Ask HN: What are the problems you'd love to work on if you have time?
1. We can find remembered points easily when reading because of centuries of development of layout and typography (headers, paragraphs, punctuation, spaces). What are the equivalent techniques for audio, video, and hypertext that would bring them to parity with printed text?
2. What is the easiest to acquire auxiliary language (i.e., Esperanto, Interlang, Basic English) that we can build? Can the barrier be made sufficiently low that the language would be adopted by linking it to some body of widely desired material where translation cost would probably exceed learning the language?
3. What would a symbolic model of a "well trained" mathematical mind look like, and can it be made precise enough for a computer to train a human mathematician? Can we capture a large fraction of taste and intuition in this area?
4. What should a programming language meant for truly casual users (think a macro language for a word processor or a machine tool) look like? How do you make cutting and pasting code into a larger program trivial? What languages features save users from silent errors? What tools must the language support and have to make it explorable the way a GUI is today, but make snippets shareable?
5. A clear exposition of biology from the ground up for technical readers, going through the thought processes and experimental results, and articulating the universal principles that arise, even if we can't state them mathematically yet (I've got lots of notes for this one).
6. How are statistical sublanguages carved off and adopted by fields? For example, statistics in biology hasn't really changed since Fisher published Statistical Methods for Research Workers. Six Sigma is another sublanguage carved off for a certain segment of business users. What are the social and technical factors that let such a sublanguage survive separate from the ongoing study of statistics?
7. What is the abstract mathematical structure that arises from the study of centers as Christopher Alexander defines it in 'The Nature of Order'? It has something to do with equivalence classes of sublattices of the set of all subsets of space, where the equivalence relation is something like a homotopy but tied into basins that have strong coherence. What is the equivalence relation, and what are the mathematical properties of the structures that result?
At the moment these are the problems that I keep in my head, and file away anything that seems like it might help on any of them. What are yours?