Tegenlicht: Money & Speed — a touch documentary

Posted: January 31st, 2011 | Author: | Filed under: Publications | Tags: , , , , , , , , , | 3 Comments »

Last Tuesday Dutch broadcasting corporation VPRO launched the iPad documentary “Money & Speed” (free now in the App Store) they had been working on on and off for a large part of the past year with catalogtree (design), Noodlewerk and Systemantics (development). I was present and got an advance copy of the documentary to review here.


The documentary deals with the flash crash of May 6th, 2010 when the black box trading operations on Wall Street went haywire and dropped the index 900 points to recover just minutes later. I’d already read about the possibility of such events from Kevin Slavin’s January 2010 Social Computing Summit presentation which has been noted down and blogged about by Michal Migurski. Both are recommended reading.

We have seen various troubles with the stock exchanges in the past year and this event especially seems one worth investigating because it exemplifies the complexity in todays exchanges and the total lack of control humans have over the process.

Another reference which I thought was important is the concept of the macrospcope by John Thackara which I first heard used and expanded upon in Matt Webb’s Reboot keynote. The related reading on the BERG blog is interesting but primarily the definition by Thackara: “A macroscope is something that helps us see what the aggregation of many small actions looks like when added together.”

So let’s see.


For an international audience the concept of Tegenlicht may need some explanation. Tegenlicht is a documentary that examines world events by interviewing experts interspersing the interviews with visuals and a voice-over to create a dramatic storyline. The app contains the entire show in high quality which is in part why it is so heavy.

The shelf life of Tegenlicht documentaries is quite high. For another concept we recently rewatched their 5 year old documentary ‘De dag dat de dollar valt’ (Eng. The Day the Dollar Fell) because it was still relevant and interesting. The 45 minute length with drawn out shots can be a bit taxing for today’s YouTube attention spans, but byte-sized information is not their game. TED is much better at that. They concern themselves with the documentary as a dramatic art form that needs to engross its audience.


Given that concept —highly traditional television, cinema almost— it is interesting how you would interject/overlay/add interactive features into the narrative whole. This was touched upon briefly in the presentation, but that is not what this app concerns itself with. You can view the documentary and jump back and forth through the various segments while additional content is presented for your perusal.

It is clear though that traditional broadcasters are still very much struggling not only with the internet but also with the spectrum of television, cell phone, laptop, iPad and the locus of interactivity (if there is any interactivity). Tweede scherm is one such recently award-winning concept that displays supplementary information to add context to the main experience on the large screen.

Infographics & Visualization

The app is decked out with a nice cadre of infographics and visualizations and those are indeed its most important selling point. There is a list of them on the home screen. Several time series, an animated display, a multi-layered map overlay and a world map with live stock updates:

The visualization that gets the most emphasis and also is used often in the documentary is a time series display of the stock price around the crash:
Stock start

It starts out nice and flat with a display of trading velocity (not quantity) and pricing information along with the time. There is a global display with scrubber that you can use to navigate over the entire run of the data and the crash is nicely colour coded. The vertical scale is a bit confusing as the one above does stop at zero (and then goes on for a bit more) but the other ones don’t. So confusing in fact that in the documentary one woman remarks ‘Apple is going to zero.’ which it is in fact not.

And then it goes South:

Stock Crash

You can see the drop in prices and the variability. It becomes even more clear if you also enable the Bid & Ask information which is available as an overlay and shows you the differences between the prices asked and bid for the stock at that moment:

Stock Crash with Bid & Ask

So that is a one-dimensional time series (with two extra dimensions available on request) with a beautiful presentation and animation. Another interesting piece of information is the potential locations for data centers around New York and what factors they need to take into account to carry as little risk as possible. You can see the map and the various factors involved and slice it yourself:

Data Donut

Another visualization is a map of the world (in catalogtree’s signature geographical bubble display) with semi-real time updates of the world’s exchanges and how they are doing right now:
World Exchanges detailed state

The visualizations add a lot of panache to the documentary and are aesthetically very pleasing to behold which gives them a high show and tell value.

From an information design point of view however they are underwhelming. The information density is low, it is difficult to compare several datasets and the visualizations do not offer different types of information at different zoom levels. Also: the interaction is nearly trivial.

The issue of game design and game-like experiences was touched upon during the launch event to conclude that none of the makers had a lot of expertise (or even any affinity) with games. That is unfortunate because game design with its experience in dealing with highly interactive experiences of high density information spaces can add a lot to a data visualization.


The whole issue of interactive television and how to combine a long dramatic form with visualizations (and what kind) seems to be a difficult one to solve and not the one being tackled here. As Erwin mentions in his review in Bright, added value is a highly pressing issue when it comes to traditional media trying to produce content for the iPad. That is exactly what this is: a nice packaging of a traditional television program with interactive features in a combination that will most probably remain interesting and relevant in the future.

The documentary is very attractively presented on the iPad. The extra content especially is more prominent than normal when it would have been put on a back page somewhere on the website. The video playback is also one of the first cases in Dutch broadcasting where the presentation is native to the device.

photo 2.PNG

Another benefit is that this experiment can probably live on as a packaging format for other documentaries. The richness of the experience combined with the quality of the video, the pairing of additional content and the clear payment model make a lot of sense for something which already has high production costs. With magazines and newspapers you are adding a lot of extra weight to something that has a low margin and is ephemeral (daily, weekly). A well produced documentary such as Tegenlicht can live on for a long time and this seems to be a more suitable incarnation for that than most.

