About once a year, I have posted a thread here about dribdat, a code-name, a project, and possibly a product, that supports the work of the Swiss open data community. A brief history: we started in 2011 with a good old Dokuwiki, a service which continues to be referenced and is maintained today by the OpenGLAM community. Then, in 2016 we deployed our own application written in hacked together in Python, originally at an event run with IoT Zurich. Rough and ready, it was put to use at first just mirroring the wiki pages. The basic idea was to have open aggregation of hackday results, accessible and available to all participants, self-hosted and totally hackable if people wanted to make improvements.
One year later, the project was already on issue #100, written as a kind of pulse check in hopes of getting some course correction. Around this time I researched the ‘playing field’ and started tracking the emerging market of hackathon platforms a little more closely. Thank you again and again to all the Contributors of this small but for me personally important endeavor! <3
Throughout the course of these years we had a lot of great MakeZurich events on the platform, and in 2018 together addressed the wider (open source / digital sustainability) communities at DINAcon, where @gnz and I ran a workshop to dig into some of the more fundamental and critical questions.
Later that year there were some super interesting collaborations with student groups, at the school of design in Bern and an information management class in Sierre. They contributed provocative observations, designs and code prototypes, from which we continue to draw ideas and inspiration to this day.
At the turn of the decade, I was busy with…other things. It was hard to find time for dribdat in 2019, but when I did, it was for quietly writing a whitepaper, maintaining the code, closing another 100 or so tickets and helping to support our first international users and just independently-run events. I wrote about this recently, in particular about some of the great data we have collected over the years.
We talk about quick hacks and launching early, but in the case of this project it took 3 or 4 years to get some real external attention. That said, we never really took much trouble to advertise the fact of its existence. Even today, you won’t find dribdat mentioned anywhere on our public-facing websites. It’s just infrastructure … humming quietly and getting the job done.
It is 2021, and I think it is time to change the status. Or even to start with a clean slate. It is with this that I addressed the board of Opendata.ch last week with a proposal to review our Community Infrastructure. In particular for Hackdays, with many upcoming, my ask is to review not just how to support dribdat, but also the chat platforms, registration tools, this forum - in short, all the things we depend on or could build upon for successful events and sustained open efforts.
In an evaluation last year, we already listed and tried to prioritize most of the many online platforms we support and administer on behalf of the community. There is a spreadsheet documenting dozens of services that provide value not just to the association, but also to association’s members and hackathon participants. Thank you to Darienne for refreshing this effort and helping me to reframe the discussion:
Above is a diagram made in attempt to clarify this. There is a lot to unravel here, and our overview was detailed with links to further documents and elaborate discussions. There is a roadmap we are supposed to be moving along, prioritizing on surely an infinite number of things we could potentially provide and improve on, giving focus to our very limited resources. Even deciding to what extent a particular tool serves the needs of the association board and team, as opposed to that of our members/followers/users, can be a fuzzy business.
The Analytics of the hack.opendata.ch platform over the past year can provide some indication about the lifecycle of our events, and help us to run a Hackday Platform Improvement Project.
My proposal is to address these top questions:
-
Who are our hackathon participants? (Du != Nutzer*In) Let’s better identify them as an audience to try to study the common needs, base our priorities on user research, find out what our core constituency would be willing to contribute to or (through membership fees or more directly) pay for.
-
When & how do people engage with our hackathons? (Deliver services) There is a planning, preparation, and post-production phase to every hackathon. It is arguable that our platforms do not cater equally, or equally well, to each of them today. Therefore we should make sure to understand both the diverse backgrounds and specific engagement points of our team and audience.
-
What key platforms to operate and partner on? (Good services are designed) As opposed to outsourcing or even just recommending platforms for open data work, there are a handful of tools which we have become very familiar with and maintain ourselves. Platform development can be an impactful source of activity and funding. With the outputs of the first two questions, we should structure investment and partnership-building accordingly.
Backed up by the following pros and cons:
+++ pro of working more on platforms
- as a detailed archive of former projects, codes and teams (sustainability and openness) create an important resource - see hackopendata-archive
- the application design and principles could match our branding and in general more closely reflect our goals and aims as an organization and movement
- we could prioritize workflows that make most sense to us, make sure the development improves capabilities of Prototype Fund and other programs
- when we become highly agile expert users, we can create exciting new services that connect to the central hub of data and activities
- hackathon teams are made of people with important skills and backgrounds - invest into utilizing these connections as an alternative social platform
- a product we develop/operate/repackage can be commercially licensed or supported professionally as an additional income stream
— con of working more on platforms
- software development and maintenance can be very expensive, and needs quite a different budget and potentially type of management than we have
- danger of losing sight of the bigger picture for want of constantly improving our training wheels - i.e. does the benefit here outweigh the cost
- hackathons have become an important area of activism and instrument for the association, but it is impossible to know how long interest will last
- we already have tough competition from an emerging market of hackathon service providers and platform builders
These are issues that I am sure many organizations face, whether they aim to address digitization topics or not.
The next actions I have suggested are:
-
Decide to take more ownership of developing the online Hackday platforms (pro) or should we continue to contribute or even outsource more of our dev & ops (con).
- Update the documentation around Hackday infrastructure, make sure we know what essential components are connected to what and to whom.
- Find out who on the team or board or wider community is interested to be part of this discussion and assist with next steps, such as defining roadmaps.
-
Do more user research to get input from people in our community over the course of upcoming hackathons to clarify and validate our plans.
- Distribute a short survey (drafted here) to get critical user perspectives.
- Run a workshop along the lines of “platforms for sustainable results and applying best practices” to get more direct feedback.
- Contact people on a 1:1 basis, starting with folks already following the Slack and Mattermost channels.
- Ask participants of upcoming hackathons.
-
Run a “hack the hackathons” event or challenge - use the hackathon format itself to stimulate participation from our community.
- Collect an interest group for a call or workshop and follow up on their requests.
- Talk to Open Knowledge, who are planning their own hackathon in October, and are interested among other things in Frictionless Data support in dribdat.
- Pitch a challenge for HACKnight.DINAcon.ch on October 29.
That’s at the top of my head. What priorities do you see for making our hackdays great again?