Daniel Turek

@paciorek Please let me know if this runs into conjugacy code that you want me to update / deal with - otherwise I'll presume it's sufficiently in the `stick-breaking` realm...

@paciorek I wish I had more useful feedback for you. The proposed solution would handle common cases of the `inprod` problem, which maybe is sufficient for the time being. I'm...

@paciorek The new information seems helpful to have. Only two comments: 1. As you mentioned, I feel some of this new content might be better moved to `help(modelInitialize)`, with a...

@paciorek Thanks for tagging me on this. To address your highest level design questions upfront: - I agree, I feel this should be added as part of the `RW` sampler,...

@paciorek What you said makes sense; I certainly understand your point that "in the past, we're never able to anticipate and handle all use-cases; users always come up with something...

@paciorek I looked over this carefully, and it looks great. This will be a nice addition.

@perrydv Agreed.

@paciorek I, personally, think this `returnComponents` option is useful to have. But that said, if the momentum is to clean up and remove it, I would understand.

@perrydv To be consistent with the method name `findMLE`, would it make more sense to have the covariance-finding method be called `findCov`, rather than the current name `vcov` ? How...

Sounds good to me.