It's been a week, so time for a writeup of what went down at DWS. The press below covers some details, but I'm going to talk about the _feels_.
tl;dr - Electric atmosphere, technology on the cusp, very unclear future.
I found out about this too late to attend the first day, but I followed along via the live stream while reading up on the underlying technologies and chatting with attendees using federation features of Slack.
I went in person for day 2 and immediately felt the deja vu. O'Reilly FOOCamp meets early Google I/O meets the original GopherCon. You had wise sages (or as Wendy Hanamura put it _Orignal Gangstas_) working side-by-side with the new blockchain Gangstas. The only thing missing was a game of werewolf.
The breakout sessions were tech heavy but the crowd didn't need their hand held to pull down git repos and run/modify code. Many quick demos were created.
Lightning talks (available online) had thoughtful live questions and were broad enough to cover both the underlying technology and the potential results of applying it to society. I appreciated the inclusivity and diversity.
That said the ghosts of breathless tech conferences past were all there:
*Mobile* No real demos on phones. Many talks started with 'enter this on the command line'. That said most of the p2p systems on display have really good mobile properties: eventual consistency. offline sync, etc.
*UX* Little to no focus on UI/Usability problems. There was some discussion on the "Why PGP failed" talk, and passing references here and there. But very little about how this tech would be better for users.
*Economics* This new tech is competing with dirt-cheap VPS hosts out there and a generation of software designed for centralized client-server. Privacy and long-term effects on the ecosystem are low on users feature list. With the fintech bubble about to pop, who's going to fund the development; let alone the ongoing governance. Will new bitcoin funding models be the solution? Will the incumbents embrace or reject?
*Complexity* writing cryptographic serverless code is difficult and it's easy to make mistakes. libp2p is a good start, but the tooling isn't there yet.
Despite these serious issues this is the most excited I've been about our technology space in a long time. The electricity and optimism about what might emerge from this soup of technologies was palpable. I have hope that people won't want to repeat past mistakes, and that the new stack can achieve some 10x gains. Here's some initial thoughts:
- What if your phone could pull down entire sites for use offline and have deltas propagated when connected?
- How about having all of your physical devices syncing between themselves instead of up and down to the cloud.
- How about a better UI for managing your identity public/private keys?
- What about your OnHub being your persistent home on the network?
- What about being able to archive and 'play back' entire web sites like you would a git repo?
- How about having easy micropayments as a way to break free from our current ad-supported mess?
- How about Android APKs that travel from device to device with the security of knowing that you're running the exact same code as everyone else.
... more to come.