Last week I attended the Decentralized Web Summit at the Internet Archive in San Francisco. I was also part of the planning committee for the schedule and topics. As an organiser, I helped facilitate a discussion on security during the builders day (which was invite only and took place one day before the public days) and moderated a lightning talk session on the last day. As a rep for MaidSafe, I attended the builders day and got to meet a bunch of extremely interesting people as part of it (and tell them about/update them on SAFE), was on a panel about security in decentralized networks on the second day (and hosting 1-hour booths to discuss SAFE on each of those days) and finally hosted a 90 minute workshop on the last day.
It was an exciting, yet exhausting week. Glad @frabrunelle was there to back me up at times, too. So thanks for that Francis.
The workshop was fun but it mostly was me describing the network consensus. I did show some basic launcher demos at the end but will aim to integrate that more in the future. As far as I could tell, the group who came seemed really interested in talking/asking about the design of the network. I think I would have seen a more general crowd had the session taken place in the afternoon but overall, the design seemed to sink in pretty well and folks walked away with reasons to research more for themselves and check out the launcher more. I also gave short demos during the “science fair” portion of the first two days which essentially allowed people to ask questions and talk about specific projects.
I went to the Freenet workshop which took place right after the one I did and installed it on my Mac. It was cool to meet more Freenet developers. They’re really a great group of intelligent people and have really interesting experiences with building a dark net that we could probably learn from. We should hire them all.
I walked away with better knowledge and understanding on a couple things: what properties SAFE brings to the greater community of decentralized technologies and importance of the human element when building these systems. To the first point, there are relatively minimal attempts at a fully autonomous network to provide data and node integrity properties and also not too many focused on anonymity. The ability for the network to not only manage itself in terms of keeping connections and relay data but also make sure that data is, in fact, continuously stored on the network is unique. There’s a ton of projects which are bittorrent + bitcoin (or some dht + blockchain variant) which is a worthy avenue to explore but as a result, there’s a lot of overlap of features and will inherently include scaling problems of blockchains and Sybil vulnerabilities of standard dht implementations. Some alternatives to autonomous data integrity I see are incentive based (Storj, keep paying data host nodes) and trust based (a la friend to friend style in Freenet and Tahoe LAFS). Regarding anonymity… I really only saw SAFE and Tor representing that priority at the event.
To the second point about the human element… I basically heard some horror stories from Freenet about the worst of the worst kinds of people using Freenet for sharing the worst imaginable things. This is not to say the foundation of a decentralized web requires preventing this, but rather that technology literally cannot prevent this and social strategies are required to block or filter this sort of content with the help of third party tools (ie. blacklisting).
There was a ton of interest in the realm of content addressing and more and more people understanding the natural solution to locating data within a decentralized web is to search via content hash rather than a location where data is stored. Most projects are using this method and if you haven’t heard about Van Jacobson’s named data networking which is essentially building a standard for this sort of thing to replace IP.
Finally, I think there’s a lot of potential for collaboration with the Internet Archive community. They are extremely interested in the concept of reader privacy which is currently dependent on using Tor to access their archives. SAFE will certainly be able to provide a viable alternative to this and I really look forward to have this be a major use case. There’s already a few tools on https://archivelab.org/ which seem relatively trivial to also give an option to reading from a decentralized network such as SAFE.
Photos from the event here: https://archive.org/details/DecentralizedWeb20160607Images
Videos here: https://archive.org/details/decentralizedwebsummit2016
Some articles about the summit:
New York Times – "The Web’s Creator Looks to Reinvent It"
EFF – "Locking the Web Open: Dispatches from Morning One of the Decentralized Web Summit"
The Next Web – "The Inventor of the Web thinks we need a new one that governments can’t control"
Business Times – "Tim Berners-Lee wants to reinvent the web"
Gizmodo – "The Web’s Creator Now Wants to Unfuck It"
The Inquirer – "Sir Tim Berners-Lee: Internet has become ‘world’s largest surveillance network’"
Breitbart – "The Inventor Of The Web Wants To Give Power Back To The Users"
Fortune – "Here’s How Two Visionaries Want to Fix a Broken Internet"
Boing Boing – “How we will keep the Decentralized Web decentralized: my talk from the Decentralized Web Summit” <-- Edit: Highly recommended