He is the author of several popular projects including the web framework "Flask". This makes him a person very respected in the Python community - personally I love his taste for interface design.
I wish he could interact better with the core team, because some of his rants are not as constructive as they might be.
Whoa... Armin has never, to my recollection, been malicious in his criticisms of Python 3. He simply disagrees with some of the core team's decisions. And he's far from the only one.
It's not Armin's duty to interact in any way with the core team. Hell, the core team should be working to please Armin if you ask me. He's your target user, and he's disappointed with your product, it's not his fault.
I think Armin does a wonderful job providing a voice for those of us who are increasingly disenchanted with Python 3.
Simplicity is a core value of the Python community; this kind of change goes through a long discussion before getting implemented, so I wonder how we ended up with such convoluted design (the reasoning is documented in the PEP system but most of it is beyond my wizardry level).
Perhaps this is me showing some cultural bias (I'm Brazilian), but in my social context it is considered rude to make this kind of criticism in public unless you tried your best to address the problem in a more restricted exchange. The more authority you have, more you are expected to come with a better proposal instead of just pointing out the mess.
asyncio is not my construction site and I do not intend to make it mine. I am not an expert in anything asynchronous and as such will not try to make myself that person.
This entire adventure started because I was looking for a way to get logical call contexts working in it and this shows that there is not enough internal machinery to support that at the moment.
I'm not going to fight for that however. I hope my post was rude.
Thanks for taking the time to write out your concrete experiences and qualitative impressions in such detail. It's really a tremendous service and I'm sure it'll be well received.
For what it's worth I would like to do a PEP on accessing event loops from coroutines and logical call contexts but having read some of the conversations that already took place to improve asyncio I am less thrilled about being involved.
I suggest to proceed with the PEP if you have a good idea about the API. Get it out there, it's not required to spend hours on mailing lists - other people can do that if they like it.
> unless you tried your best to address the problem in a more restricted exchange.
That sounds like orders of magnitude more work. In case you are not interested in performing that work, it seems more constructive to me to share your thoughts rather than to say nothing.
He is the author of several popular projects including the web framework "Flask". This makes him a person very respected in the Python community - personally I love his taste for interface design.
I wish he could interact better with the core team, because some of his rants are not as constructive as they might be.