Hacker Newsnew | comments | show | ask | jobs | submit login

The Imus map does look nicer, but as a computer geek I am more interested in how one could capture the heuristics he uses and better automate the process. Surely all (or most) of the label placement, typography etc. ideas he uses could be captured in a set of criteria that could improve the state of the art in computer-mapmaking. This is similar to what Knuth did to maths typesetting.



It's possible that some of those map layout problems are NP-hard, as similar problems in graph layout are proven NP-hard.

Point well taken either way.

-----


Approximate solutions are fine though.

-----


>Surely all (or most) of the label placement, typography etc. ideas he uses could be captured in a set of criteria that could improve the state of the art in computer-mapmaking

Yeah. The following quote really annoyed me:

>He used a computer (not a pencil and paper), but absolutely nothing was left to computer-assisted happenstance.

Happenstance!

-----


What's wrong with that word?

-----


Well, if Buddy spent two years writing the algorithms to dynamically place the names on the on-line maps, how is that placement "happenstance"? Just as much effort went into it as went into the hand-drawn map...

-----


The algorithm isn't happenstance, but any individual label placement on the map could be.

-----




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: