Totally agree. I tried the JIT approach; I could never get it to work, and I've never seen anyone else do it either. The wisdom has always been to keep everything flowing and accept spoilage (or kludge it with lots of bots and then move on). This patch makes it a more feasible, I think.
I'd love to see splitter filtering by freshness (e.g. nutrients at >=80% freshness) but I don't think that's in the cards.
I just use small buffers. I don't know if that's technically JIT or not, but a smaller restricted buffer with high throughput works great for me.
> keep everything flowing and accept spoilage Yeah, I still have to do this too though.