Diaspora* - One of the Founders tells the Background Story

What would an alternative Social Networking platform to Facebook look like. How would it respect its users´ right for privacy and allow them to self-control their data over the long term? What would its architecture be and how could people help to improve it? So far, no other project has engaged as intensively into this terrain as the team behind Diaspora*.

Maxwell Salzberg, one of the founders, tells the story of how he and his friends got involved in this project of a “personally controlled, do-it-all, open-source social network” after a talk by FSF lawer Eben Moglen at their university in New York. He answers questions about their technical infrastrucutre, first problems they have been run into and many more. “Our goal is to empower people with their data.”

DISCLAIMER The transcript for this file is not yet finished. If you want to help, please download the file and send us more transcribed text to info@transformingfreedom.org. It will help to get this speaker / project to be better known and researched in the future. If you want you name or project to be mentioned, please say so. DISCLAIMER

  • Date of recording: Sun, 2010-09-05
  • Language(s) spoken:

hypervideo Player

Download this file

Illona Koglin: Okay Maxwell, I guess hundreds of people have already asked you and I apologize for asking you again: how did you all start, how did you come to get the idea of Diaspora*?

Maxwell Salzberg (MS): It kind of started when my friends and I from school, - we always got  into trouble in one way or another and we started looking for something to do next, had finished one of our, sort of, primary projects, which had been building a maker bot. Actually, Bree Pettis is showing one over there, go check it out. (Pointing to TELEINTERNET area at the Ars Electronica 2010, curated by Aram Bartholl).
As we were looking at Eben Moglen, who is a lawyer at the Software Freedom Law Center. He wrote the GPL 3, which is the new version of the GPL software license. And he is a very well renounced Software Freedom Fighter. He gave a talk at my university about “Freedom in the Cloud” which really was ,-  in a way, he is a very passionate speaker, he is amazing, the man can tell a story. And, basically, he just said that: why are there no alternatives?
He used basically giving this talk (_?) to a bunch of geeks and hackers. People like myself. And __? why is it that there’s only Facebook?

01:21 Even if you wanted to be a tinfoil hat and jump through huups __? and share just like photos and ___?
There is not a good platform that we can do that on and it is not a huge technologically hard problem. Something that doesn’t make sense why there is’nt something better and really was kind of like a cold arms [?]. I think the thing that circumcurt[?] me and my friends that he did like point out, he kept a front[?] Facebook as `PHP dudats´ (ed.: simple building blocks)PHP is a programming language. Because intrinsically, e.g. sharing a status message with your friend is not the technological marvel of the world, you know, it is not really that hard. 
Facebook has ___a range of computer science PhDs because __ trillion photos. Every time you click a photo it has to look through a trillion things which is a really hard software problem but you wouldn’t actually have the problem if there wasn’t one place that had trillion photos. 
Right, so I think when we ___ looking for something to do on Friday nights like -you know- and you kind of like – we just started messing around kind of

We started of ___ that’s more distributed, that’s more about like connecting between two people and less about like [the] network you’re on.
[]As we got more into
it’s not that hard. That certain concessions you have to make. [] Eben (Moglen) is fantastic, because he is uncompromising as a figure.
But we __
once you kind of give restraints for yourself and sort of say: I am not going to worry about solving this problem right now, someone else who is really smart might come along later and solve it. With the few of those in the place like: we can build something that works in three months in the summer. So wouldn’t it be cool if instead go and get jobs? We can just work on Open Source software for some time.

So we decided – we had some sort of working code and then, I guess, our professor had a hackathon – it is a kind of where a bunch of startups come together with all their technologies and say, like, “in 24 hours we build the coolest thing where we mash up these three technologies”. And he raised extra money for us on this website kickstarter. And he was very successful – we were able to have pizza and online. It was great.

We kind of . Maybe his kickstarter, maybe a couple of hundred people [] mostly like friends or family of friends, maybe they all throw a bit of money out of way[?]
We thought, we asked for 10.000 dollars, which for 4 guys for 4 months is like – you know – not so – you know —
Then we wouldn’t have to worry about [] how would we eat and anything like that. We could just worry about coding.

And then it just went crazy. We’ve been raising over 200.000 dollars – it’s over 2000 percent we’ve asking for. Yeah, and now [] here in ars electronica in Linz.
GR1: 04:20 You’re famous!

MS: Jesus! [laughter] That’s scary!

04:25 Nina Stuhldreher: And you also have a business model? Or do you think about a business model?

MS: Our goal is to empower people with their data. That is the thing we want. That’s __aquite__ we started doing this. We haven’t really spent too much of our money. 
The only really upgrade we have from 10 to 200.000 dollars was that we got our own rooms in a ‘shitty’ appartement, right, vs. like sharing a room. (laughs)

But we have a lots of money — the first goal is to make it really awesome and make people really wanna use it. That’s really the most important thing.
I think there are some cool things around helping people, empower their data a step further. Perhaps we can [] provide some sort of business service on the software side. But I think our primary focus right now is just making it something, a thing that people need.

