Hacker News new | past | comments | ask | show | jobs | submit login

The problem with inventing languages is that we don't fully understand natural languages. For example, there is no satisfactory formal semantics for natural languages. Categorical grammar and Montague grammar are some inroads into formal semantics, but they don't get very far. The work of Roger Schank got further in terms of expressiveness, but then fizzled out, probably because it was ad hoc and lacked an underlying mathematical structure. So inventing a language is working in the dark.

There are some natural languages which were in part designed in a top down manner, like Mandarin and Indonesian. But still, on the scale of natural to artificial, these are much closer to natural languages. They were both based on some other natural language(s), borrowing words and grammar from them.




So inventing languages is a good thing, as a research topic into how natural languages work (with plenty of provisos, of course).

And, as a human nature hack, marketing them as tools for world-peace, etc. makes them much more likely to be adopted by idealistic humans who may have the motivation to actually test them for a few decades...


>So inventing languages is a good thing, as a research topic into how natural languages work (with plenty of provisos, of course).

That might be your opinion, I don't think it's a useful research project. Are you aware of any insights into natural language that have been arrived at via artificial languages?

And, as a human nature hack, marketing them as tools for world-peace, etc. makes them much more likely to be adopted by idealistic humans who may have the motivation to actually test them for a few decades...

That seems very cynical. Why do you think artificial languages are so important that misleading marketing is justified?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: