Newsgroups: comp.lang.dylan
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!news.moneng.mei.com!news.ecn.bgu.edu!psuvax1!news.pop.psu.edu!news.cac.psu.edu!newsserver.jvnc.net!nntpserver.pppl.gov!princeton!phoenix.princeton.edu!mfterman
From: mfterman@phoenix.princeton.edu (Dark Mutant)
Subject: Re: Yes, but how much will it cost?
Message-ID: <1995Mar31.193221.24061@Princeton.EDU>
Originator: news@hedgehog.Princeton.EDU
Sender: news@Princeton.EDU (USENET News System)
Nntp-Posting-Host: phoenix.princeton.edu
Reply-To: mfterman@phoenix.princeton.edu (Dark Mutant)
Organization: Mutant for Hire, Inc.
References: <gwynne-2403951245370001@tchm04a14.rmt.utk.edu> <davidc-2703951108200001@macq106.psy.uwa.oz.au> <1995Mar27.185511.19398@VFL.Paramax.COM> <tad1-3103951110330001@tad1.cit.cornell.edu>
Date: Fri, 31 Mar 1995 19:32:21 GMT
Lines: 49

In article <tad1-3103951110330001@tad1.cit.cornell.edu>, tad1@cornell.edu (Tom Dimock) writes:
>Not necessarily true.  If a language offers enough advantages, I'llk
>cheerfully undertake the training needed.  I have done this at least once,
>and I'm certain we'll do it again.  If by spending several months
>re-training people to use a new language I can cut half a year off the
>development time (and I think Dylan may be able to do that) then the lack
>of trained Dylan programmers is a non-issue for me.

The problem is that you are an individual case. Most people I see in
the business world don't tend to go for such abstract things. They
look at "well, what can it do that C++ can't, and where can we get
people who are experts in it." And if you claim that you can develop
faster and more reliable code using Dylan, people will want to see
studies proving it, preferably backed up by lots of real world
experience.

>It reminds me of a cartoon I used to have on the wall of my office.  A
>medieval king was brushing off a machine gun salesman, saying "I'm sorry,
>but I don't have time to look at this.  I've got a war to fight!"  Any
>manager who will not consider a new tool just because it might require
>training is a total loser in my opinion.  If he uses "good business sense"
>and develops in C++, and I beat him to market by re-training for Dylan and
>speeding up my development, who wins?  Who had the better "business
>sense"?

Not a good analogy. The thing is, that C++ offers an upgrade path from
C. At one business I know of, they seriously considered shifting all
their support/installation scripts from ksh to Perl, which had all these
nice features. Didn't happen because they lacked the time to rewrite
all their code from the ground up in Perl. Likewise, in order to
persuade a company to switch from one programming language to another,
you need to convince them that the language offers big advantages and
has a viable future in front of it. Not only things it can do better,
but things you simply can't do in the old system.

If Apple wants Dylan to take off, they need to encourage people to write
code in it, lots of code in it. There should be an advantage to writing
in Dylan on the Mac, on OS/2, on other operating systems. (high levels
of support for SOM, DSOM and Opendoc are a necessity) Come up with a
system that allows for fast prototyping, fast development, easy support
and upgrading. Companies want systems that allow them to get code out
the door faster and more reliably.


-- 
 Martin Terman, Mutant for Hire, Synchronicity Daemon, Priest of Shub-Internet
 Disclaimer: Sticks and stones may break my bones, but flames are just ignored
  mfterman@gnu.ai.mit.edu  terman@pupgg.princeton.edu an174376@anon.penet.fi
"Sig quotes are like bumper stickers, only without the same sense of relevance"
