[rfc-i] Natural Language Processing (NLP) applied to RFCs

Carsten Bormann cabo at tzi.org
Wed Mar 27 01:10:32 PDT 2019

On Mar 27, 2019, at 00:44, Michael Welzl <michawe at ifi.uio.no> wrote:
> Sure - it’s easy to imagine such a tool just producing a long list of useless garbage,

Well, we have the not-quite-NLP “idnits”, and it certainly is useful.
What is missing from idnits, and should be in anything that has false positives, is a way to acknowledge each false positive that has been identified as such, in the source and suppress the listing of that.

Grüße, Carsten

More information about the rfc-interest mailing list