Hacker News Reader

Comment by qwertox | Hacker News Reader

I was expecting some pretty, colored graphs, but the book is in black and white.

From Wikipedia [0]:

In graph theory, graph coloring is a special case of graph labeling; it is an assignment of labels traditionally called "colors" to elements of a graph subject to certain constraints.

[0] https://en.wikipedia.org/wiki/Graph_coloring

View on HN

Lots of academic authors have decided color is a gimmick (color printing is expensive and not clear how long-lived; grayscale printing kills yellow and light-green lines; then there is colorblindness). Certainly I agree that load-bearing color is to be avoided, and I do so in my own writing.

I'm surprised too, especially given that this seems primarily intended as a book to read online so the extra printing costs for color wouldn't be an issue. I imagine there are some places in the book where there are large numbers of "colors" and so it wouldn't be helpful to show them as actual colors - but surely there are some places where actual colors would be useful.

it makes more sense in the applications, asking AI, you get examples like how to schedule exams so that no student has a conflict, scheduling flight crews, radio frequency assignment to avoid overlap. e.g. "By representing flights as vertices and connecting edges between flights that share crew members, a coloring algorithm assigns crews to flights to minimize conflicts"

If you think you must ask AI for this, wait till you learn Wikipedia article you replied on says just that but with more sourced detail and less possibility for accidental falsehood

the falsehoods on wikipedia are usually purposeful deceptions so I tend not to bother with it as a go-to. net-net AI has about the same level of accuracy and lacks the actual mendacity.

almost none of us are going to move any discipline forward, let alone discover new math, but as part of the long tail of people who could benefit from recognizing classes of problems with solutions in the domain of graph colouring, AI is sufficient for most purposes.

sorry that nobody appreciates your effort though.

sorry to break it but AI is nothing but roughly speaking sum / average of data it is fed, if you believe it has some magic layer that excludes mendacity present in that data or in people who select which data to include...

Reply on news.ycombinator.com