GR1: 05:19 And what would you say is the killer function?
MS: The killer function – I think actually the thing that’s gonna be really cool at least for the first version is we really wanna help promote sort of small group sharing
There is [] this interesting phenomenon – these are American stats, I’m sorry if this is not true [here/everywhere] – but at least in America 80 percent of all cell phone calls reach 4 people. Right? 
So there’s ___ And people have 255 Facebook friends and thier email are between 20 and 30 people. […] Why isn’t there a social software tool that’s build for like the people you actually care about? For communication with people you actually [sort of//…] see everyday but at the same time [like//…] you can enrich that relationsship with [sort of//…] a social tool. So, I think that actually kind of comes down to serve ___ privacy. Right, privacy is not about not sharing on social networks – it’s about sharing stuff. It’s about beeing really, really clear where everything is going and [and//…] not feeling you’re sort of out of control […].
That’s the spectrum. On the one hand privacy is like ‘I don’t want my mum see this __ photo of me at night’ and the other one is ‘I am a Chinese dissident and I want to make sure that I not gonna be arrested for posting this’. [//] I there is battle-hard??? Diaspora to be close to this, but we are at least at first trying to make it so – you know it’s a privacy model more based on trust rather than these crazy granular functionalitites that even Facebook now can keep me basically do everthing in terms of privacy. It’s just a matter of [like//…] figuring it out and [like//…] posting something becomes like brain surgery.
The idea is you wanna be able to post stuff and not think about it. But you don’t wanna inherently not thinking about it as sort of in a way what causes a lot of privacy problems. So, you have to sort balance, you have to…
I think the killer feature is [that sort of//…] group sharing we’re gonna implement and I think that the first version people gonna start seeing in October [2010] is very basic and stripped down.
The thoughts behind is that you really in order to do user-centred design you have to [like//…] get it out there and see people use it and see what they understand, what they don’t understand. It’s really important that that part is so blatently obvious that […] they don’t have to worry about it.
It ___ ridiculous simple.


GR1: You have been talking about [at ars electronica] some tools you use, could you please tell us a  bit more about that?
MS: Okay, how techno do you want me to dive into it?

GR1: You can get quite technical, because our readers are interested …

MS: OK. We are using Ruby on Rails, version 3, which just came out last week, but we’ve been running on the beta.

So it is Ruby on Rails 3, using the MongoDB database, which is from a company called 10Gen. It is a  new Open Source document storage database, it is really cool, because I think in the beginning it is gonna be very simple, we’re not going to leverage MongoDB as much as we can, but it can do crazy, crazy things when we get to these scenarios where you are pointing to all sorts of different content types, because it is basically schemeless. So you do not have to, sort of, predefine what it is going to be. You can just put stuff in and get stuff out, it is really, really cool.

Some of the technologies we’re using to [sort of//…] make this, sort of communication between our work our pubsoul pubub which is [sort of//…] a message hub, distributed open and free message __ hub which is actually an Open Source project from two guys at Google – Bernd Slatken and Bret Fitzpatrick.
It’s really cool piece of technology because it’s really effective but it’s also really really simple. Which is – it kind of sounds funny but generally on the Web again and again the [sort of//…] simpler dumber 09:25 protocoll has won basically every time.
It’be simple for developers to use and for people to sort of get??? time and time again. That’s sort of a proven thing. And it works.
    •    Using webfinger (http://code.google.com/p/webfinger/), which is a distributed [sort of//…] — you can annote metadata to your email, but it’s not only email, [sort of//…] disgracionary__? data. That was really cool too.
It’s very, very new. 
___finger protocol on Unix. Again, that’s good for, so I can say my email address is xyz@gmail.com and I can attach where my Diaspora seed is located on the Net. [] As long as I know my email than you can find me very easily
    •    The Zen protocol, which is a protocol, which [sort of//…] handles the problem when things are reposted so often on the Internet

This post is mine.

They’ll send you the comment.

12:40 you could run a website

every seed can have


on a technical level

MS: Everything is encrypted.
Everybody is

It’s also about verifying

So, the idea

Fundamentally the privacy model

Girl1: Acutally my question is obsolete.

MS: So silent blog(?)
across a distributed network

I also don’t want to pass

The first battle is

opt in

TN: I’m asking for a special feature
MS: Oh, version 3.
communication procedures
professional self observation tool

MS: Diaspora tracks

19:10 like opt-in thing
Letting people leverage their own data.

n to. I know t



An interview by Illona Koglin (weave.de) as well as the artist Nina Stuhldreher and Tobias Nöbauer from transformingfreedom.org. Recorded in September 2010 at Ars Electronica Festival “Repair”, Linz,