Maybe Micropython needs more people?

C programming, build, interpreter/VM.
Target audience: MicroPython Developers.
greetasdf
Posts: 2
Joined: Mon Oct 19, 2020 9:07 pm

Maybe Micropython needs more people?

Post by greetasdf » Sat Jan 02, 2021 12:09 am

It seems with the recent delay of release 1.14 and >300 pull requests and >800 open issues, doesn't Micropython really need more people developing this project?

Was this discussed yet? I wonder what Damien's thoughts are on this, is it a money issue, do people not have the time or not the skill to contribute effectively? How do people contribute effectively? Is the main issue advertisment and there any plan for public PR? Should people contribute to Circuit Python (because some say that's the future) or to Micropython, since it's the base?

Maybe wrong branch to post, but it kind of felt appropriate for 'development of micropython' as in growth of the whole project

User avatar
pythoncoder
Posts: 5096
Joined: Fri Jul 18, 2014 8:01 am
Location: UK
Contact:

Re: Maybe Micropython needs more people?

Post by pythoncoder » Sat Jan 02, 2021 5:22 am

I'm loath to comment in detail but I think that @Damien's laudable focus on code quality limits the bandwidth for user submissions.

I do think that the difference between MicroPython and CircuitPython needs clarification as their focus is quite different. CircuitPython aims, above all, to be beginner friendly. As such it omits numerous performance-focussed features of MicroPython such as interrupts and (currently) asynchronous programming. To get the most from MicroPython requires some investment in time. In my opinion this is well worthwhile. Many of my projects would have been impossible in CircuitPython.

There is no question that MicroPython is well ahead in crucial features, not least an entirely new and much improved uasyncio library. Given that asynchronous coding is at the heart of most serious firmware applications, this is a big deal. IMO, of course ;)
Peter Hinch

greetasdf
Posts: 2
Joined: Mon Oct 19, 2020 9:07 pm

Re: Maybe Micropython needs more people?

Post by greetasdf » Sat Jan 02, 2021 1:10 pm

Thanks for the answer even though not detailed! In other words the 'inventor' of micropython is at his limit to quality check everything and that creates a bottleneck. Is there anything that fellow community members / readers / enthusiatsts can do to help with this sort-of bottleneck?

User avatar
Roberthh
Posts: 2566
Joined: Sat May 09, 2015 4:13 pm
Location: Rhineland, Europe

Re: Maybe Micropython needs more people?

Post by Roberthh » Sat Jan 02, 2021 1:59 pm

There are a few people working on Micropython. Especially contributtion to ports are in control other people too than Damien. But it is also true that Damien controls the core software of MicroPython. And that is good. It ensures consistent quality. The feedback time is reasonable, the quality of feedback excellent.
If you look at other branches of MicroPython, like the Pycom branch, you see the downside of several people working on the product. No one seems to have full view and understanding of the firmware. There seems no process for reviwing and eventually integrating PRs. Occurrences of bug regressions, ... All caused by communication problems. Adafruits seems to be a little bit better organized, but also there it seems to be a very small group of people controlling it, like 2-3.

stijn
Posts: 596
Joined: Thu Apr 24, 2014 9:13 am

Re: Maybe Micropython needs more people?

Post by stijn » Sat Jan 02, 2021 6:21 pm

greetasdf wrote:
Sat Jan 02, 2021 1:10 pm
Is there anything that fellow community members / readers / enthusiatsts can do to help with this sort-of bottleneck?
Helping out on github can do quite lot. For instance there are always issues being raised which aren't actual issues and/or belong on the forum; pointing that out and helping people means those issues can get closed without any interaction from maintainer(s) required. Likewise: reviewing PRs, within your capabilities of course. Just simple things as checking spelling/grammar get the PR in a better shape, i.e. less work. Another simple one is double-checking against contributing.md rules for code style/commit message format, people often forget those so if someone points that out it's again less time/effort needed from others. I'm just mentioning these because they don't even require one to know C or the MicroPython internals. Of course if you do, review code as well.

User avatar
jimmo
Posts: 2262
Joined: Tue Aug 08, 2017 1:57 am
Location: Sydney, Australia
Contact:

Re: Maybe Micropython needs more people?

Post by jimmo » Mon Jan 11, 2021 12:28 am

greetasdf wrote:
Sat Jan 02, 2021 1:10 pm
Is there anything that fellow community members / readers / enthusiatsts can do to help with this sort-of bottleneck?
Documentation is usually my top suggestion.

Especially documentation that facilitates high quality PRs, or helping people get started with their first PR.

User avatar
pythoncoder
Posts: 5096
Joined: Fri Jul 18, 2014 8:01 am
Location: UK
Contact:

Re: Maybe Micropython needs more people?

Post by pythoncoder » Mon Jan 11, 2021 8:46 am

jimmo wrote:
Mon Jan 11, 2021 12:28 am
...
Documentation is usually my top suggestion...
On which topic, whatever became of the Season of Docs? I thought this provided 3 months full time effort from a professional technical author.
Peter Hinch


User avatar
pythoncoder
Posts: 5096
Joined: Fri Jul 18, 2014 8:01 am
Location: UK
Contact:

Micropython Internals doc

Post by pythoncoder » Tue Jan 12, 2021 5:37 am

Wow. There's some serious bedtime reading there! From a quick look, highly recommended. :D
Peter Hinch

User avatar
deshipu
Posts: 1385
Joined: Thu May 28, 2015 5:54 pm

Re: Maybe Micropython needs more people?

Post by deshipu » Sat Jan 16, 2021 9:55 am

So sad it will be outdated even before it merges.

Post Reply