I find email list VASTLY easier to deal with than any newfangled web-based custom discussion forum. Part of that is that it is a uniform interface to every list I belong too, and I can choose my own MUA. With all those web things, every site works a little bit different from every other one, that imposes an unnecessary cognitive burden (and usually simply lacks some desired capability) On Aug 13, 2016 11:39 AM, "Donald Stufft" <donald@stufft.io> wrote:

...

On Aug 13, 2016, at 2:21 PM, Oleg Broytman <phd@phdru.name> wrote:

...

As Donald pointed out, there are people who are not going to create custom email processing toolchains.

In what way they will be helpful to the development of Python? Contributors have to install, learn, configure and use a lot of development tools, comparing to which email tools are just toys.

Well, I personally generally do not have the time to sit there and craft some sort of complex email toolchain to deal with it. I just leave lists if their volume are too high for me to deal with or they don’t provide me the tools to interact with them in a non frustrating way. For example, I’m no longer subscribed to python-dev because of these reasons. I *think* I’ve positively impacted the development of Python, but maybe not! In any case I think that falling down the trap of thinking that anyone who is willing to contribute to Python is also willing to maintain a personal toolchain for dealing with the deficiencies in mailing lists is not a place we should be in.

That’s not to say that a traditional mailing list may not represent the best trade off— I have my opinion, you have yours, but we should not start thinking that adding an obstacle course that a user must complete before they can meaningfully contribute is doing anything but self selecting for people willing to run that particular obstacle course, not selecting for skill or likely impact.

— Donald Stufft

_______________________________________________ Python-ideas mailing list Python-ideas@python.org https://mail.python.org/mailman/listinfo/python-ideas Code of Conduct: http://python.org/psf/codeofconduct/

On 13 August 2016 at 19:44, David Mertz <mertz@gnosis.cx> wrote:

I find email list VASTLY easier to deal with than any newfangled web-based custom discussion forum. Part of that is that it is a uniform interface to every list I belong too, and I can choose my own MUA. With all those web things, every site works a little bit different from every other one, that imposes an unnecessary cognitive burden (and usually simply lacks some desired capability)

For me, the important point is "uniform interface to all lists". I like email for the forums I participate in, simply because I only need *one* browser tab open (gmail) and I don't have to remember or bookmark a variety of URLs. On forums that have their own web interface, I participate much less frequently, and tend to fall into much more of a "drive by" interaction, only contributing to "my" threads, rather than fully participating like I do on mailing lists. Whereas with my email lists, I read pretty much everything (sometimes only skimming, of course), which leads to me participating in threads I would otherwise have ignored. I can't really offer any opinion on the "mailing lists are more efficient" debate, as I simply dump all my lists into gmail, with a label per list, so I'm not exactly a power user, but the "single website for everything" aspect is the huge bonus for me. Would I follow python-ideas if it moved to a different forum? Certainly, if the new forum let me just click on something and from there on interact solely by email. Maybe, if I found that having a python-ideas tab permanently open was worthwhile. Otherwise, I don't know. Likely not, except on specific topics (but without an email feed, I don't know how I'd find out about such topics). And my participation would be much less frequent. (I leave it to others to judge whether that would be a good thing ;-)) In my opinions, forums tend to encourage a much more focused style of discussion. In one way, that's a good thing (and I'm sure many people would prefer python-ideas to have more focus). But it *also* tends to deter people from contributing - I can't quite express why, but there's somehow less of a sense of being an open debate with a forum. Maybe that's just me - it's certainly a subjective thing - but in a forum I'd expect the quality of discussion to increase, but the quantity (and breadth) to decrease. While I'm mentioning random thoughts, email replies to a forum like a github tracker are often a little disruptive, because etiquette is different. Email users tend to quote extensively for context, and often include signatures. Neither of these things is typically as necessary on a tracker (minimal, careful quoting is frequent, but not the extensive quoting common on mailing lists). So I could imagine a "mixed" interface actually being *less* comfortable for both types of participant. Paul

On Sat, 13 Aug 2016 at 11:39 Donald Stufft <donald@stufft.io> wrote:

...

