2 ms·
This is a really great answer, I appreciate you taking the time to thoughtfully address Doctorow's issues. 2.5-2.7 are really hard problems. I think that lots
by bane 12y ago
This is a really great answer, I appreciate you taking the time to thoughtfully address Doctorow's issues.
2.5-2.7 are really hard problems. I think that lots of people working in the field get lost on these by trying to achieve some sort of perfect model, or by trying to aggregate every possible option into their model, but neither of them have really been terribly satisfactory or provide the kind of subtle decision framework that humans feel comfortable with.
- dllthomas 12y ago2.5-2.7 might be resolved by decentralization. If everyone could pseudonymously tag anything, and you could ask questions about the tagging of various sets of pseudonyms, these don't seem to be really hard any more. You do need some degree of normalization, but we don't need One True Taxonomy - we can support multiple perspectives on the world. Hopefully, with proper tooling, we can find a few that are sufficiently consistent and useful.