[Python-Dev] Draft proposal: Implicit self in Python 3.0 (original) (raw)

Ian Bicking ianb at colorstudy.com
Sun Jan 8 21:43:17 CET 2006

Tim Peters wrote:

[Thomas Wouters]

My point isn't that it isn't archived somewhere (mailinglists, wiki, FAQ, the minds of many, many people, not just Python developers) but that it isn't easily findable and it isn't easily accessible in a single location. I thought PEP's where supposed to be that, and if I have a particular idea for new syntax or new semantics, PEPs would be the place I'd look, not the FAQ or a Wiki. Luckily, in his benevolent infinite wisdom, I expect Guido reserved PEP number 13 for exactly this purpose: for a meta-PEP to record the unlucky PEP ideas that are so unlikely to get accepted that it's not worth anyone's time to write an actual PEP for them. I like the title: Don't Bother: PEPs Rejected Before Being Written No, I'm not kidding. At least I don't think I am.

+1. I think it's rather patronizing to send things back to python-list when you know people are wasting their time discussing them because they will never be accepted. People on python-list have useful things to do too, they don't just read to waste their time.

I would prefer PEP form over a wiki page, as I'd rather this be truly authoritative, and only Guido can really completely reject an idea. Justifying the rejections can be done by anyone though; maybe each idea could link to a wiki page on the topic.

I think it's also important to be clear on what's being rejected. Often these rejected ideas address a real issue, and if you think about the issue from another angle you might be able to solve that without falling into the trap that the oft-rejected proposal fell into. But it's easy to confuse that the issue or use case is being explicitly ignored, rather than the particulars. For instance, Thomas said "changing all statements into expressions (e.g. logix): python isn't (going to be) a functional language" -- and that's what shouldn't be in the PEP. All statements aren't going to be expressions; the editorialization that Python isn't going to be a functional language is both rather inaccurate, misses the real reason for statements, and needlessly alienates people who like functional programming (and they have been needlessly alienated by discussions about lambda and filter).

So... maybe Guido or python-dev should write/vet the justifications too. When you are putting up walls and specifically discouraging community participation on certain issues, it should be done in a sensitive way.

-- Ian Bicking | ianb at colorstudy.com | http://blog.ianbicking.org

More information about the Python-Dev mailing list