Andrew Gibiansky

We can parse the line. If the parse fails then say it's incomplete, unless the previous line was a newline. That way two newlines always submit the code. This makes...

I believe `stack ghci` should work; I usually do `stack build` and then rerun, though.

It should be pretty straightforward to do this, if the GLSL code can run with WebGL. Generating JS and having it displayed is pretty easy. Is that the goal? If...

In addition to that issue, it seems like we should disable warnings before the IHaskell internal code, and then re-enable warnings afterwards (if they were enabled at all). We'll need...

Messages introduced [here](http://ipython.org/ipython-doc/dev/development/messaging.html#history)

I am not sure how this is actually used, so delaying implementation until I, or someone else, runs into this and needs this.

@stallmanifold I think you're on the right track. How are these requests / responses actually used in the frontends? (How can I trigger them?) Also, if there's anything you have...

No, IHaskell doesn't support this (yet).

@miked-mellanox We have more info and speedup comparisons in the [blog post](http://research.baidu.com/bringing-hpc-techniques-deep-learning/), along with performance / scaling plots.

I'm having a bit of difficulty getting the full error (with segfault/memory corruption), so maybe that aspect isn't even `tabix`'s fault. However the following code ``` import multiprocessing import tabix...