Michael Kirk

> easier to integrate with third parties without having to learn about the ecosystem. I agree that eco system discovery is a bit of a challenge. I wonder if having...

Also maybe related: #564

@urschrei didn't assert that it would impact performance to implement a trait. And he can correct me if I'm wrong, but if his concern is similar to mine, it's a...

One of my reservations is that Wkt and geo-types do not map one to one. For example, how do you represent a ~~geo-types::Triangle~~ and `geo-types::Rect` as WKT? In the `wkt`...

Upon reflection, it probably does make more sense to just use WKT as our debug format rather than inventing our own weird similar but different thing like I proposed in...

> re-implement WKT writing directly in geo-types (outputting WKT is simpler than parsing it), without read support as our debug format. Another thing to consider is that currently the WKT...

It didn't occur to me that you could even do this. I guess it's something like: ``` pub struct Coord where T: CoordinateType, { pub x: T, pub y: T,...

Sorry, I'm just now finally spending a little time reading some of these resources @urschrei. Thank you for gathering them! In taking a look at GEOS, it appears to be...

> This is an interesting comment: https://github.com/anvaka/isect#bush-algorithm Indeed it is interesting! I don't know that those benchmarks are authoritative, but that tldr seems to be: - BO is good when...

> I would suggest to add the functionality using the geo-clipper library as a version 1.0. We could implement our own algorithm ourselves as a version 2.0 in the future....