Matthieu Viry

Thanks for your PR and sorry it took me a while to reply. Indeed, making value type generic and independent from coordinate type is a great thing :+1: (and it...

Hi @netthier! Is there anything I can do to help finalise this PR? Would you like me to continue the work you have started?

> Who knows what kind of numbers people might put in there, maybe some can indeed fail serialization. I don't really know either but let's say "better safe than sorry"....

Thanks! > This could use some tests tbh, but I'm not sure if I have the time and knowledge required to write good ones I don't really have an idea...

> if users upgrade and want to keep using floats, they'd get a slowdown in exchange for having error handling on a bunch of f32->f32 conversions I wonder if using...

It's a good idea, let me think about it for a moment and discuss the solutions available to us here.

Thanks for proposing to contribute this new feature, it sounds like a very useful addition. And I'm on your side regarding the deletion of the `Float` type to avoid the...

Yes, these are two totally valid use cases that should be supported by this library. I think it would also be an opportunity to properly support the presence of NaN...

Thanks for bringing this up. No, there's no technical reason for that, and having a `parallel` feature in this crate would indeed be a good thing. Indeed, the improvements of...

> But it could be possible to add a new, alternative implementation in this crate with the [cleaner isobands](https://en.wikipedia.org/wiki/Marching_squares#Isobands), taking care not to look at any AGPL code while doing...