Berlin does civic tech. Großartig!

I’m back in the UK (is that still a thing?) now — but I have two more EuroTrip stops to blog about. 

First, Berlin. As you may know, Berlin is excellent. It also has a bit of a reputation for tech startups. And for doing interesting political things with tech.

It seems a long time ago that there was excitement about ‘Liquid Democracy’ — the German Pirate Party software that was going to revolutionise representative democracy. It allowed constant ‘delegative democracy’: you could choose to delegate your vote to someone on a certain topic, but take it away from them again or choose a different delegate at any time. In addition, your delegate could delegate your vote, and so on. Hence the ‘liquid’ bit — power would flow as voters chose and changed their representatives at will. The software made this practically feasible for the first time. There were some excitable blogs about it. But the revolution never came. I wondered what became of it — my trip to Berlin revealed the answer.

And there’s much more happening in Berlin today — in a way that is perhaps more realistic and more understanding of how most people want to engage. 

The scene

Berlin’s civic tech scene seemed not quite as scene-y as Paris’. This could of course be a result of the wholly unscientific sample of people I met. Or maybe that Paris seems smaller somehow. Anyway. Germany’s chapter of the Open Knowledge Foundation seems to fill in for a kind of central ‘hub’ of civic tech for Berlin, in the way that Newspeak House now does in London, and that Paris’ anticipated Civic Hall will. I do think having a physical space for lots of different projects to come together and work under one roof, share ideas, critique and funding does make a positive difference. There is probably research on this somewhere. I bet the Return on Investment is impressive.

Berlin, like Paris, seemed to have a broader range of civic tech projects that had been running longer than those in the UK. And only now are some of these projects reaching a sustainable financial place and, in some rare cases, gaining official interest. Hopefully this suggests that things will only get better over time for the UK.

I’ve not written on the whole security, privacy, distrust-in-big-US-digital-corporations issue, though it was well in evidence — Tom Steinberg takes a look here — and there’s probably enough in this amazing street art (actually from Cologne, but still):

A giant NSA/CIA eagle, complete with drones, watches over the Sheeple. (Ehrenfeld, Köln)
A giant NSA/CIA eagle, complete with drones, watches over the Sheeple. (Ehrenfeld, Köln)

Anyway, projects. First up, one project notable for its sustainability, and its amazing uptake by parliamentarians. Lots of lessons here for the day job:

Parliament Watch

Haus der Bundespressekonferenz (Ansgar Koreng CC BY-SA 3.0 (DE))

This is the Haus der Bundespressekonferenz, built by Germany’s national press association. The symbolism here is that the media invites the politicians to answer questions in its own building — specifically in that big jutting-out glass box — rather than heading to the federal press office. It’s a great building, packed with journalists from international media and German local media. And it’s home to, who’s Gregor Hackmack also runs or ParliamentWatch.

ParliamentWatch’s name suggested to me that it was something along the lines of TheyWorkForYou (mySociety’s classic parliamentary monitoring website) — but actually Gregor suggests that the Bundestag’s website is good enough not to worry about this.

Instead, ParliamentWatch is about trying to connect parliamentarians and voters, most notably through a Question and Answer system, but also through research and blogging on any iffy practices that parliament gets up to. PW keeps a close eye on parliamentary expenses, for example.

The Q&A was the most interesting bit for me. It’s a simple idea: people log-in, ask a public question to a politician, then receive a public answer. This is an idea replicated across the web, but I’ve never actually seen it work on such a scale and so successfully. Ninety per cent of politicians in the Bundestag have answered a question on the platform. 90%! That must be some kind of world record.

The politicians are asked an average of 7,000 questions per month — and 80% of questions in the last parliament (09-13) were actually answered. My German’s not good enough to trawl through them to guess at the quality of answer, and I would love to see user feedback on this, but it’s still an amazing number. I wonder how much repetition there is — and whether improvements could be borrowed from something like Quora (perhaps even Reddit) on designing to improve the quality of debate. But they’re still way ahead of any other legislature I know of. 

The Q&A platform is not completely open. The questions have to be approved first, and PW has a team of paid interns all working a few hours a week to review questions.

