Open Source nostalgia — Can we just move forward please?
This is #3 of 5 posts I had in draft state for a few months, that I decided finish up & post. Here’s hoping my research helps others. I started writing this in May.
“Inessential weirdness of open source”
This term (crediting to Katrina Owen at Github) perfectly describes a conundrum of open participation, whereby we hold onto symbols, processes, and idiosyncrasies of open source in a mix of nostalgia, delusion and … I’m going to say it – arrogance , as the primary (nearly holy) measures of ‘being open’ in community building .
‘But’ve always done x’, is a very common response to change in open communities. Whereby we unintentionally (yet deliberately) avoid change because we believe that that purity of ‘open’ is the only way to innovate further . We even avoid change despite huge potential to grow more diverse and healthy open communities – because… there are slivers of non-open. gah!
Two years ago I ran the ‘Open Hatch Comes To Campus’ workshop at the University of Victoria. I spent 1.5 hours teaching people the skills they needed to ultimately… type ‘hello’ on an IRC channel.. Our workshop implied IRC was a critical doorway, and on-ramp to participation in open source. Saying hello, asking for help – with an instructors guidance: 1.5 hours. What?
I’ve often heard project maintainers say, that obtuse processes like these actually help ensure the success of those who are truly serious about contribution. As if asking basic questions is a holy grail of volunteering- one where only those willing to waste ridiculous amounts of time on discombobulated, obtuse processes and tools are worthy of participation. I call bullshit on any process that makes connecting with others, in an ‘open project’ – an obstacle.
“open and accessible doesn’t beat usable and intelligent”
In the last couple of years we’ve seen open communities faced with an interesting choice of using tools that work really well for working open, but are not themselves open. Github being the most obvious example. Similarly I’ve also followed the Open Data communities use of + Slack + Slackin!
Still in the voice of nostalgia asking us to remember our legacy IRC.
Anyway….what exactly do we need our community software to do? Here’s a short list I used when measuring chat solutions (and sure I am missing things)
- Open source – I want the ability to inspect, and improve-on software we use for community conversation, and to propose improvement via pull requests.
- Data is discover-able via web search. So much success of ‘open’ is that people can stumble on conversations that push innovation further.
- Open Conversations – no login or registration required. Anyone can ‘lurk’.
- Easy to grasp & intuitive – Lets not ask newbies to install software to ask for help. Lets’ not expect that contributors are technical contributors.
- Github feed (my own requirement, that everyone can see new issues, and comments they subscribe to).
A clever human-connection setup should allow new contributors an ability to answer these questions with some clarity:
- Who is here?
- Am I welcome here?
- What’s happening in this community?
- How can I contribute?
- How do I ask for help?
With this criteria, and questions in mind, here are the results of those I researched for education contributors at Mozilla:
Mattermost – Has potential, but seems unfinished, and little ‘alpha’. Without installing myself ,I couldn’t figure out how to enable a Github feed.
Gitter – I discovered this when looking around Free Code Club. I liked the UI, and possibilities for multiple channels easily toggled, searchable and friendly. Plugins tend to be more developer-friendly, which was a drawback for non-technical contribution – but not a show stopper. Has a great search option for communities. Chat rooms are associated with Github Repos, which has huge potential for building communities around projects and initiatives.
I think Gitter is doing with Github, what Github should be doing for Github projects interested in nurturing participation.
Discord – I found found Reactiflux development via Facebook React’s repo, but was nervous about jumping in.
Seems more like a team project, than community. I found it intimidating, especially with voice, and it wasn’t clear what preferences where. Quickly left.
I revisited this after comments were left about this project portal being community organized (as it had been months since I was there). Aside from struggling to switch login/register status, I do have to say it’s a very easy to lurk into – and has desktop versions (it seems I didn’t have a lot of time to test). I’m not clear on how discover able conversations are outside of this app, but the community has set things up very well to ask questions in a number of ways (which is awesome). Still on the fence about voice chat, but maybe that’s because it’s harder to stay gender-anonymous with voice. Thanks for the comment that made me take another look Mark!
Rocketchat – It’s open source, it looks great – it has the potential to do what Gitter is doing for communities, but it feels very single-instance and Slack-replacement focused. Don’t get me wrong, it’s beautiful, very capable of being a good alternative but I want more – I want ‘open’ feel like more than code. If I had to choose an alternative it would be this one.
Rivr – I couldn’t find inspiration other than free, and not-Slack. Guessing it’s a great alternative too.
Slack should be thought of as first generation example of how community might meet, connect with participation and community, but not as a template, and not as a ‘bar’ that we now try to replicate openly. Reactiflux community has also demonstrated that a cohesive collection of support vrs any one solution is often the best way to go as well.
It’s time we prioritized connection of humans ‘ in the open’- lets end the inessential weirdness of open source.