Thank you, I remember your participation back in that era. The main useful thing that emerged somewhat recently is the system called 'spawngrid' which leverages fossil rootscores, venti replication, and independent namespaces provided by ANTS to spawn independent plan 9 environments on demand and make them available on multiple nodes. It was a bit too ambitious for the real world so I'm going to stop wasting the money on running the unused servers but it did work pretty well in the technical sense. The lighter and simpler public grid will continue because it has an active user community.

I also developed a lot of interest in more theoretical computing topics which I may or may not be able to bring toward Plan 9 namespaces at some point, but I don't have the time to summarize that here, the blog on my site covers some of the ideas.

It's about two things mainly: updating the design of the distributed architecture of Plan 9 to be less centralized, and using Plan 9 for ad-hoc resource sharing and communication for hobbyists. These things overlap because I provide grid resources for people to use which run on ANTS-powered servers/architecture.

No, its not wise from a security perspective, the grid operates in a very fragile zone of 'nobody except a small number of plan 9 users knows or cares about this and there is no incentive for malicious behavior' - we have been very lucky in that it seems that using Plan 9 and taking the time to learn how to connect to the grid seems to filter out people who just want to be destructive or malicious. At some point it may prove to be a problem but we have gotten along ok for a few years. Users also have the choice of what services they use and some users dont use the plumber for auto-sharing in this way.

It's mostly hobbyist Plan 9 software development. There is some shared disk storage and people upload code they are working on for other people to test and help with, so it is an active workspace for Plan 9 development. There is also a lot of general discussion of computer related topics in the chat, sometimes with a rather abstract mathematical focus on limits of computation and dependent type theories and the logic of infinity, with only a vague connection to existing Plan 9 software - far out brainstorming.