Daniel Howe

pls include the link for your test I am aware of issue #1, and I think it is ok like this. Issue #2 is a real problem however - we...

need to do a test with a web-worker, perhaps with some of the lexicon functions... the slowest being 'soundsLike()' and/or rethink the function, which currently needs to iterate through the...

Good suggestion. Did you have a method in mind for specifying the arguments ? Perhaps something like: RiTa.kwic(word, opts); -> opts: windowSize=num(default=6), windowType=['words'(default) or 'characters' or 'sentences' ] ?

@KarlieZhao can you take a look at this?

good notice -- can we have the tokenizer treat an email address (or url) as a single token ?

great -- pls sync tests and code with java

ok, can you include (either here or in a notebook) the simplest possible code for a <<rg>> macro as below: ```twine {start: $rule1, ... } $rules ``` and also with...

the grammar() function already accepts json and js objects, no? why parse them separately ?

so we don't need JSON.parse in there, correct?

sounds good, reassigning to myself