Jeremy R. Manning

Hi @ryankirkpatrick57, thanks for the request! Is this something you’d be interested in helping out with?

ok-- thanks for the information and we'll keep this in mind as we continue to develop quail.

The code should fail if bad arguments are passed: http://www.martinfowler.com/ieeeSoftware/failFast.pdf

i think the flexibility is useful, even though this violates the "fail fast" principle. so...let's go with 1. we should tell the user where their arguments are going at least...

Could be done using https://cloud.google.com/speech/docs/async-time-offsets

Tips and tricks tutorial: https://github.com/ContextLab/CDL-tutorials/tree/master/coding

Could also model multi-list data, e.g. [this dataset](http://memory.psych.upenn.edu/files/pubs/LohnEtal15.data.tgz) ([Lohnas, L. J., Polyn, S. M., and Kahana, M. (2015). Expanding the scope of memory search: Intralist and interlist effects in free...

@KirstensGitHub's idea during lab meeting: show a single image and then model recalls unfolding in time (static stimulus, time-varying recall)

bonus points: - output an admonishment if user is using python 2 - construct sample tweet

Given our recent proxy server approach, implementing this is now more complicated. We need to limit which prompts are "allowed" in order to protect against misuse of the proxy server....