[python-advocacy] that guy that decides py or not to py?

[Tennessee Leeuwenburg]

Great thread everyone. I've been really pleased to read all the
contributions so far.

For me, I started using Python to develop something due to a fault in the
Java core APIs. I had heard about Python, and thought it might suit me. So I
started working, and solved my problem using the same algorithm as Java,
transliterated to Python, in about half a day. Win!

So, I kept using Python for other things. I fell in love with semantic
whitespace, list comprehensions, the interpreter and everything that makes
Python great. Clean, meaningful code.

Now, I work for the Australian Bureau of Meteorology on the Graphical
Forecast Editor, which is a NOAA product written in Python with some C/C++
deep down. I deliberately sought out this job because it was in Python (not
to mention being a promotion and more interesting work). In this case, the
application came first.

One reason that Ruby may have been so popular in the Enterprise is Ruby on
Rails. Yes, I know, it's just a web platform, and Python has some too. But
consider this -- a project manager has at his disposal either a Ruby or
Python programmer, and the need for some web application. To choose Ruby, he
only needs to choose one thing (RoR, the obvious choice). To choose Python,
he needs to choose two things -- Python, and then the web framework he will
use.

To my eyes, as a professional programmer using Python exclusively, working
in the domain of desktop applications with no web interface or interaction,
it is *difficult* to choose a Python web framework of any sort. If I have to
do any heavy lifting, I will most likely revert back to Java because 3 years
ago I knew how to do web stuff with servlets.

The python web apps are to my mind very unusally oriented. I need to know
about Python, Javascript, browser incompatibilities, CSS and a whole list of
things. I need to be really, really expert in order to get a Python web
application working well. In practise, I have never been able to just "do
what I want" using a Python web framework.

Where am I going with this? Rather than continuing to argue for the point,
let's just assume I'm right and examine that assumption later.

Ruby can get into organisation easily, because it has a single,
well-advertised preferred solution which developers can learn. It is stable.
The language isn't the important choice here, the web framework is the
important choice -- at least initially. But if you look at the PSF website,
there is no shiny web framework to be picked up. Even the various tutorials
on the framework websites (thinking of Django, TurboGears) only show how to
(essentially) press 'go' on a pre-packaged module. They don't show you how
to get into the guts of the thing and really make it do something you want.

There are no good books on doing new application development using the web
as a front-end. Shopping carts and Wikis? So yesterday. While I may be
arguing, perhaps, against my own inexpertise, I don't think this is the
whole story. There will be others in my situation -- new to Python or new to
web programming -- who find the body of knowledge too large to assimilate,
yet not find any acceptable solutions in what they do choose to explore.

This is a barrier on two fronts -- Management has too much to choose from,
and developers have too much to learn.

Sorry if this has turned into a rant. I will leave it there, but if anyone
is interested, I have a few more points to make.

Cheers,
-T

On 3/15/07, Mark Ramm <mark.mchristensen at gmail.com> wrote:
>
> On 3/13/07, Carl Karsten <carl at personnelware.com> wrote:
> > I took on python as the result of a personal choice.   I am sure there
> is a huge
> > population of people in that boat.
>
> And the recent success of Ruby on Rail's anti-enterprise marketing
> campaign proves that it is often the influence of these people who
> bring a language into the enterprise.
>
> So paradocically Ruby is being used in  large companies more now than
> a year ago because David Hanson made fun of the Java Enterprise
> Architect bloggers a lot.
>
> Ruby is gaining ground in companies because of individual converts
> marketing it to their bosses, not the other way round.
>
> Providing tools for people to sell to their Corporate Managers is a
> usefull enterprise, but getting people that are willing to invest time
> and effort to make that sale one-on-one is, in my opinion more
> productive.
>
> We need to focus on the people who choose for themselves, and who know
> enough about different languages to have their choices be respected by
> those around them.
>
> > I think there is a bigger boat: the groups that are lead by a 'manager'
> (pointy
> > hair or otherwise) that is a single person directing a group: "We are
> going to
> > use language X."
> >
> > That person's choice of X will cascade - his group will learn about X,
> and each
> > member be able to speak about X, and recommend X to other leaders, and
> the told
> > two friends...
>
> The cascade only happens if people fall in love with python, and very
> few people fall in love with a technology forced upon them by the
> Pointy Hared Boss.
>
> In my opinion "Marketing Python" is most effective when we convince
> respected, visible, interesting people to pick python to do cool
> things. Oh, and it doesn't hurt if they publicize those cool things
> too.
>
> --
> Mark Ramm-Christensen
> email: mark at compoundthinking dot com
> blog: www.compoundthinking.com/blog
> _______________________________________________
> Advocacy mailing list
> Advocacy at python.org
> http://mail.python.org/mailman/listinfo/advocacy
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/advocacy/attachments/20070321/e25fccd2/attachment.html