Hi Elena,
I'm not questioning the ability of community members to submit PRs, but
I do share the concerns of Mats regarding the PR review and approval
process.
At the end of the day, someone will have to review the PRs, request
changes, possibly rebase them, etc. All this is extra work, which tools
such as wikis or Google Docs don't require.
*A change in technology won't fix the inherent problems:*
My point is that even when you make it really really easy for users to
modify content (often at the risk of attracting spammers and people who
enjoy vandalism), you still have to find editors who really care for
parts of the topics covered in the resource... and that's the really
hard part.
Developers tend to not be very keen on producing documentation, or even
in-line comments in their code. The few who really do care, often find
themselves chasing changes in the code base and updating documentation
after the fact.
With resources such as the wiki, the situation is similar. Many people
use these resources, but only very few actually care about topics enough
to make the (small) extra effort it takes to update, correct or curate
changes for the better.
*As for the idea to run a poll:*
I don't have much faith in results of Internet polls. People will
typically submit choices without actually contributing afterwards.
If you really want to test drive a community based resource based on a
Github repo, I'd suggest to set one up, run it for a year and then
compare the outcome to other platforms or strategies.
*What may help improve things:*
IMO, a better idea would be to ask on some of the platforms for help
with curating the Python Wiki.
Not just simply: "Can you please help ?" But instead with a strategy we
would have to develop beforehand to help guide people to areas in the
wiki, which could benefit from more attention and also with a clearer
set of rules to follow and perhaps some extra discussions around changes
on this ML.
As with all such projects, some level of management is required to make
things happen in a productive way.
If you think that such a request could DDoS the wiki, we could gate this
by only taking up, say 50-100 new editors. They would still need to be
made editors by us to be able to get write permission to the wiki, so
that's certainly possible.
In order to prepare for this, we could have a discussion on the
strategy. I think this would be more productive than seeking to
implement new technology first.
PS: I've seen too many instances where people thought that new
technology was the key to improving things, only to find out that they
were just replacing one set of problems with another. And not making
progress on the overall problems they were trying to address.
Ironically, lack of documentation was often a driver for such attempts...
Thanks,
--
Marc-Andre Lemburg
eGenix.com
Professional Python Services directly from the Experts (#1, Mar 02 2025)
Python Projects, Coaching and Support ... https://www.egenix.com/
Python Product Development ... https://consulting.egenix.com/
________________________________________________________________________
::: We implement business ideas - efficiently in both time and costs :::
eGenix.com Software, Skills and Services GmbH Pastor-Loeh-Str.48
D-40764 Langenfeld, Germany. CEO Dipl.-Math. Marc-Andre Lemburg
Registered at Amtsgericht Duesseldorf: HRB 46611
https://www.egenix.com/company/contact/
https://www.malemburg.com/
On 02.03.2025 00:19, Elena Williams via pydotorg-www wrote:
Hi Marc-Andre,
Thank you again for the detailed response.
I would like to challenge your assertion here regarding the ease of
making edits.
While technically what you say should be true, and it indeed was true,
the development ecosystem has evolved in an unexpected way in recent
decades.
I assert that in this era, in 2025, there is a lower psychological
barrier for a technical person (particularly a younger one) to send a
PR to a documentation github repo than contribute to a wiki.
The evidence I have for this is in my own teaching of software
engineering and computer sciences students at one of Australia's most
elite universities. Both this and last semester 4-500 students have
passed through the software engineering professional accreditation
capstone project, which I am heavily engaged with. First hand I can
tell you that 100% of these student know how to make a commit, ~75%
know how to make a PR, 60%+ will be confident in managing PRs in a
github context by the end of their degrees.
As compared to when I did my studies, or even when I started teaching
this course in 2017 these numbers are incredible, virtually
unbelievable, yet it is now the expectation for these graduating
university students. Of course all of us as professional software
engineers these days must know how to make PRs.
As for counter-evidence I don't have such big numbers but I do have
the responses of people with whom I've discussed this, which have been
firm in the negative, though only anecdotal.
I wonder how many takers/responses you get when you drive for editors?
--
I do have an idea for a way to get some concrete numbers.
