2:58:02beachBike: Maybe, but I think this problem has to do with not assigning the correct values environment and dynamic environment to the new instructions.
2:59:56beachSo, I have another chapter to edit from Dyomkin. And my colleague and his two sons are coming over for lunch today.
3:40:15beachI wasn't looking forward to debugging this problem.
3:40:56beachIt occurred in a HIR program that was so big that the visualizer took so long to start that I had to interrupt it.
4:49:51beachImproved stepper with new special instructions: metamodular.com/clordane.mp4
6:00:26beachI think the next step is for the compiler to find all those BREAKPOINT-INSTRUCTIONs, and create a code object that contains all locations that can be stopped at.
6:03:27beachOh, one more thing, the boot procedure is not noticeably slowed down by having those additional instruction to interpret, and by having to call Clordane in the interpretation method for them.
7:54:06ck_Shinmera: do you know whether quicklisp needs an issue posted to update float-features?
7:54:38beachI started editing the latest chapter of Dyomkin's book, but I am making very slow progress.
8:03:27ck_Shinmera: ok, thanks. I thought the bit access you put in would have propagated by now
8:03:38ck_oh, but it was only a week ago! seems like forever
8:38:05ck_beach: I was momentarily startled by the use of := in the key-value chapter you're probably reviewing right now. What do you think about that choice?