On Aug 13, 2016, at 2:21 PM, Oleg Broytman <phd@phdru.name> wrote:

...

As Donald pointed out, there are people who are not going to create custom email processing toolchains.

In what way they will be helpful to the development of Python? Contributors have to install, learn, configure and use a lot of development tools, comparing to which email tools are just toys.

Well, I personally generally do not have the time to sit there and craft some sort of complex email toolchain to deal with it. I just leave lists if their volume are too high for me to deal with or they don’t provide me the tools to interact with them in a non frustrating way. For example, I’m no longer subscribed to python-dev because of these reasons. I *think* I’ve positively impacted the development of Python, but maybe not! In any case I think that falling down the trap of thinking that anyone who is willing to contribute to Python is also willing to maintain a personal toolchain for dealing with the deficiencies in mailing lists is not a place we should be in.

That’s not to say that a traditional mailing list may not represent the best trade off— I have my opinion, you have yours, but we should not start thinking that adding an obstacle course that a user must complete before they can meaningfully contribute is doing anything but self selecting for people willing to run that particular obstacle course, not selecting for skill or likely impact.

To give another example of how mailing lists put people at the mercy of how much time and expertise they have with their MUA, python-dev received https://mail.python.org/pipermail/python-dev/2016-August/145815.html today. Now in the archives that email looks fine, but the two opening paragraphs of that email came formatted a bit differently into my inbox (which is Google Inbox): ``` Hello, We are experimenting with a tool for inspecting how well languages and libraries support server certificate verification when establishing TLS connections. We are getting rather confusing results in our first major shootout of bundled CPython 2 and 3 versions in major, still supported OS distributions. We would love to get any insight into the test stubs and results. Maybe we are doing something horribly wrong? ``` That's a mess and the whole email is formatted like that. I actually have not read the email because of the formatting issue. As Oleg pointed out, when you go with a federated solution like mail, you are the mercy of whatever tools people choose to use with the service. But when you use a centralized approach you know the experience is consistent for everyone and thus there's a certain level of quality control. Now some say we don't need to include more people in discussions and we should let the power users continue to use their powerful workflows, while others say we should make it easier for all to participate. For me this community is known for being welcoming and I want to foster that, and if that means some of us have to use slightly less powerful tools to manage conversations so that everyone gets a better experience overall then I say that's worthy tradeoff.

On Sat, Aug 13, 2016, 16:26 Chris Angelico <rosuav@gmail.com> wrote:

On Sun, Aug 14, 2016 at 9:17 AM, Brett Cannon <brett@python.org> wrote:

...

That's a mess and the whole email is formatted like that. I actually have not read the email because of the formatting issue. As Oleg pointed out, when you go with a federated solution like mail, you are the mercy of whatever tools people choose to use with the service. But when you use a centralized approach you know the experience is consistent for everyone and thus there's a certain level of quality control.

IMO that's an argument in favour of the federated approach.

I don't understand how me receiving a badly formatted email due to some disagreement between the sender's and my email client is a point of support? -Brett

ChrisA _______________________________________________ Python-ideas mailing list Python-ideas@python.org https://mail.python.org/mailman/listinfo/python-ideas Code of Conduct: http://python.org/psf/codeofconduct/

The world probably wouldn't better, but I think python-ideas will be better on something like GitHub. People who like it will benefit from it, people who don't will still be able to use their email setup (as pointed earlier - at least for the majority of cases). Regards, Eugene On Sun, Aug 14, 2016 at 12:08 PM, Chris Angelico <rosuav@gmail.com> wrote:

On Sun, Aug 14, 2016 at 2:57 PM, Brett Cannon <brett@python.org> wrote:

...

On Sat, Aug 13, 2016, 16:26 Chris Angelico <rosuav@gmail.com> wrote:

...

On Sun, Aug 14, 2016 at 9:17 AM, Brett Cannon <brett@python.org> wrote:

...

