Nick

> I could not get the __float128 case to work without forcing the user to link -lquadmath if using GCC which would be an unwelcome breaking change. Wait, I thought...

> I could not get the __float128 case to work without forcing the user to link -lquadmath if using GCC which would be an unwelcome breaking change. Could we use...

@mborland : Any reason this didn't get merged?

@ryanelandt : Will try to get to this this weekend! First impressions are good though.

@ryanelandt : I'm so sorry! I'm really trying to get to this! (FYI: I hate it when my PRs don't get reviewed-I feel your pain.)

Approved the CI run, but this is quite a complicated PR. I think @jzmaddock (who wrote the rootfinder) may need to take a look.

The build failures do look like a pre-existing problem, but unfortunately I don't have time to investigate; have they failed in any other builds? I have read the diffs and...

@pulver : How do I get the cool "Approved" button you put on the PR? (This is the sort of thing I have time for!)

So there is a broader question now: This now explicitly uses Eigen in the boost source directory, but traditionally we only test that our code is "generic enough" to interoperate...

> All coefficients are accessible and mutable, and the class can be used as a container. Why make the coefficients mutable?