Gameplay vs. Education vs. Reality – Work in Progress!
A response to Christian Sandvig's extensive but harsh comment on Data Dealer
Thanks for your extensive feedback, also for all the positive points you made. You're one of the first reviewers to take such a close look at our game design and you've mentioned many things we've been discussing intensely since we started the development of Data Dealer back in 2011.
In general we didn't want to develop a complex data mining simulation game which offers a 100% accurate representation of today's personal data ecosystem. We had to simplify things to achieve what we wanted to achieve: Creating an easily accessible, casual game – which will attract people who didn't previously engage in issues of personal data and privacy so much. This translation of our topic into a simple, casual game design was one of the hardest parts during development. We repeatedly had days and weeks of discussion about these issues and we've changed some concepts several times. Besides being a small nonprofit group who started to develop this game without any money (and alongside our jobs) this is one of the main reasons it took us such a long time to develop it.
I. Demo vs. full version
I believe some of your acute critique is very true for our demo version, but not for the full version we're planning to finish in the next months:
- The selection of data sources, attributes and clients in the demo version is FAR from perfect, yes. For the demo we just put in some of the many ideas we had collected. This also explains why some aspects are over-emphasized, others under-emphasized, others completely missing. In the full version (it's already 80% finished) you'll have more than 100 database attributes, and of course credit card information as well as browser cookies will be included. And there will be much more variation in the kinds of companies and online ventures you'll be able to run as a player, e.g. public records crawlers, phone surveys, product registration cards, smartphone apps, self tracking apps and of course your own Smoogle, Tracebook and online advertising networks! And there will be many more kinds of clients as well.
- You' wrote that algorithms are not being covered at all in our game and mentioned the "absence of any depiction of the combination of disparate data to produce new insights or situations". And you're right, this is completely missing in the demo. But the good news is: It's already included in the current internal development version! There will be different ways of combining attributes. A very simple example: You'll be able to increase the amount of age and gender information by "analyzing" your profile's first names. There will be "summary" attributes (e.g. combining all kinds of identity information or summarizing all kinds of friends/network information) and score attributes (not only credit score, but also health score and so on). To refer to your "political attitudes" example: In the full version you'll collect the books people are buying and aggregate this information to create a "political attitudes" attribute in your database. In the full version it will be possible to "play" the database.
- In the full version there will be several more ways to get into hot water as a player when acting too carelessly: Privacy regulations, whistle blowers (ehm) and more. And many additional ways to respond to those incidents as a player. You wrote: "It even feels like the makers of Data Dealer are trying to demean themselves". I believe that this ironic turnaround of blaming privacy advocates in our game could be experienced quite differently.
II. The Data Dealer's underground economy
You criticized the "criminal underworld" part of our game. You wrote that no legitimate data mining company would use personal data from sources like that. First, some of them certainly do. In Austria we recently had a case where a very small consumer data agency had systematically paid money to people working at court to make them hand over the data of bankrupted people. The small agency sold this information to a bigger, more serious agency. The boss of the small agency was convicted, the bigger one not. Or let's take for example Choicepoint (now: "Lexis Nexis Risk Solutions"): They've bought so many data collections and whole companies (including their data collections), nobody exactly knows where all their data is from. It's not transparent at all.
But that's not the main point. What we want to show by putting data sources like "Uncle Enzo" (refers to the Sony Playstation online network hack 2 years ago) into our game: Everywhere massive amounts of data are stored, there exists a risk of abuse. On the one hand personal data is abused (semi-)legally by companies or governments, on the other hand there is the risk of abuse through negligence, security holes, disloyal employees, or hacks. So, what could be the motivations for people to abuse personal data? How does social engineering work? But this part won't be as important in the final version.
III. What else?
Moreover you criticized that we included things like celebrity endorsements and ad campaigns. It's a game! When players e.g. can do "fake postings" to advertise their companies ("seeding") then this is not about personal data, right. "Data Dealer" is not just about personal data – there is much more to it...
Finally you wrote that one of the major problems in educational game design is how to offer both effective gameplay and effective education. And you're certainly right. We want both. But it's still a big challenge. We're the first ones trying to make a game about "personal data", so it's crucial for us to get as much feedback as we can.
By the way: Investigating the mistakes in the simplified model we use could maybe even play an important part in our game. Many gamers are used to analyze every tiny single detail of the games they're playing. We've already had the idea to explicitly ask players to discuss what might be "wrong" in our game compared to reality – as a part of the game!
In general "Data Dealer" aims to raise awareness of, but also to provoke conversation about surveillance, personal data & privacy in a new way. But we also want it to be a fun game! It's not easy for us to achieve that with our really small budget. But we're doing the best we can. If you have further ideas on what to change or add, please let us know! We'd be excited (and grateful!) to hear them.
I hope this helps to clarify some of your concerns. If interested: Here's a small report on our background research, here's a TEDx talk I did about our project and here's an interview I recently did with ThinkProgress.
Wolfie Christl, Data Dealer