As well as the product, ParliamentWatch is interesting due to its model of financial sustainability. It has managed to get several thousand ‘sustainers’ across Germany to give at least €5 per month so it can afford a small permanent team and lots of freelancers. How did they manage to convince several thousand people to donate? Gregor tells me that democracy is too meta for most people, who think and care about issues, not the democratic process. So to build its supporter list, PW campaigned on aspects of parliament that might outrage people — such as closed nature of expenses information — and grew its list. If I recall correctly, I think PW ended up suing the parliament for this information — which needed supporters to chip in to pay for the suit, and those supporters stuck around afterwards. Gregor’s also a dab hand at opportunities for media coverage, so now PW is a go-to source for German media needing commentary on the practices of parliament.

Getting together a large group of people who care about transparency and accountability and converting them into regular donors has worked brilliantly for PW. It took them a while — they’ve been working on the project since 2004 — and sustainability didn’t come quickly. Early on the team used grant-funding, delivered some consultancy work and did something cunning with loans for equity, which they’re now en route to repaying, getting back the equity, and becoming totally independent again. It was clear that being sustained by the people, and nobody else, was extremely important to Gregor — and I like this model too. (Interestingly for WhoCanIVoteFor — a Democracy Club project — PW also charge politicians €200 to get a photo and a few other benefits for their profile page. I wonder if DemoClub could offer a similar product…).

I think ParliamentWatch is one of the most successful civic or democracy tech projects I’ve seen, full stop. It’s perhaps found the best balance of capacity, independence and sustainability of any such organisation in Europe, other than mySociety. Impressive.

Code for Germany


Meanwhile, in a quiet corner of East Berlin, I met with Julia Kloiber of Code for Germany. This borrows its name and basic strategy from Code for America, which takes bright young coders and puts them on a Fellowship programme embedded into a keen local government office. In so doing, CfA tries to inject new skills into the system and to create a cadre of people across the nation doing good stuff for government. 

However, Code for Germany ( has found that local government isn’t quite ready for this level of commitment. This is a problem because the plan relies upon local govt to pay the Fellowship’s salaries. So instead, are building a network of local groups — so far 25 strong — and centrally setting an agenda or challenge before encouraging the groups to just get on and build stuff to showcase the value of civic hacking. The HQ also runs a couple of workshops per year that bring all the chapters together. Eventually, goes the theory, the local governments will realise the value and start putting some money behind it and request a ‘fellow’. At the time we met, there were signs that the government of Hamburg was about to get involved. 

The successful creation of thriving local chapters is in part thanks to piggybacking upon an existing network of local Open Knowledge Foundation groups. It’s not any weaker for that — just a useful note when thinking about whether we need a Code for Britain. I think this might be one model for a ‘Local Government Digital Service’ that occasionally gets talked about. Or, as we’ve wondered about at Democracy Club, a ‘Code for Democracy’ where we get bright young things to come and do some volunteer/short term work on an idea that could benefit democracy. 

Recently, has managed to land significant funding from the German federal government (the Ministry of Education) for a prototype fund, which will be awarded to civic tech projects — many of which will presumably come from the existing groups and community. The prizes are capped at €30,000, which will ensure a wide range of projects get more development cash. The fund closes at the end of September and it’ll be great to see what gets funded.


Civocracy screenshot from Berlin

Civocracy is a local consultation platform. It has an attractive design and…wait for it…it has clients! And having proved it can win clients, it’s managed to receive some venture backing. In that sense it’s definitely a civic tech ‘startup’. On the platform, citizens can open discussions, make propositions or ask questions about solving a local problem.

It strikes me that there are a few of these consultation platforms that local government can choose to use, such as the UK’s But this approach reminded me of the problems of start-up land. The incentive for companies like Civocracy is to sign up councils, not to increase engagement. Ben Snow, the founder, explained that Civocracy do offer and deliver training and advice on what works for creating public engagement, but that the local government body has to be ultimately responsible for the levels of engagement. A payment-by-results model would be too risky. That might seriously incentivise user-research though; I dare a council to try it. 

