Top 10 Language Constructs (Phyton)

[Tim Rowe]

In article <8kmtab$4or$1 at pollux.ip-plus.net>, GustavsB at ch.sibt.com (Bruno 
Gustavs) wrote:

> "Dinu C. Gherman" <gherman at darwin.in-berlin.de> wrote in message
> news:396EE7BC.DA2FA7B3 at darwin.in-berlin.de...
> > What do you mean by "language constructs", Python keywords
> > like "for" or "class"? Are you expecting some statistics of
> > used keywords over one's own code?
> >
> > I doubt this will be very useful, like the answer to the
> > question: "Which are the ten most used words in English?".
> > But maybe I'm not understanding well how you think you'll
> > benefit from such an answer...
> 
> Lets assume you're going to design Phyton2. What would this
> language contain? What should Phyton2 support? What would you
> like to solve with Phyton2?

People's "top 10 constructs" would be a disastrous basis for such a 
language. I'm trying to remember the name of a language I used many years 
ago that was designed by government committee. It had about 10 ways of 
doing some obscure things (because everybody on the committee had 
contributed their favourite solution) but no ways of doing quite ordinary 
things (because everyone on the committee used those things but none of 
them noticed them). The result was like a car with three different 
air-conditioning units and no engine.

/Much/ better IMHO is to see what application domains Python is 
particularly strong in, and to try to find out what distinguishes Python 
in those domains. Then look for ways it can be improved in those specific 
domains and expanded into closely related domains. At the moment there are 
things I use Python for and things I use Ada for. If the next version of 
Python tried to muscle in on the mainstream Ada territory it would almost 
certainly find that Ada still did those things better but the attempted 
shift had broken it's advantage in those things it was good at before.