The VPRO has brought forth a great release of a beautiful looking product. Tegenlicht Money & Speed is out in the App Store in a Dutch version now with an internationalized release forthcoming.

Personal Mapping Platforms

Posted: January 18th, 2011 | Author: | Filed under: Research, Statlas | Tags: , , , , , | 4 Comments »

For the project Statlas we are looking into making a personal mapping platform for journalists. We submitted the grant proposal for this almost half a year ago and the idea had been alive for far longer (we started about this time last year).

It’s good to sea that there is a wider trend in consumer mapping platforms right when we are underway with ours. Here’s a brief survey of the ones we found during a cursory examination. There are bound to be more. If you know them, please let us know in the comments.


ArcGIS has a mapping platform based probably on the ArcGIS server, a paid for cloud mapping platform.

ArcGIS - My Map

Looks nice, like a web based version of Google Maps combined with Google Earth with all the different overlays you can put on there. I tried to create a map and share it on Facebook which oddly enough did not work. The sharing, embedding and standalone map versions do look well thought out but if they don’t work they’re probably not tested well.

View Larger Map


ESRI the company behind ArcGIS has another ‘Make a map’ tool which is a lot more restricted but because of that provides a clearer experience.

Make a Map | Free Embeddable Maps | Embed Map Web Page | Embedded Maps

This doesn’t offer a ridiculous amount of options, but it is very clear and nicely done and the sharing options are also very straight forward. An embed of that map is below:


Dotspotting is Stamen‘s platform for putting dots on a map currently in its ‘SUPER ALPHA-BETA-DISCO-BALL VERSION’. As they describe it, it’s intended to make the process of visualizing city data easier, more open and more robust.

That is pretty much the same reasons we started on this road in the first place. Mapping and data literacy are necessary in web development as well as the other way around: web literacy is necessary for those that make the heavy-duty maps. The two need to meet to create the applications and ease of use we are looking for.

Dotspotting - "My Flickr Photos (Sept 2010)", a sheet of dots by straup

A script to export my Foursquare checkins in an easy way and create a sheet with those is forthcoming. Anyway, Statlas is best described as that: a way to project values onto regions and enable people to play with that dynamic.

Weet meer

Weet Meer got launched very recently in a beta release and is limitedly available up until next month. It does a decent job in displaying the statistics offered by the CBS and offers some statistical relations and tools to compare things with.

Weetmeer.nl - Aantal inwoners

That is a brief overview of what is already out there. We’re glad that we have hit a nice timing to be able to develop ours and fulfill an actual need out there: to be easily able to make maps of a set of values to a group of regions.

Glanceable news using the new NOS API

Posted: January 4th, 2011 | Author: | Filed under: Manifestations | Tags: , , , | 2 Comments »

I had been aware of the fact that our Dutch public broadcaster NOS was going to open up their news using an API. During the run up to this I gave them a couple of pointers concerning what we think are great APIs and must-haves for this kind of thing.

Then they launched open.nos.nl to a limited amount of people at first to play around with and see the initial beginnings. It is quite modest right now with three end-points and not a lot of functionality or content to play with, but it’s a solid beginning and it is clearly meant to grow.

Ons Nieuws — Het laatste nieuws van NOS.nl

Busy though we are right now, I got a bit of an itch on Monday and whipped together this simple example: Ons Nieuws which is a glanceable news display of the latest news on NOS.nl meant to replace the tt101 screens that run at pretty much every news/publicity department.

Ons Nieuws is heavily influenced among others by: DEXTR. Glanceability and ambiance seems to be getting back in vogue also with apps such as Trickle. This thing we whipped up is really a very simple proof of concept version mainly meant to exercise the API, but it may lead to more serious applications.

Bloggers, transparency and our Catch-22

Posted: January 3rd, 2011 | Author: | Filed under: Events, Publications | Tags: , , , , , | 2 Comments »

This opinion (in Dutch) by Ernst-Jan Pfauth about how bloggers and other people online can help make the government more transparent is in line with most current thinking on open data. It is good that message is being spread wide and far by influential people in different spheres.

Some projects we have been involved on in the past (Vervuilingsalarm, Schoolvinder) are even name checked by Ernst-Jan which is very nice except for one thing: these are prototypes we slapped together some years ago, in less than a day with minimal support AND these are still the best examples we have in the Netherlands of proper open data projects? It’s a testament to Google App Engine that these are still running and it’s a testament to the climate towards these kind of projects in the Netherlands that they have found little to no backing.

We are running into this problem when we are promoting open data in the Netherlands. It’s difficult to find sustainable, complete, thriving examples of applications built on open data like there are ample of in the USA and in the UK. So they ask: “Do you have any good examples of apps that will convince us why we should open up the data?” “Well,” we reply, “obviously if you haven’t opened much up yet, there won’t be too many good Dutch examples yet. But look at these from abroad!”

It’s a rather annoying Catch-22 that is used to justify institutional inertia. So yes, the growing mindshare around the subject is good and it’s flattering that the stuff we built is being used as the leading examples, but we will have to do better still. Stay tuned. We’re setting a bunch of stuff in motion that will nudge the status quo forward by a fair amount.