Perhaps these kind of local engagement products will work by building the business in reverse. Get the users first, then sell that data to local government as ‘consultation’. It probably doesn’t feel right to the cities, cos they probably want to dictate the question — but this is a more user-friendly way. Then the question shifts to ‘how do we generate the viral growth of people discussing and debating local policies’, which I think relies on great UX, design and things that will benefit citizens, not necessarily civil servants. Which is probably the right way around. Sites like, or the seriously VC-backed, might be examples of that approach. There might be a unicorn here, or perhaps Facebook will just win again.

Liquid Democracy

It lives! Gregor from ParliamentWatch introduced me to an organisation called LiquidDemocracy, e.V. — where I could find out what became of the revolution. Actually, it’s quietly going fine. In short, the folks behind it split in two — some to work further with the Pirate Party and some to work less on voting, more on improving online deliberation and discussion. The latter group is now the company whose team I met over pizza in hip Neukölln: Nils, Rouven and Moritz. (Aside: these good people are currently offering free office space to civic tech Berliners.)

The company has some interesting clients and projects. They’re working with a few local authorities to run democracy experiments in schools: the kids get to come up with motions, debate and vote on them — using both offline and online discussion. The important point is that the schools agree in advance to enact whatever the kids decide — so the kids have a strong incentive to take part. Happily, for worried headteachers everywhere, the kids have not proved terribly radical. But the process gets the pupils into thinking about the democratic process and engaging in online deliberation before making decisions. This is good long-term democracy-building stuff.

From runway to playground (CC Tony Webster)

The team has also worked on the Templehofer Feld consultation — the conversion of Berlin’s wartime airport into a public park, which happened only after a significant citizen activism. And they’re working with the city of Berlin on — an online participation platform that is still in Beta. 

Meanwhile, all is not lost for the revolutionary system of deliberative democracy — Liquid Feedback — the delegative voting platform — does still live as open-source software for voting, though it’s not clear whether anyone is using it.


FragDenStaat (or ‘Ask the state’) is Germany’s equivalent to the UK’s WhatDoTheyKnow – the freedom of information request site. Housed in Open Knowledge Foundation’s office and managed by one charming, sleep-deprived German, they’ve found it a little difficult to fundraise, but the site’s been used to make over 10,000 requests. There’s also Verklag Den Staat for when things get more serious: Sue the State. 

Arne (the sleep-deprived German) has some exciting plans for a sue-the-state fund to help with legal costs if users actually do end up in court. If the person pursuing information wins the case, the state pays costs, so the fund can get the money back and put it aside for the next attempt. I think Arne said that the money is coming from a partner organisation, but you could easily imagine a crowdfunding (Crowdjustice?) aspect to this. (As an aside, a nice example of a request FragDenStaat made and is now suing for, concerns the VW emissions scandal. The way Arne tells it, VW is such a treasured German brand that the elite have closed ranks around it, meaning that VW is getting off more lightly in Germany (and the EU) than in the USA, for example, where the manufacturer is paying thousands of dollars to every VW owner. Suing will help open information about the political decisions taken when the scandal broke. Good stuff.)

Civic non-tech

Lastly, I did meet a not-very-tech project that deserves a mention. EuroTopics is a Herculean project that tries to bring together opinion/editorial from across the 28 member states’ newspapers to get some sense of a pan-European conversation. In one daily digest. It’s kind of fascinating to see an issue from so many different viewpoints. At the moment it’s not very techie: freelancers across Europe read, digest and translate for a central editorial team in Berlin. We chatted about a future where algorithms might do some of the comparing, highlighting and digesting, with human oversight.

‘What does Europe think?’ is an amazing question — and one that might help create the European public square (see back to conversations I had in Brussels). I also think there would be commercial demand for a euro-news roundup of ‘what is Europe thinking about Brexit’ — even if the only buyers are BCG/McKinsey/PA or whoever is going to earn £millions organising Brexit. Sob. But do sign up to the email digests.

Next stop, Amsterdam.

Catch up on Brussels or Paris here.


Corrected:  Frag Den Staat is Ask The State; Verklag Den Staat is Sue the State. Thanks Alex!


Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s