Johannes King

Hey, thanks for your reply. The result on master is ``` import pandas as pd data = dict( df = pd.DataFrame(data=data) print(df) ``` So, it's better I'd say - exactly...

I came here from [this](https://tex.stackexchange.com/questions/99613/asymmetric-errors-with-exponential) rather old post on stackexchange. Has there been any development towards multi-part or asymetric uncertainties?

I will make a PR

I noticed that this will be a larger project because it cannot be achieved with the current XML serialization module. The issue is that different XML parameter map to the...

> How about deleting the parameter registry and putting something like this instead? That would have been my first guess as well :+1:

> I would not use a system based on axis number, or axes quantities though. I think it is equally confusing. I agree, it is for example not clear if...

@adonath and I discussed about the script on friday. > Yes, one could create SkyCoords from PixCoords, but that won't change the SkyCircle's being drawn incorrectly. Right, it does not...

[This](https://github.com/astropy/regions/blob/master/dev/regions_pyregion_comparison.csv) is the table with the current status > With a functional parser like the one we've started on, we can guarantee 100% coverage given enough programmer time. I agree,...

> For example, when taking a multiwavelength SED from several different papers, it is highly unlikely that the user has the info for all the required columns in the spec...

> I'm still curious what people thought of the idea of creating an unbinned flux spec that uses a subset of the columns from the current binned flux (SED) spec....