[transcript of a talk Ritchie gave on a paper on the development of C]
Key points and Ritchie's notes on them (and me making obvious reflections to Python; I'll be updating these, just need to get 'em written down first):
- Neither elegance nor formality is enough
"It's important that people really be able to understand the the language."Python has been hailed as a really easy to learn language and learnability is indeed one of it's strong points. But Python's not just a teaching language, far from it, it holds also a lot more, too (witness list comprehensions, functionalshy features, etc.).
- Connection with what people need and can get
- Availability: (compiler techonology, implementation, distribution)
- Appropriate interaction with environment (usability as tool, adaptability to unexpected uses, ways of dealing with extra-linguistic issues)
- Ability to age gracefully
- It's best to get it mostly right the first time
- Be lucky
There's still no commitee for Python and there's active development. Significant changes are getting more difficult, though.
GvR and Tim Peters could elaborate on this, but I think they're not too keen to do it. Rather, check out Kuchling's gatherings on Python's warts. [oh, it's not up, luckily Google's got cache: cached copy of Python warts]
This might need some elaboration.