Hi Brittany!
Yeah, I was one of the co-editors of the Solid Protocol, and I have moved on. That is not to say that I didn’t like Solid, I had been involved in a lot of what lead up to Solid, the whole Semantic Web thing was something I was involved in since 1998. To me, Solid was what could finally make Semantic Web useful, and I also viewed it as a last-ditch effort to do so, we had been doing it for 20 years without much coming out if it, and I wasn’t prepared to let it go for much longer.
Solid was to me kind of a “minimal add-on to the Web” to allow for personal control of data, a UNIX for the Web. It had an architecture where you’d use the verbs of HTTP to manipulate the resources and it had private and group data control from the outset.
Unfortunately, it didn’t work out. Some of us who was involved early on have been thinking about writing up a post-mortem, but I’m not sure. Let me just point to two very bad decisions that happened: 1) Deciding not to pursue a community effort to build a new server. We had quite a lot of people wanting to do that together, instead Tim decided to try to salvage the failing NSS implementation (and I was tasked with doing it), and when that didn’t work out, an academic community was contracted to do it, but they wanted a research system. 2) Inrupt’s decision to make a server their main product. This resulted in that every significant decision was made in-house, completely opposed to the public communication which stressed openness. Solid became a commonswashing effort, and continues to be so to this day. Since Inrupt focused on Enterprise customers, every effort was made to make the permissioning system focused on enterprise use cases. But setting permissions to data in the enterprise is a expert focused task, resulting in a system that would only give you control over your data if you were an expert. That is not to say that what we had was good either, it is to say that it completely sucked the oxygen out of the room for every other effort. I wrote a lengthycomment on Leigh Dodds blog too.
AT Protocol has a very different focus, its focus is not on the Web and its infrastructure as it were, its focus is on the blocks of data, and how you can trust those data even when you don’t necessarily trust the underlying infrastructure – an important consideration. It didn’t bring in the permissioning on data from day 1, but that’s an effort that’s going on now. And it doesn’t use the verbs of the HTTP protocol for resource manipulation, it instead uses the RPC paradigm. I’m not a fan, but at this point in time, I believe that this is not a fight worth taking.
So, in terms of alignment… I don’t know. I guess Solid could go somewhere, but it seems pretty dead by now apart from Inrupt’s commercial endeavors, which also doesn’t seem to amount to much. It has had a decade to take off, and it hasn’t. I’m pretty saddened by the fact, it was such a big part of my life for so many years.
To me, the Lexicons efforts of AT Proto look a bit like something that hasn’t met yet the challenges that RDF failed at a couple of decades ago. People just don’t tend to agree much upon semantics, despite every effort to make it clear and well defined. 
I believe there are lessons that could help AT Proto not fail the same way RDF failed.
Also, I hope that the mistakes made around private data in Solid would not be made when designing the system here.
There might be ideas around stream query languages that never made it anywhere on Solid that could perhaps come in querying the firehose.
So, in conclusion, there isn’t a lot in terms of actual tech, but there could be many ideas that didn’t make it to fruition on Solid that could make it here. And overall, the governance must be much better here than it was over there.
)
