Dollar signs entered into Builder loop nReps fields are being compiled to JS in the TrialHandler nReps parameter, causing a reference error in browsers, e.g., see Discourse post

@stevenbird
Initial release of the corpus available at: https://doi.org/10.7488/ds/1411

Suggested NLTK name: ARCOSG

I have updated and corrected the corpus for inclusion in NLTK. (The one at the link above is older and shouldn't be used).

Corpus reader code verified:
`arcosg = LazyCorpusLoader(
'arcosg',
CategorizedTaggedCorpusReader,
r'.*.txt',
cat_file='cats.prn',

Documentation for case 3 and case 1 in definition extraction methods is the same (ref: https://github.com/LexPredict/lexpredict-lexnlp/blob/f3920be16dac588b2f38e17811ea5482b417954d/lexnlp/extract/en/definition_parsing_methods.py#L134) however, case #3 seems to only work for Title case or upper case words followed by something from the strong trigger list, and case #1 must necessarily have the word

Summary

Currently, the FLAT can be configured to limit the number of sentences/paragraphs loaded at once. If a document is longer than the limit, it is sliced into multiple pages which can be navigated via the page dropdown on the top-left of the editor view.

Problem

The page is too small and the total number of pages are not displayed anywhere. So, after the first page is annota

When merging all the new inventory resources, the process introduced duplicate bibtex keys (some from the original data providers themselves).

TODO:

• remove duplicates
• double-check that all inventory ID's have a bibtex (currently at least two from ER are without citations; others are URLs to online materials which need a reference added)
• make all entries valid bibtex format (when using

https://creativecodinglab.github.io/TextAnnotationGraphs now points to docs/index.html. We should include a link to the live demo under the corresponding section in the docs.

This is something that will greatly help your code quality, especially if you have to target multiple platforms. You also then get substantial type checking benefits from mypy and other tools since there's no type ambiguity between strings that are representing strings and strings that are representing file paths.

It would be helpful if TextGridTools would provide more information when it encounters an error when loading TextGrid/ELAN files, e.g., the type of error and the line number on which it occurred.

We should have our policy on Glottocode assignment and maintenance documented somewhere (also in the web app).

1. Language level Glottocodes are always valid - i.e. the way to mark obsolence for these is moving the languoids to the Bookkeeping pseudo family.
2. Sub-group level Glottocodes may be become obsolete and will then be removed from the current version. The codes will never be recycl