I’ve been moving things around on the SG portfolio all day, with a little flurry of announcements, some good, some regrettably retiring projects I either don’t have time to work on anymore, or that I feel are holding back the whole effort.
The problem I’m working on is that I have a tendency to want to build all the pieces, which is what the goal is, but, I will sometimes wait until those foundational pieces are built before building the rest of the product when, if I had used an out of box component for those pieces, I’d be able to retrofit my components in when they are built later, with a more built out portfolio in the long run.
So, that’s what I’m going to do. No more projects will be held up on IDM/A or SMQ/S implementation. I’ll use RabbitMQ and FreeIPA for those components until I have time to circle back and build them out.
I’ve decided to pull SMQ/S from the active projects list at SILO GROUP.
That problem was one of data consistency. For a high transaction volume distributed system like that, there are tradeoffs that the designer is forced to make, and none of them are great.
I found that to really build what I was envisioning there I would have to create some kind of amalgamation of Chord and Kademlia and resolve some pretty old problems that, while fascinating, will take years to do.
So, I’m going to make it a fun personal project and stop jamming up the SG portfolio progress with it.
I’ve also decided to shelve the SAL/M project from the community engagement portfolio at SG.
It was an ambitious experiment but it’s just going to take too many years to develop and it’s one of those “nice to have” projects that are jamming up the whole portfolio.