On Thu, Apr 7, 2016 at 11:08 PM, Ethan Furman <ethan@stoneleaf.us> wrote:

On 04/07/2016 07:17 PM, Alexander Belopolsky wrote:

...

You greatly expanded the scope of the discussion. The original thread was focused on a single feature, and a not so widely used one. You asked 6-7 (rhetorical?) questions below that have nothing to do with the original topic.

Thank you, Alexander. I appreciate the feedback.

It's entirely possible I'm talking to myself, but in case anyone is listening:

Guido, it felt to me that you were being a jerk. I was definitely an ass in response. For that I apologize.

Thanks for that. I didn't mean to be a jerk, just to prevent the thread from derailing way out of scope. And I was on my cell phone, where I can't be as eloquent as when I have a real keyboard. So I apologize for sounding like a jerk.

As for broadening the scope of the discussion I will say this:

So far I have authored or helped with three successful PEPs; The first was to amend an incomplete feature (raise from None), and the third was to add back a feature that had been ripped out (%-interpolation with bytes).

Why were those two even necessary? I suspect because the folks involved were only thinking about their own needs and/or didn't have the relevant experience as to why those features were useful.

Perhaps I am only flattering myself, but I think I try hard to see all sides of every issue, and the only way I can do that is by asking questions of those with more or different experience than I have.

I think the current pathlib discussions are a fair indicator: I don't particularly care for it, and might never use it -- but I hate to see it tossed and Antoine's work and effort lost; so I'm working to find a reasonable way to keep it, and not just in asking questions and offering ideas -- I volunteered my time to write the code.

I'm starting to ramble so let me close with this: I'm not sorry for asking questions and trying to look at the broader issues, and I'm not going to stop doing that -- but I will stop pursuing any particular issue when asked to do so... but please don't be insulting about it.

Starting a new thread with a broader (or different) scope is always totally fine. Bringing up a whole bunch of contentious things that distract from a relatively simple issue is not.Thanks for asking for feedback! -- --Guido van Rossum (python.org/~guido)

On Apr 8, 2016 3:09 AM, "Greg Ewing" <greg.ewing@canterbury.ac.nz> wrote:

Ethan Furman wrote:

...

Can anybody please enlighten me as to what, exactly, I did wrong?

I think Guido was objecting to talk about things such as making bool no longer subclass int, which is not only going a long way beyond the original proposal, but is almost certainly never going to happen, so any discussion of it could be seen as wasting people's time.

While reading the thread, I was honestly wondering about whether it was Pythonic to have bool be a subclass of int. (It's definitely convenient.) So that, at least, might be a justified "Should we also...".

On Thu, Apr 7, 2016 at 11:08 PM, Ethan Furman <ethan@stoneleaf.us> wrote:

On 04/07/2016 07:17 PM, Alexander Belopolsky wrote:

...

You greatly expanded the scope of the discussion. The original thread was focused on a single feature, and a not so widely used one. You asked 6-7 (rhetorical?) questions below that have nothing to do with the original topic.

Thank you, Alexander. I appreciate the feedback.

It's entirely possible I'm talking to myself, but in case anyone is listening:

Guido, it felt to me that you were being a jerk. I was definitely an ass in response. For that I apologize.

Thanks for that. I didn't mean to be a jerk, just to prevent the thread from derailing way out of scope. And I was on my cell phone, where I can't be as eloquent as when I have a real keyboard. So I apologize for sounding like a jerk.

As for broadening the scope of the discussion I will say this:

So far I have authored or helped with three successful PEPs; The first was to amend an incomplete feature (raise from None), and the third was to add back a feature that had been ripped out (%-interpolation with bytes).

Why were those two even necessary? I suspect because the folks involved were only thinking about their own needs and/or didn't have the relevant experience as to why those features were useful.

Perhaps I am only flattering myself, but I think I try hard to see all sides of every issue, and the only way I can do that is by asking questions of those with more or different experience than I have.

I think the current pathlib discussions are a fair indicator: I don't particularly care for it, and might never use it -- but I hate to see it tossed and Antoine's work and effort lost; so I'm working to find a reasonable way to keep it, and not just in asking questions and offering ideas -- I volunteered my time to write the code.

I'm starting to ramble so let me close with this: I'm not sorry for asking questions and trying to look at the broader issues, and I'm not going to stop doing that -- but I will stop pursuing any particular issue when asked to do so... but please don't be insulting about it.

Starting a new thread with a broader (or different) scope is always totally fine. Bringing up a whole bunch of contentious things that distract from a relatively simple issue is not.Thanks for asking for feedback! -- --Guido van Rossum (python.org/~guido)

On Apr 8, 2016 3:09 AM, "Greg Ewing" <greg.ewing@canterbury.ac.nz> wrote:

Ethan Furman wrote:

...

Can anybody please enlighten me as to what, exactly, I did wrong?

I think Guido was objecting to talk about things such as making bool no longer subclass int, which is not only going a long way beyond the original proposal, but is almost certainly never going to happen, so any discussion of it could be seen as wasting people's time.

While reading the thread, I was honestly wondering about whether it was Pythonic to have bool be a subclass of int. (It's definitely convenient.) So that, at least, might be a justified "Should we also...".