I blame Bret Victor. He wrote an influential paper that was longish and people read up to the "last value predictor" section, had an a-ha moment, went to implem

I blame Bret Victor. He wrote an influential paper that was longish and people read up to the "last value predictor" section, had an a-ha moment, went to implement it, and ignored the rest. He hasn't update the paper since so no one knows what else to do. The marginal improvement was to take the last value, relate it to a bunch of other people whose last value was the same, and show me their next values.