Following up from stepping back

Thank you to those who responded to my previous post with thanks and words of encouragement. Just to be clear, I'm not stepping back from Python altogether; far from it. I'm simply giving up certain background responsibilities which I've found difficult to sustain. I realise that, by doing so, I've left the others who continue to manage those tasks more burdened than before. So, again, I encourage you to offer to help on whatever basis you think you can manage.

Python has grown. When I entered the community about 20 years ago, it was a moderate-sized mostly technical language community. It's always been growing in popularity, but over the last 10 years it's reached the point where it's used and relied on -- directly or indirectly -- by millions of people around the world. Indirectly by virtue of its use in some very big-name companies. Directly because it's become the go-to language for science and data projects. And it's the text-based programming language most likely to be taught in secondary schools in the UK and maybe elsewhere, especially with the encouragement of the Raspberry Pi foundation and its resources. The audience is huge.

Because of its size and significance, many people assume that there's a commercial organisation running things. Or at least a well set-up NGO or institute or something. But there isn't. There is a series of semi-detached volunteer teams who typically only interact because one or more people on one of the teams are also on one or more of the other teams.

The webmaster@ address is the fallback communication point for the Python community. It handles all sorts of queries. There are general requests for help; requests from official bodies for us to complete some complicated export license document; more questions about the cost of licenses than I care to think about ("No really, it's free; yes you can get a bulk discount of 100% because IT'S FREE"); questions about copyright and logo use; technical questions; questions from school administrators who don't grok the new Python Windows installers; people who've bought a car security system with the brand name "Python". And all those queries are fielded by about three people (now about two people!) working with whatever time they have available.

The Python mailing list (also comp.lang.python newsgroup) is the "official" general-purpose communication channel for questions about Python. But it's a mailing list / newsgroup, both of which are slightly ageing concepts. Its origin as the forum of discussion around a mostly technical audience still informs its general attitude. And, by its nature, it assumes that you're interested in joining a community and not simply asking a drive-by question. There's some experimentation going on at the moment about using a Discourse or a Zulip instance or both, which I think might be a move in the right direction for the wider casual audience.

People raise issues on https://bugs.python.org asking questions about https://pypi.python.org; they write to webmaster querying why they haven't received a confirmation email when they signed up on the https://python.org website; they write to mailto:docs@python.org when downloads aren't available on https://docs.python.org for the latest version of the documentation; and they raise an issue on the website tracker when there's a spelling mistake in the online documentation. And although I've put those examples forward as mistakes, it may not even be obvious to you why those are mistakes. python.org contains quite a few silos.

The easy majority of volunteers behind these various forums are technical people who've stepped up to help. Which is clearly very much to their credit. Back in the day, most people posting through Python forums would also have been those with a technical background who'd chosen to use Python: students, professional programmers &c. So there was an easier meeting of minds when it came to answering questions and giving advice. Nowadays they might be, for example, schoolteachers or data scientists who use Python as a tool (for teaching or analysis) but who have no knowledge of or, really, interest in the underlying mechanics of the language or its community. Or even schoolchildren looking for help with their homework or with a personal project. So there can be a mismatch between the people asking questions and those answering.

On top of all that, we're increasingly in a world where Codes of Conduct are needed, if only as a reference answer to "Who says I can't do that?". But that puts a significant burden on list owners and forum volunteers who, in their spare time and often without much guidance, need to respond authoritatively to complaints and responses.

What am I suggesting? Whatever I suggest runs the risk of sounding like "We really should be doing..." when I've just announced that I'm stepping back from doing things. However, in a spirit of constructive criticism I think I'm suggesting:

That the PSF take an holistic view of the different public-facing python.org properties (including bugs.python.org, pypi.python.org, python.org itself, the more public Mailman2/3 lists on python.org, planet.python.org, webmaster@python.org) with a view to creating a common understanding of the roles and responsibilities of those volunteer teams who manage those resources, including cross-fertilisation of ideas and a common meta-forum for dealing with misdirected enquiries and CoC issues. Perhaps even synthesising a single team to handle requests coming in through all of those and other channels.

Just as I was backing away from my python.org responsibilities, I think there may have been the beginnings of something like this: a meta-list for list admins. If there was, then I think that's a step in the right direction. Without question, there are wider technical issues which could be addressed given time and resources (eg the number of accounts needed to operate on different python.org properties). But I recognise that there are technical and even political challenges to making that work to everyone's satisfaction.

I don't know what's in store for Python in the future. Maybe it will continue to grow and evolve; maybe it will be overtaken by a new kid on the block. I'm not saying that it doesn't matter. But what matters more is what Python can offer now for the many, many people who find it useful, or who merely find themselves using it, and want support from its community.

Comments

Comments powered by Disqus