That's a mess and the whole email is formatted like that. I actually have not read the email because of the formatting issue. As Oleg pointed

out,

...

...

when you go with a federated solution like mail, you are the mercy of whatever tools people choose to use with the service. But when you use a centralized approach you know the experience is consistent for everyone and thus there's a certain level of quality control.

IMO that's an argument in favour of the federated approach.

I don't understand how me receiving a badly formatted email due to some disagreement between the sender's and my email client is a point of support?

That in itself isn't. But your next point is that email lets people choose what they use, whereas centralized systems force everyone to use the exact same client (or whatever clients the one central authority provides - eg Slack offers web and desktop, and I think mobile). In fact, the entire *point* of the centralized systems is to force everyone to use a restricted set of clients, instead of having the freedom to choose. Would the world be a better place if everyone were forced to write all code in Python?

ChrisA _______________________________________________ Python-ideas mailing list Python-ideas@python.org https://mail.python.org/mailman/listinfo/python-ideas Code of Conduct: http://python.org/psf/codeofconduct/

On Sun, Aug 14, 2016 at 03:57:44AM -0500, C Anthony Risinger <anthony@xtfx.me> wrote:

Mobile dominates my non-work net-time today. I don't want to get out a laptop to respond pretty.

In what ways are you going to contribute without getting out to your laptop?

Email is almost by design static and unable to change. Text. Walls of it!

We are talking about Python development, and the development is performed with texts. Python is written in C, Python code is written... well, in Python, documentation is written in reStructuredText. So discussions about all of this is, naturally, textual. Oleg. -- Oleg Broytman http://phdru.name/ phd@phdru.name Programmers don't die, they just GOSUB without RETURN.

A uniform interface works well enough for issue trackers. And the "freedom of choice" idea doesn't overrule all other concerns. Maybe we should just start a python-ideas tracker and see who comes. There's no reason it couldn't exist in addition to the mailing list. (Before you scream fragmentation, we have many lists already.) --Guido (mobile)

I find email list VASTLY easier to deal with than any newfangled web-based custom discussion forum. Part of that is that it is a uniform interface to every list I belong too, and I can choose my own MUA. With all those web things, every site works a little bit different from every other one, that imposes an unnecessary cognitive burden (and usually simply lacks some desired capability) On Aug 13, 2016 11:39 AM, "Donald Stufft" <donald@stufft.io> wrote:

...

On Aug 13, 2016, at 2:21 PM, Oleg Broytman <phd@phdru.name> wrote:

...

As Donald pointed out, there are people who are not going to create custom email processing toolchains.

In what way they will be helpful to the development of Python? Contributors have to install, learn, configure and use a lot of development tools, comparing to which email tools are just toys.

Well, I personally generally do not have the time to sit there and craft some sort of complex email toolchain to deal with it. I just leave lists if their volume are too high for me to deal with or they don’t provide me the tools to interact with them in a non frustrating way. For example, I’m no longer subscribed to python-dev because of these reasons. I *think* I’ve positively impacted the development of Python, but maybe not! In any case I think that falling down the trap of thinking that anyone who is willing to contribute to Python is also willing to maintain a personal toolchain for dealing with the deficiencies in mailing lists is not a place we should be in.

That’s not to say that a traditional mailing list may not represent the best trade off— I have my opinion, you have yours, but we should not start thinking that adding an obstacle course that a user must complete before they can meaningfully contribute is doing anything but self selecting for people willing to run that particular obstacle course, not selecting for skill or likely impact.

