Does anyone have any context or background on this paper and why its just now appearing in HN? Is this the seminal paper on random sampling? Has this recently become contextually relevant for some reason, or just an old gem that people are rediscovering?
Any background info is appreciated!