Adithya Kumar

> (maybe write a paper?) @snowleopard Quite intriguing! Let me know if I can help :-)

Hi, Please go through the code in the following repository and let me know if it could work as a possible representation for acyclic graphs. [https://github.com/adithyaov/acyclic-graphs](https://github.com/adithyaov/acyclic-graphs) The function `unsafeConvertToAcyclic` in...

@snowleopard , I think that there is a guarantee of acyclicity. It is guaranteed from the representation of the acyclic graphs themselves. In the above representation of acyclic graphs, there...

@snowleopard , It seems I am a slow learner, I believe I started off on the wrong foot. But I think I'm finally understanding what you meant! 👍 > this...

See https://github.com/composewell/streamly/pull/1838

We can do this after hosting `streamly-core` on hackage. We will need to write up some small migration tasks accordingly.

Is there a reason to not use `nilM`? Too many options lead to confusion. If `before action nil` and `nilM` are exactly the same, we might as well remove `nilM`.

I'm pushing this to the next milestone as it isn't high priority.

Pushing this to the next milestone.

Will merge it after the regression CI passes