— Donald Stufft

_______________________________________________ Python-ideas mailing list Python-ideas@python.org https://mail.python.org/mailman/listinfo/python-ideas Code of Conduct: http://python.org/psf/codeofconduct/

On 13 August 2016 at 19:44, David Mertz <mertz@gnosis.cx> wrote:

I find email list VASTLY easier to deal with than any newfangled web-based custom discussion forum. Part of that is that it is a uniform interface to every list I belong too, and I can choose my own MUA. With all those web things, every site works a little bit different from every other one, that imposes an unnecessary cognitive burden (and usually simply lacks some desired capability)

For me, the important point is "uniform interface to all lists". I like email for the forums I participate in, simply because I only need *one* browser tab open (gmail) and I don't have to remember or bookmark a variety of URLs. On forums that have their own web interface, I participate much less frequently, and tend to fall into much more of a "drive by" interaction, only contributing to "my" threads, rather than fully participating like I do on mailing lists. Whereas with my email lists, I read pretty much everything (sometimes only skimming, of course), which leads to me participating in threads I would otherwise have ignored. I can't really offer any opinion on the "mailing lists are more efficient" debate, as I simply dump all my lists into gmail, with a label per list, so I'm not exactly a power user, but the "single website for everything" aspect is the huge bonus for me. Would I follow python-ideas if it moved to a different forum? Certainly, if the new forum let me just click on something and from there on interact solely by email. Maybe, if I found that having a python-ideas tab permanently open was worthwhile. Otherwise, I don't know. Likely not, except on specific topics (but without an email feed, I don't know how I'd find out about such topics). And my participation would be much less frequent. (I leave it to others to judge whether that would be a good thing ;-)) In my opinions, forums tend to encourage a much more focused style of discussion. In one way, that's a good thing (and I'm sure many people would prefer python-ideas to have more focus). But it *also* tends to deter people from contributing - I can't quite express why, but there's somehow less of a sense of being an open debate with a forum. Maybe that's just me - it's certainly a subjective thing - but in a forum I'd expect the quality of discussion to increase, but the quantity (and breadth) to decrease. While I'm mentioning random thoughts, email replies to a forum like a github tracker are often a little disruptive, because etiquette is different. Email users tend to quote extensively for context, and often include signatures. Neither of these things is typically as necessary on a tracker (minimal, careful quoting is frequent, but not the extensive quoting common on mailing lists). So I could imagine a "mixed" interface actually being *less* comfortable for both types of participant. Paul

On Sat, 13 Aug 2016 at 11:39 Donald Stufft <donald@stufft.io> wrote:

...

On Aug 13, 2016, at 2:21 PM, Oleg Broytman <phd@phdru.name> wrote:

...

As Donald pointed out, there are people who are not going to create custom email processing toolchains.

In what way they will be helpful to the development of Python? Contributors have to install, learn, configure and use a lot of development tools, comparing to which email tools are just toys.

Well, I personally generally do not have the time to sit there and craft some sort of complex email toolchain to deal with it. I just leave lists if their volume are too high for me to deal with or they don’t provide me the tools to interact with them in a non frustrating way. For example, I’m no longer subscribed to python-dev because of these reasons. I *think* I’ve positively impacted the development of Python, but maybe not! In any case I think that falling down the trap of thinking that anyone who is willing to contribute to Python is also willing to maintain a personal toolchain for dealing with the deficiencies in mailing lists is not a place we should be in.

That’s not to say that a traditional mailing list may not represent the best trade off— I have my opinion, you have yours, but we should not start thinking that adding an obstacle course that a user must complete before they can meaningfully contribute is doing anything but self selecting for people willing to run that particular obstacle course, not selecting for skill or likely impact.

To give another example of how mailing lists put people at the mercy of how much time and expertise they have with their MUA, python-dev received https://mail.python.org/pipermail/python-dev/2016-August/145815.html today. Now in the archives that email looks fine, but the two opening paragraphs of that email came formatted a bit differently into my inbox (which is Google Inbox): ``` Hello, We are experimenting with a tool for inspecting how well languages and libraries support server certificate verification when establishing TLS connections. We are getting rather confusing results in our first major shootout of bundled CPython 2 and 3 versions in major, still supported OS distributions. We would love to get any insight into the test stubs and results. Maybe we are doing something horribly wrong? ``` That's a mess and the whole email is formatted like that. I actually have not read the email because of the formatting issue. As Oleg pointed out, when you go with a federated solution like mail, you are the mercy of whatever tools people choose to use with the service. But when you use a centralized approach you know the experience is consistent for everyone and thus there's a certain level of quality control. Now some say we don't need to include more people in discussions and we should let the power users continue to use their powerful workflows, while others say we should make it easier for all to participate. For me this community is known for being welcoming and I want to foster that, and if that means some of us have to use slightly less powerful tools to manage conversations so that everyone gets a better experience overall then I say that's worthy tradeoff.