I am a moderator of Django discord which has 27.9K members, and I was
considering running a poll there (particularly as I would have
moderation control of it), but the better place would the be the
official Python discord which is one of the biggest discord servers in
the world with something like 370K members. It is truly a phenomenon,
they are very successful there. I was pondering how/where running such
a poll would make sense in that sprawling community, when I noticed
that my friend from PSF EOW, KeithTheEE is a moderator in this
server (so the story comes around).
*How would you feel about the idea of running a poll about the wiki in
the python discord?*
*Specifically we could ask the question as to what interface would be
"easier" for the community there to engage with.*
I would be very happy if you wanted to write such a poll.
For the record: based on my dealings with wiki.python.org
<http://wiki.python.org> in the last few months I'm a bit concerned
that even just asking this question may effectively DDoS the wiki.
Irrespective it's unambiguous progress in terms of putting the project
in the minds of a pool of potential new editors.
---
Elena Williams
Github: elena <http://github.com/elena/>
On Sun, 2 Mar 2025 at 09:43, Marc-Andre Lemburg <m...@egenix.com> wrote:
Hi Elena,
such repos are good for documentation, but accessibility is rather
limited and requiring PRs to add/change content will not exactly
be an invitation to people wanting to help curate the content.
Our wiki provides very easy ways to edit content and we make
access very easy as well. The level of entry is much lower than
for a documentation repo such as the devguide (and I'm not even
talking about the necessary CD to turn the repo into the actual
content).
As mentioned, the tech is not really the problem. It's finding
enough people to engage is curating the content, that is.
PS: I've tried lots of ways of making content editing easier for
contributors in my day job as CTO, software architect and project
lead. By far the easiest was creating a set of Google Docs, but
even with those, it was hard to find and motivate people to
provide and curate content. It's really a people problem, not a
tech one.
Thanks,
--
Marc-Andre Lemburg
eGenix.com
Professional Python Services directly from the Experts (#1, Mar 01 2025)
>>> Python Projects, Coaching and Support ... https://www.egenix.com/
>>> Python Product Development ... https://consulting.egenix.com/
________________________________________________________________________
::: We implement business ideas - efficiently in both time and costs :::
eGenix.com Software, Skills and Services GmbH Pastor-Loeh-Str.48
D-40764 Langenfeld, Germany. CEO Dipl.-Math. Marc-Andre Lemburg
Registered at Amtsgericht Duesseldorf: HRB 46611
https://www.egenix.com/company/contact/
https://www.malemburg.com/
On 01.03.2025 18:22, Elena Williams via pydotorg-www wrote:
Hello Mats,
Not using the "wiki" part of a github repo -- I agree this is
awful! Upon exploring this and another upfront fatal flaw is that
file trees are not at all supported.
Perhaps more following the pattern established here:
https://github.com/python/devguide
https://devguide.python.org/
---
Elena Williams
Github: elena <http://github.com/elena/>
On Sun, 2 Mar 2025 at 04:10, Mats Wichmann <m...@wichmann.us> wrote:
On 3/1/25 10:01, Elena Williams via pydotorg-www wrote:
> Dear Marc-Andre,
>
> Thank you for your prompt and detailed response!
>
> Would it be an option to consider moving the wiki to
github? It's
> unambiguously demonstrated that these days that content of
the scope of
> the wiki is well within what they are accustomed to.
There's an absolutely fatal flaw with github-hosted wikis,
even more
than that it's an absolutely terrible wiki implementation:
content is
not made available to search engines for indexing. If we
think it's too
hard for people to find stuff via the wiki now, it would get
that much
harder. On a project I'm involved in, we're actively looking for
someplace else to put ours, but we're not a funded project so
haven't
found a solution yet.
_______________________________________________
pydotorg-www mailing list
pydotorg-www@python.org
https://mail.python.org/mailman/listinfo/pydotorg-www
_______________________________________________
pydotorg-www mailing list
pydotorg-www@python.org
https://mail.python.org/mailman/listinfo/pydotorg-www
_______________________________________________
pydotorg-www mailing list
pydotorg-www@python.org
https://mail.python.org/mailman/listinfo/pydotorg-www
