Hello Bruce,
What is the easiest way for CS users to report typos? Is this expected at all?

As a sidenote, you’ve recently commented that it was not worth the hassle to move the codebase on a service like github or Bitbucket, but would you consider opening up its textual parts (i.e., LIVEDATA & RAWDATA)? In my opinion, that could potentially enrich the dictionaries while mitigating development interference.

easiest way to report typos is to email me.
Is it expected at all?  Am I human?

What do you mean by opening up its textual parts?  Do you mean putting them on github? How is that different from putting the codebase there?

If I moved stuff to github, are you suggesting that anyone can check in changes or merely that they can “get”?

By “is it expected” I mean “should we report bugs/typos that do not affect matching or normal CS operation”.

As for opening parts of CS:

From what I understand, the DICT, RAWDATA and LIVEDATA are quite independent from the rest of the program. As such they could be reviewed and improved by the community easily because the bugs are mostly typos, missing synonyms… Duplicating this part on a service like github would allow you to harness the eyes and fingers of the community to improve and raise bugs from these mostly english language files, while not changing your build process for Chatscript. You would only have to merge the latest revision of the github repository into CS before doing a release.

Since duplicating it to github or Bitbucket would allows anyone with an account to suggest changes, they would need to be approved before being integrated. You and/or a group of administrator would be able to allow or deny any change to the DICT, RAWDATA and LIVEDATA files, to make sure that no new bug are introduced.

I appreciate all bugs being reported, including spelling errors in the documentation.

Actually, the DICT folder is generated from data I have, it is not “editable” in the normal sense and is not independent from CS internals.  LIVEDATA has some files that are “arbitrary” like interjections or british and some that are not, like “queries” which is again tied to the CS engine.

RAWDATA is engine independent mostly. The ONTOLOGY section has some integration with the engine.  WORLDDATA does not.

I would say this… you are welcome to create a github repository for RAWDATA files and LIVEDATA files. And I would be willing to try diffing changes against my version before release and see if I am willing to incorporate changes. And I will post on this site the results—- what I did or why I did not in various cases.

Great, thanks! I will duplicate RAWDATA and LIVEDATA from 5.31 into a repo and post it in the forum once it’s done, see if other users want to contribute.