[python-advocacy] Forrester Research Survey Answers for Review

[Roy Smith]

My first comment is that this is an excellent document, and obviously  
the product of a large amount of effort.  Thank you for doing this!

Regarding F14, "This is reflected in its use of whitespace  
indentation for enhanced readability, and freeing the developer from  
repetition by avoiding type predeclarations"

I'm not sure where it fits, but you might want to mention  
retrospection.  I'm currently working with a C++ unit test framework  
which requires me to type the name of each test case FOUR times (once  
for the prototype, once the definition of the function implementing  
it, once for where it is printed out when the test runs, and once to  
register the test to run).   Because Python has introspection, a  
framework like unittest lets you type it just once.  The ratio of  
real code that does something useful to the total number of lines you  
have to type is several times higher in Python.

F66: History
      How mature is the language?

You might want to mention backwards compatibility, i.e. that  
something written in 1.5 (7 years ago now?) will run on any newer  
version of the interpreter.







On May 26, 2007, at 8:33 AM, Jeff Rush wrote:

> Whew! I've finished filling out the survey response to Forrester  
> Research and
> it was much harder than I expected.  I'll be delivering it to them  
> on Tuesday
> morning, so I'm making it available online to the community for  
> correction and
> final input.
>
>   http://dfwpython.org/uploads/Forrester/forrester-survey.txt
>
> At this stage I'm looking for blatant errors, embarrassing claims and
> overlooked significant features.  Some of the questions called for  
> an essay
> answer but I'm only permitted a few paragraphs and had to boil  
> matters down
> quite a big.
>
> Anyway, if anyone is around this holiday weekend, I'd appreciate a  
> quick skim,
> especially by anyone associated with Jython or IronPython, as I  
> might have
> misunderstood something and made claims that aren't precisely  
> accurate.  It
> goes into a spreadsheet and the cell numbers are next to each  
> response, so you
> can refer to those.
>
> I think afterward I may turn this into a whitepaper, with somewhat  
> expanded
> answers, to recycle the effort.
>
> -Jeff
>
> _______________________________________________
> Advocacy mailing list
> Advocacy at python.org
> http://mail.python.org/mailman/listinfo/advocacy


--
roy at panix.com