[Python-Dev] Re: Re: Call for defense of @decorators (original) (raw)

Tim Peters tim.peters at gmail.com
Sat Aug 7 19:33:44 CEST 2004

[Guido]

I still think it shouldn't be needed. Do we have to add 'currently' to every statement about the language? That doesn't make sense. The reference manual's title page already includes a version number. Shouldn't that be sufficient warning for those who want to interpret any part of the manual as a promise for all future?

Yes.

I really want to take a hard stance on this, because I believe the only reason this came up was that someone needed to find an argument against '@'.

At least two reasonably popular Python tools use @ heavily now, and their authors didn't appear to give a rip about decorators one way or the other. The use of @ for any purpose in the core would have elicited similar concern.

I don't think their argument would have a chance in court,

I believe they agree with that (partly because they both said so ).

so there's no reason to give in to them.

Courts are adversarial. You don't want an adversarial relationship with Python users -- there are lots of things to consider besides what a court would say.

Fight the trend to add silly disclaimers everywhere!

+1. OTOH, I'm also +1 on picking a character and promising (in the reference manual) that the language will never use it, to give authors of these kinds of tools a way to live peacefully with Python evolution. @ seems like a good choice for that.

More information about the Python-Dev mailing list