search

hypertidy / silicate

A general form for complex data
https://hypertidy.github.io/silicate/
53 stars 4 forks source link

guidelines and rationale #117

Open mdsumner opened 4 years ago

mdsumner commented 4 years ago

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 commented 4 years ago

Also, original rationale: https://github.com/hypertidy/silicate/issues/80

and the core https://github.com/hypertidy/silicate/issues/41

mpadge commented 4 years ago

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

mdsumner commented 4 years ago

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: https://github.com/dcooley/sfheaders/issues/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