After four days of back to back action and three nights of very short sleep, IPFS Camp 2019 is a wrap. Unfortunately not everyone could be there in person, so I wanted to capture some notes to share for everyone.
The structure of the event wasn’t a traditional conference, like those that have a couple of tracks and prepared presentations. While there was some of that, Camp had a bigger on creating and capturing ideas and feedback from small groups working on specific topics. This was done through Poster Sessions (where posters are created on the spot to share ideas not selected via abstract submissions), Lightning Talks, and Deep Dives. There were also educational courses (both core and electives such as the Textile course), sci-fi fair booths, unconf sessions, and a surprising amount of karaoke. Many of the topics where new information was created were (and continue to be) captured on the shared GitHub repo, so it’s worth browsing and reading through topics that you are interested in.
People at the event were so engaged. I for one, had a hard time finding room to sleep. Mornings jumped right into group breakfast, followed by content; evenings ended with group dinners that went right into some of the best conversations I’ve had all year about the future of this space and how different teams were trying to solve it. Which reminds me of another significant part of this event, the number of teams that are trying to solve similar problems that are working in collaboration instead of competition, it’s incredible, and very energizing.
IPFS to solve core use-cases of Internet scale.
A topic that has been emerging over the last few months is the idea that IPFS can make software distribution on the web more robust and resilient. What does that mean? It means that IPFS can become the defacto network to distribute software libraries that are often critical to many apps we use on the web. At Textile, we’ve been fans of this idea since GX, but now IPFS plans to take what they’ve learned in small experiments like that, and build solutions for the whole web. Already they have an IPFS hosted mirror of the NPM registry and plan to improve and expand that service.
More Camp Links:
- Package Managers: What now / What next (deep dive)
All about IPFS
There are parts of the general community that, when you talk about IPFS, have it very linked to Filecoin. What I saw at the IPFS Camp was that there is actually a pretty clear division, and that division is evident in conversation, in practice, and implementation. Filecoin was not the focus of any projects I spoke with, and only came up in a few general conversations. To me, that signals that IPFS is going to be a robust and independent technology for as long as I can see, and the success of Filecoin will be independent from the success of the IPFS network.
Improvements in UX continue and continue to pay off
User-experience on IPFS is a many layered problem. We have the developer UX: how do we make it easy for apps, services, and networks to adopt IPFS and related protocols. Then we have the end-user UX: how do we help the users of the internet easily access and understand the benefits of IPFS in their daily use. They are connected problems but are solved with different approaches. What I saw was that the UX is improving on both fronts and that the payoffs are starting to materialize.
On the developer front, we’ve seen the advancement of easy to use tools like Textile or Pinata that help develops easily pick up IPFS and start using it. Many of us see the work being done by Cloudflare as helping to lower that barrier even further for developers otherwise unable to dedicate the time to start using IPFS in their work.
That progress has directly fed into improvements in the user experience for end-users. For a long time, we’ve been using Textile Photos as a way to experiment and learn what that UX can look like. At IPFS Camp we saw demos from teams like Anytype, Permaweb, and Berty. Anytype, for example, created lots of buzz at the event with their new dashboard for your files, notes, and lists. Permaweb gave a great lightening talk on why we need to rethink our relationship with data on the www, and Berty impressed everyone with their progress on p2p over low energy Bluetooth.
With projects like Anytype and Permaweb already building on Textile, we knew there were even more teams we wanted to connect with at the event. What we found at our packed room workshop was that far more teams than we expected are trying to learn how to deploy apps on IPFS now. It’s about to get very exciting out there.
More Camp Links:
- Building DApps with Textile, the iCloud for the DWeb (workshop)
- Developing Apps with IPFS API (workshop)
- PeerPad Internals (poster)
- Textile Internals (poster)
- Offline First (deep dive)
- IPFS on Mobile (deep dive)
Bridging the gap(s)
It felt like because IPFS Camp was so focused on problem-solving, there weren’t many fluff presentations or conversations. This pragmatic grounding meant that there were many conversations about how we can bridge the WWW of today — dominated by centralized APIs and a handful of popular web browsers — with the web we believe we can (and should) build. One of the highlights for me was the “Making the Browser a True User Agent” elective course, where we explored how to get the browser to work on our behalf more. Some proposed approaches worth looking into included Lunet and libdweb.
There were other topics around bridging this web gap, including explorations of how to simplify website hosting on IPFS, how to get IPFS and libp2p into more mobile applications, and how to speed up IPNS across the network.
More Camp Links:
- Making the Browser a True User Agent (workshop)
- Managing Datasets with QRI (workshop)
- Deploying IPFS Infrastructure (workshop)
- IPFS for Websites (deep dive)
Still more to solve
IPFS isn’t complete, and it’s a mistake to think it ever will be. One of the goals of IPFS that Juan Benet tries to remind us of often is that IPFS aims to create a network that is up-gradable by design. There are so many exciting areas of progress inside or adjacent to IPFS that it’s hard for any one person to keep track. The Camp did a great job of giving us all a rapid tour of those areas. I highly recommend looking over all of the poster session topics, the deep dive topics, the outcomes, and watching the videos when Protocol Labs releases them. IPFS is leading such an exciting project and a fantastic community, I’m more excited than ever for where we are headed.
Come join our Slack channel if you want to chat about anything above!