 11:35:38 -0400, Cameron MacKinnon wrote:

> Younger programmers would be [...] more focused on the skills
> listed in the employment ads.

We don't have much choice. As a young(ish) programmer, I don't have a lot
of successful projects I can list when I apply for a job. I can give a
list of languages that I am more or less fluent in, but the only one that
people seem to pay any attention to is Java. Most interviewers I've
talked to have been a *lot* more interested in the fact that I can write
Java than the fact that I have a Ph.D. in mathematics, or that I've been
programming computers since I was 9 years old. I do have projects I've
done in my spare time that I consider successes. Unfortunately, again,
the interviewers I've come into contact with ask me to restrict the
discussion to "commercial" projects. I would hope that, as I get older,
I'll be able to list successful commercial projects and then when asked
what language I would use for the project I'm being interviewed for I'll
say "Lisp". Perhaps I'm naive to think the successful projects will
outweigh the "weird" choice of languages in any interviewer's mind?

Cheers,
Bill.
--
Dr. William Bland www.abstractnonsense.com
Computer Programmer awksedgrep (Yahoo IM)
Any sufficiently advanced Emacs user is indistinguishable from magic

Cameron MacKinnon

2004-04-23, 2:36 pm

Duane Rettig wrote:
> But to practice a little one-upsmanship on your pedantry, that is not
> actually what PG said.
>
> Consider the four phrases:
>
> "I don't expect for myself to convince ..."
>
> "I don't expect us to convince ..."
>
> "I don't expect this article to convince ..."
>
> "I don't expect anyone to convince ..."

"I don't expect not to convince ..." - You can't just add words and say
it's a valid interpretation. English doesn't work that way.

Phrases one and three are, effectively, the same. To argue otherwise is
to suggest that PG had better arguments than those he presented in his
article, but he was saving them for some other time. In rhetoric, you
have to assume that your opponent is marshalling his best arguments.

> All four interpretations are equally correct, and might be applicable
> depending on the context. If you look at the article from which the
> statement came (http://www.paulgraham.com/avg.html) and especially the
> surrounding paragraph, it is clear that he intended the third meaning,
> not the first.

From the top of the article:

"April 2001, rev. April 2003

This article is derived from a talk given at the 2001 Franz Developer
Symposium."

So I'd say PG was preaching to the converted, or at least to those
willing to be swayed.

> However, even with that interpretation the statement
> is false, since many I have talked to have attested to coming to
> Common Lisp based on his books _and_ his articles, including this one,
> and I wouldn't doubt that some are over 25.

That he was wrong doesn't nullify his expectations. We have to assume
that his statement, about what he expected, was true at the time he made
it. Or do you think he was pulling our collective leg?

--
Cameron MacKinnon
Toronto, Canada

Duane Rettig

2004-04-23, 5:06 pm

Cameron MacKinnon <cmackin+nn@clearspot.net> writes:

> Duane Rettig wrote:
>
>
>
>
>
>
> "I don't expect not to convince ..." - You can't just add words and
> say it's a valid interpretation. English doesn't work that way.

Well, of course English doesn't work the way in which you've just
demonstrated, but it certainly does work by elision; a good portion
of communication skills is based on what is _not_ said.

The antecedent of the prepositional phrase "to convince..." is elided.
It is not the preposition at the beginning of the sentence, which is
in fact the subject for the transitive verb "expect". Note that expect
is missing its object, as well, and that object ends up serving as the
subject of the prepositional verb-phrase "to convince".

The reader/hearer of such elision is the one who supplies the elided
object/antecedent, which can be any word whose subjective form can serve
as the subject of the phrase "<subject> <tense> convince(s) anyone ...
to learn Lisp", and which can thus be any one of "myself (I)", "us (we)",
"this article", or "anyone" (or other objects, as well), but
which _cannot_ be "not", which is never a noun (unless enclosed
in double-quotes).

> Phrases one and three are, effectively, the same.

I disagree.

> To argue otherwise
> is to suggest that PG had better arguments than those he presented in
> his article, but he was saving them for some other time.

So argued. PG has written many articles at various times, and I'm sure
if you looked closely they would not all be completely consistent,
but even if they were, and he had said all that he had had to say,
then why would he write the other articles? If he writes on top of the
pinnacle of his own writing, then further writing would be redundant.

> In rhetoric, you have to assume that your opponent is marshalling
> his best arguments.

If you do that, you'll lose some of your arguments. There are a
number of reasons why a rhetorical opponent doesn't always marshal
his best arguments.

--
Duane Rettig duane@franz.com Franz Inc. http://www.franz.com/
555 12th St., Suite 1450 http://www.555citycenter.com/
Oakland, Ca. 94607 Phone: (510) 452-2000; Fax: (510) 452-0182

Kenny Tilton

2004-04-23, 7:48 pm

Duane Rettig wrote:
> Cameron MacKinnon <cmackin+nn@clearspot.net> writes:
>
>
>
>
> Pedantry is only effective (in annoying people) if it is correct, and
> you've gone and refuted your own pedantry.
>
> But to practice a little one-upsmanship on your pedantry, that is not
> actually what PG said.
>
> Consider the four phrases:
>
> "I don't expect for myself to convince ..."
>
> "I don't expect us to convince ..."
>
> "I don't expect this article to convince ..."
>
> "I don't expect anyone to convince ..."
>
> All four interpretations are equally correct, and might be applicable
> depending on the context. If you look at the article from which the
> statement came (http://www.paulgraham.com/avg.html) and especially the
> surrounding paragraph, it is clear that he intended the third meaning,
> not the first. However, even with that interpretation the statement
> is false, since many I have talked to have attested to coming to
> Common Lisp based on his books _and_ his articles, including this one,
> and I wouldn't doubt that some are over 25.

