From the desk of the CEO and Founder of

The (almost) Endless Week

Whoops, it’s been a while since my last post. What happened? Well Berlin Summer came and took its toll. And so I got stuck:


The typewriter project I’ve announced was taking of course just a bit longer than expected. The initial idea was to build a text editor for the web that saves everything the moment you type it under the exact URL you’re currently on. No big deal, but at some point features started creeping in. It started with Markdown support, then drag ‘n drop image and text uploading and suddenly I found myself adding realtime collaboration. Ooops, no wonder it took ages to get it done. But here it finally goes, say hello to


During these weeks there was also another time waster creeping in. Mr. Lee Martin. Back in SF on one of my last nights after leaving SoundCloud Lee and me made great plans to work together in the future. Given that we were just half way through a bottle of whisky at Frankies, who would have thought that it would actually happen? There were 9 timezones and 2 egos against us. Turns out we’ve made it work! Last week we’ve released the first projects that kept us busy the last weeks.

The Social Connectome

The first one was a teaser campaign for the new Muse album. Its artwork contains a visualisation of the neural network of a brain made by the Human Connectome Project. So we’ve tried to translate this idea into something we’ve called the Social Connectome. But instead of the human brain this Connectome maps and visualizes the social connections between Muse Fans. And to make that happen each visitor gets a unique link to share, which allows us to identify from which other user a visitor was referred to the site. And with that data the app was drawing a line for each of these users, connecting him to the referring users all the way down to Muse. Along the way we’ve figured that it would be pretty hard to explain the concept. But once released and growing the Social Connectome luckily spoke for itself, somehow:


For the visuals we’ve used Mr Doobs fabulous Three.js library and wrote some custom WebGL shaders. Pretty heavy stuff. So for the next project we went from the cutting edge back to the origins of animations on the web: Animated GIFs. What better way to create a fake 3D car driving experience?

The Road To Battleborn

This project was the premier for the new The Killers album Battleborn. The artwork made it already pretty straight forward what to build: Take a virtual roadtrip through the desert while listening to the new album together with your friends. To make it a bit interesting we threw in some personalized signs and billboards for the side of the road. Lee wrote a bit about the details. Read it and have a look at


The first projects are out and the end of the summer in Berlin came. So what’s next? I’ll be off to a little Manoir in the Bretagne for a while. Lee will come over for some whisky in front of the fireplace. So there is a good chance for some more stupid masterplans for the future.

Week II: The Adventures Of Stan

Consistently late by a day and even worse I’ve abandoned the project I’ve announced (see bottom). So on saturday I just went for better idea that I always wanted to do: Building an adventure game! To be realistic, given the fact that I only had 4 days left, I focused mainly on the engine and used some well known artwork and characters. Say hello to the Adventures Of Stan: Episode 0.1:


Stan is finally able to use his great selling talent on Edward Hoppers Nighthawks. Mission Accomplished. Of course it’s very much half baked and missing some critical features like a proper inventory, item and dialog system, but hey it’s a start!

The underlying engine I wrote is called NAIVE, which stands for Naive Adventure Invention Engine, and is using lots of div’s to make the magic happen. It even runs relatively well on iPhones. The source is already up on GitHub, so feel free to hack on it. I think it’s a good base to create a slightly more ambitious adventure game with the help of some friends soon. Maybe during the next Ludum Dare. Until then have fun playing this one.

Post Mortem: Well now I have a new ongoing project instead of a finished one. A bit of a fail, but I had tons of fun while building it, so screw that! To keep it going I’ll spend the next week to polish up another favorite of mine: a typewriter for the web.

Footnote: I’ve planned to build a simple service that accepts a screenshot of your iPhone homescreen via Email, identifies the apps and creates a profile page as a way to recommend the apps you are using to other people. But in the end I wasn’t excited enough to really kick it off and also discovered that similar services are popping up like daisies right now (Although they don’t do the automatic detection).

A fresh take on Wikipedia

After deciding to leave SoundCloud I immediatly began to think about what to do next. While looking for inspiration I made a list of startups that inspired me most. Interestingly on top of that list was a site which actually isn’t a startup: Wikipedia.

With over 22 million articles it is still the 6th biggest website. Looking up a person or a thing on the web will almost always lead you to Wikipedia. It’s so much valuable content! And all created and owned by a huge community of people instead of some commercial entity that hoards it in its data silos. I think we should all appreciate more what a great resource we have with Wikipedia.

And yet I’m also worried about the state of things and fear that Wikipedia could loose it’s traction over the long run.

New services like Googles Knowledge Graph, which is now directly embedded in Google result pages, could lead to significantly less traffic to Wikipedia. And even if visitors find their way to it they are greeted with a site that still looks like it’s 2002.

What we need is a fresh, state-of-the-art consumption experience that encourages exploration, curation in form of sharing and leads to participation.

"Ask not, what Wikipedia can do for you. Ask what, you can do for Wikipedia." JFK

Well, here is my take on it:


It is a simple client-side app running on the short domain which allows quick access to all the Wikipedia content. By hijacking the original Wikipedia page it’s possible to extend it with features like map & image overlays, but also to remove a lot of the clutter without having to rebuild everything from scratch.

Additionally it introduces a few social sharing options like Facebook & Twitter. Even Facebook Activity Sharing, just like you know it from Spotify, is possible. That means that if you opt-in all the articles you read will show up in your Facebook feed. While I understand that not everyone is a fan of this, I think it’s worth the experiment to see what kind of feedback loops are created when your friends can see and interact with your activities on Wikipedia.

But the best thing? It has a plugable architecture that makes it very easy to extend it. So if you have your own ideas about how to improve the Wikipedia Experience go ahead fork, push and pull-request the hell out of it on Github!

I think this is a good start and hopefully it will fuel the discussion a bit about how we can keep the innovation on and around Wikipedia happening. So give a go and let me know what you think.

Woohoo! The first week of Operation: is wrappend up! Unfortunately I was a day late because I had to wait for Facebooks Open Graph approval (got rejected because I used the action verb “read”. Now it’s “study”), but I’m still happy with the result. So 5 more exciting weeks and random projects to go. The coming project will be about an easy way to share what apps you’re using on your iPhone. Again I’m aiming for a release on tuesday.

So stay tuned!


Last week was my last day as a developer at SoundCloud and I moved back from beautiful, boring San Francisco to cheap ‘n ass-kicking Berlin. So I thought I better come up with a plan to keep myself busy before falling into the tempting convenience of being another lazy barfly in the sweet, sweet bars of Kreuzberg. Introducing Operation:

SoundCloud Game Cards

At my old job we had the fun fact going around that I’m the CEO & Founder of the startup and I guess now it is the right time to turn that joke into reality. So the plan is to pivot into my own personal incubator to breed out all the random hacks and apps I build and keep track of it on this blog.

In the last months I’ve collected already a bunch of ideas and prototypes that are all driving me insane and itching to be build and released. To scratch that itch and kick things of right I’ll dedicate each of the next 6 weeks to finish one of these projects. The goal is to build, polish, get them released and then explain the idea, the process and the result on this blog. The projects will be anything from totally worthless but fun to useful and revolutionary (cough). Everything goes, as long as something is out and gets documented on this blog at the end of each week.

This weeks idea is surprisingly serious: Build a beautiful and social consumer-experience around Wikipedia. Come back next tuesday to see how that worked out or simply subscribe to this blog or do the twitter thing via @johanneswagener.