1. 4
  1.  

  2. 2

    I love these sorts of papers: reasonably short explanations of interesting algorithms. There are lots of these papers from the 60s and 70s. I’ve found they became more long-winded later on. (Maybe I’m suffering from too much nostalgia. I’ve no idea.)

    The reason I love them is that they are great programming exercises, especially when you want to try using a different language. They tend to be sufficiently complex so that the exercise it not trivial, yet not too large as to warrant having to learn extensive APIs.

    1. 3

      rpg’s 2012 reflections on the paradigm shift in programming language papers ,as reflected in Bracha & Cook’s choice of what to focus on and what to elide in their 1990 mixins paper, is probably relevant to that. After about 1985 or 1990, I suspect that the kinds of papers you like to read became unpublishable; rpg says:

      In the 1990s it seemed to me that scientists in the programming community pulled back the welcome mat from engineers. I noticed that the kinds of papers that could be published then were different from when I was publishing at a good clip in the 1980s. By the mid-1990s it was clear that a paper like Moon’s from 1986 [describing the Flavors OO system CLOS was based on] would never be published at OOPSLA or any other programming language conference.

      It would be nice to have a functioning world of hackademia where this kind of thing can get published again, even if not in traditional paper form, in order to get engineers to put in the work needed to produce really understandable descriptions of their inventions. What would that look like? Like CodeCon? Usenix? GitHub? Some kind of combination?

      1. 1

        I don’t have an answer, but it’s a good question.