On Sat, Aug 13, 2016, 16:26 Chris Angelico <rosuav@gmail.com> wrote:

On Sun, Aug 14, 2016 at 9:17 AM, Brett Cannon <brett@python.org> wrote:

...

That's a mess and the whole email is formatted like that. I actually have not read the email because of the formatting issue. As Oleg pointed out, when you go with a federated solution like mail, you are the mercy of whatever tools people choose to use with the service. But when you use a centralized approach you know the experience is consistent for everyone and thus there's a certain level of quality control.

IMO that's an argument in favour of the federated approach.

I don't understand how me receiving a badly formatted email due to some disagreement between the sender's and my email client is a point of support? -Brett

ChrisA _______________________________________________ Python-ideas mailing list Python-ideas@python.org https://mail.python.org/mailman/listinfo/python-ideas Code of Conduct: http://python.org/psf/codeofconduct/

The world probably wouldn't better, but I think python-ideas will be better on something like GitHub. People who like it will benefit from it, people who don't will still be able to use their email setup (as pointed earlier - at least for the majority of cases). Regards, Eugene On Sun, Aug 14, 2016 at 12:08 PM, Chris Angelico <rosuav@gmail.com> wrote:

On Sun, Aug 14, 2016 at 2:57 PM, Brett Cannon <brett@python.org> wrote:

...

On Sat, Aug 13, 2016, 16:26 Chris Angelico <rosuav@gmail.com> wrote:

...

On Sun, Aug 14, 2016 at 9:17 AM, Brett Cannon <brett@python.org> wrote:

...

That's a mess and the whole email is formatted like that. I actually have not read the email because of the formatting issue. As Oleg pointed

out,

...

...

when you go with a federated solution like mail, you are the mercy of whatever tools people choose to use with the service. But when you use a centralized approach you know the experience is consistent for everyone and thus there's a certain level of quality control.

IMO that's an argument in favour of the federated approach.

I don't understand how me receiving a badly formatted email due to some disagreement between the sender's and my email client is a point of support?

That in itself isn't. But your next point is that email lets people choose what they use, whereas centralized systems force everyone to use the exact same client (or whatever clients the one central authority provides - eg Slack offers web and desktop, and I think mobile). In fact, the entire *point* of the centralized systems is to force everyone to use a restricted set of clients, instead of having the freedom to choose. Would the world be a better place if everyone were forced to write all code in Python?

ChrisA _______________________________________________ Python-ideas mailing list Python-ideas@python.org https://mail.python.org/mailman/listinfo/python-ideas Code of Conduct: http://python.org/psf/codeofconduct/

On Sun, Aug 14, 2016 at 03:57:44AM -0500, C Anthony Risinger <anthony@xtfx.me> wrote:

Mobile dominates my non-work net-time today. I don't want to get out a laptop to respond pretty.

In what ways are you going to contribute without getting out to your laptop?

Email is almost by design static and unable to change. Text. Walls of it!

We are talking about Python development, and the development is performed with texts. Python is written in C, Python code is written... well, in Python, documentation is written in reStructuredText. So discussions about all of this is, naturally, textual. Oleg. -- Oleg Broytman http://phdru.name/ phd@phdru.name Programmers don't die, they just GOSUB without RETURN.

A uniform interface works well enough for issue trackers. And the "freedom of choice" idea doesn't overrule all other concerns. Maybe we should just start a python-ideas tracker and see who comes. There's no reason it couldn't exist in addition to the mailing list. (Before you scream fragmentation, we have many lists already.) --Guido (mobile)