Consider adding support for morphemes

[gustafl]

It could be interesting to pick apart lexemes into morphemes. There could be a "morpheme editor", where you select different parts of the lemma, and press buttons to highlight them as one of base, bound base, prefix, or suffix (consider adding infix too). Different colors would mark different kinds of morphemes.

This is obviously a very large addition and not something we need in a first version. But imagine you could do interesting things when the user has registered a large number number of words including the separation of morphemes,

[gustafl]

When analysing a word morphologically, the user selects the part of the word which is to be analysed. Out demands on mouse precision is greater here then when selecting words, but we can assist somewhat with whitespace and non-letter characters. When the user has made a selection, he is presented with the following choices:

• Content or Function [morpheme]
• Free or Bound [morpheme]

If the user selects Content, then Bound, a third choice appears:

• Root or Non-root

Depending on the choices made, the morpheme is given a type name, which is visible somewhere in the morpheme form header. The type names are:

• Content word (free content morpheme)
• Bound root (bound content root morpheme)
• Derivational morpheme (bound content non-root morpheme)
• Function word (free function morpheme)
• Inflectional morpheme (bound function morpheme)

[gustafl]

To my understanding, bound morphemes are not lexemes. Only content and function words are lexemes. In Lexeme, the lexeme is the atomic unit of learning, which means bound morphemes are not relevant. I see little use in collecting morphemes the way we collect lexemes, and an auto-highlight feature for morphemes would be pretty useless (even if we ignore affixes and only deal with root morphemes). This is a vocabulary building tool, and as such, we are only concerned with morphemes that are also lexemes.

As such, this idea of implementing a "morpheme editor", where you can pick apart lexemes into even smaller units of learning, appears to be a dead end.