guidelines and rationale

[mdsumner]

We need an overall guiding document for principles. Motivation is coming out of anglr, for when we only have implicit topology i.e. for a surface in a sf polygon, what meshing model is used?

The cheapest one, i.e. TRI0(). TRI() is next expensive (adds visible property, and native orientation, and arbitrary subsetting) ... but cheap triangles, next is DEL0() (control over area, angle, Steiner points, Delaunay constraint, etc), then DEL() ...

What controls colour/size properties of vertices, when they are shared by higher level entities?

Precedence, i.e. which ones come first - (and, if true then we can provide control of that in the relational-forms - it's harder in the structural-forms because order is baked in).

That kind of thing

[mdsumner]

Also, original rationale: #80

and the core #41

[mpadge]

I'd love to chip in on this one, and also suspect that my somewhat "external" perspective might help things here.

[mdsumner]

Hey definitely! Getting anglr in shape has really sharpened the concepts (because I can see them ...) in a way I didn't expect. As has this: dcooley/sfheaders#72

A few more waves of updates, soon PROJ and reproj will be functional, so we'll be closing in on powerful kit.

And GEOS, seen this? https://github.com/paleolimbot/geom