That does not make his statement false. He said he did not expect to
convince. That (non) expectation going unfulfilled does not change what
he expected.

Now /that/ pedantry should be annoying. :)

kt

--
Home? http://tilton-technology.com
Cells? http://www.common-lisp.net/project/cells/
Cello? http://www.common-lisp.net/project/cello/
Why Lisp? http://alu.cliki.net/RtL%20Highlight%20Film
Your Project Here! http://alu.cliki.net/Industry%20Application

Joe Marshall

2004-04-24, 2:34 am

Duane Rettig <duane@franz.com> writes:

> Pedantry is only effective (in annoying people) if it is correct...

Technically, it would be a stretch to call it `pedantry'.

I trust I have achieved the desired effect.

--
~jrm

Christian Lynbech

2004-04-25, 4:32 am

>>>>> "Cameron" == Cameron MacKinnon <cmackin+nn@clearspot.net> writes:

Cameron> Younger programmers would be more susceptible to expensive
Cameron> marketing campaigns by the usual vendors, and more focused on
Cameron> the skills listed in the employment ads.

I think that younger programmers are mostly susceptible to the current

trends more than advertising as such. It is not that advertising
is not able to influence the trends but it does not decide it.

The current craze with Java is based kostly on Javas ability to the
hall-of-PL-fame as the internet programming language, in part inspired
by clever marketing but just as much by being at the right place at
the right time.

When I was a young programmer (late 80's) the

language, at least
in my local corner of academia, was C. This was not based on marketing
but on the macho-attitude of Real Programmers which had a great appeal
on misguided wannabees such as myself.

On the other hand, a two-point dataset is not that convincing. I do
not know whether young programmers of the 70's were flocking to COBOL
due to the prospects in the work place? It may be that the C cult was
kind of a rebellion against the mainstream, a bit like the punk
movement within music. Certainly, programming in C is the equivalent
of shaving your head and punching needles through your flesh while
bickering about the black and futureless world.

------------------------+-----------------------------------------------------
Christian Lynbech | christian #\@ defun #\. dk
------------------------+-----------------------------------------------------
Hit the philistines three times over the head with the Elisp reference manual.
- petonic@hal.com (Michael A. Petonic)

Espen Vestre

2004-04-27, 3:01 am

Duane Rettig <duane@franz.com> writes:

> It seems as though this quip to what even I thought was a generally
> true statement has caused a flurry of anecdotal evidence to the
> contrary. Of course, it is only anecdotal and self-selecting, but
> I am surprised by the numbers and especially by individuals whom I
> would have thought had started Lisp much earlier in life. Very
> interesting.

Some people decide to stop learning already in their twenties (are
these the same people that stick to their high school friends and
think that the prom (or equivalent) was the summit of their lives?),
other people keep learning their whole life.

My 70 year old mother is not too bad, she is somewhat reluctant to
replacing OS 9 with OS X on her mac, but yesterday she let my 12 year
old daughter teach her how to operate her new cellular efficiently ;-)
--
(espen)

Chris Hall

2004-04-28, 8:36 am

Rayiner Hashem

2004-04-28, 1:51 pm

Cameron MacKinnon <cmackin+nn@clearspot.net> wrote in message news:<Lp-dnYDHnMhWqhTdRVn-jg@golden.net>...
> Duane Rettig wrote:
>
> To be pedantic, PG only said that *he* couldn't convince programmers of
> a certain age of the value of Lisp.
>
> That said, I myself am a counterexample, as his writings were part of my
> motivation.
>
> I'd be inclined to think that the over 25 market would be more likely to
> be receptive to radical new :-) technologies. They've worked long enough
> with the industry standards not to believe the hype, and are receptive
> to solutions to the problems they've come to understand.
>
> Younger programmers would be more susceptible to expensive marketing
> campaigns by the usual vendors, and more focused on the skills listed in
> the employment ads.
Absolutely. Especially given that companies like Microsoft and Sun do
their best to indoctrinate college students with Java and C# hype.

Gorbag

2004-04-29, 8:11 am

"William Bland" <news@abstractnonsense.com> wrote in message
news:pan.2004.04.23.17.40.48.908441@abstractnonsense.com...

> Perhaps I'm naive to think the successful projects will
> outweigh the "weird" choice of languages in any interviewer's mind?

In order to get by the HR keyword scans you have to list languages,
projects, etc. that they care about.

Speaking from personal experience...

If you are looking for a job where they don't care about the language you
used, you aren't looking for a programming job. From what I've seen,
companies hire "C++ programmers" and "Java programmers". They don't care
what they've done before for the most part, other than to demonstrate they
know about CMM, or other methodological ways to develop software in the
large.

The back door, as it were, is to get in based on other kinds of
requirements, e.g., technologies and non-programming skills. For instance,
if you have a demonstrated ability to get DARPA awards, nobody is going to
care if you use Lisp to execute the contract; they're hiring you for your
ability to bring money in to the business. Similarly if you understand
unsupervised learning techniques deeply, they will bring you in to architect
and teach their staff those techniques (assuming they already have bought,
or you can sell them, that it's relevant to their product or interest).

Companies need people to lead them into the future. Sell that, not "Lisp."

Sponsored Links