RE: Two points of "sale"

> From: Stephan Deibel
>
> From the recent emails, I felt the need to point out that there are
> basically two points at which you can sell Python:
>
> 1) At programming project startup.  As in Strakt's case, some managers
>    are open to considering Python; remember many of them know how to
>    program and have some appreciation that tools matter.  There's
> no magic
>    marketing bullet here, just visibility to people making these
>    decisions.  They might read about Strakt's story (see
>    http://pythonology.org/success&story=strakt) and try it.
>    They might have a friend that recommends Python or read about it
>    in a magazine.  All of these matter.
>
> 2) As a solution.  *Before* a problem is solved, people are open
>    to finding out about ways to solve it.  These people care about
>    solving their problem domain -- scientific visualization, managing
>    web content, running a factory, etc -- and they want to hear about
>    how others have done that.  The language is quite irrelevant.
>
> Selling a programming language only works in (1) and otherwise amounts to
> trying to sell a better way to solve an already-solved problem, which is
> not very useful, or simply is missing the mark as to what people are
> interested in.
>
> Note that (1) is fairly rare as opportunities go:  (2) has much more
> weight (the "killer app" concept).
>
> This might explain why the eGovOS people don't want to hear about
> language -- it's already enough of a leap that they want to hear about
> OS abstractly, the rest needs to be solution-oriented.
>
> - Stephan

This sounds like a good time to remind people of Dana Moore's presentation:
Subversion from Within

http://www.infoether.com/~rich/pycon2003.pdf

"Subversion from Within. Our goal in this paper and tutorial proposal is to
preach to the unconverted rather than reinforce what some of us already know
to be true - that lightweight languages and Python in particular are
transformational and illuminate a path for the future of application design
and delivery. We note with some concern that the emergence of the Java
programming language as the "COBOL of a new generation" often clouds the
thinking of system designers and developers. Too many designers and
developers simply cannot think in terms of lightweight languages as the
vehicle for delivering sophisticated and complete systems. This paper and
presentation suggests a "embrace and replace" strategy for co-existing with
languages such as Java with the eventual goal of replacing their
functionality altogether.
We conduct a survey of strategies for Python and Java co-existence with
discussion and practical demonstration.

We suggest that in the next generation internet, there is a crucial need for
Python scripting as the way to provide the "glue" for distributed
applications, to support fast and loose prototyping, and indeed to create
applications themselves. Many in our target audience may not consider
=scripting languages to be first class application vehicles, but with Java
objects as invocation targets for scripting languages and Java applications
as a launch platform for dynamic scripting, it is an exciting recipe for
building rapid prototypes and even next generation applications.

This paper and tutorial session will explain the how and why. In particular,
we focus on: Java and Python together (Jython) and discuss and demonstrate
distributed applications using Java and Python via JXTA and Jabber, turning
native applications into services, and creating conversational knowledgebots
with Python, ECMAScript and Java."