davidcamel
1. Yes, "porting attacks" (where an adversary convinces your carrier to port your phone number to his/her own) are a real threat. You can mitigate these somewhat by choosing a carrier that has a relatively strong porting procedure. Project Fi (Google) requires a temporary PIN generated by the user's Fi app, as well as logging in to one's Google account. I don't know what the other carriers require today, but this is less than what I experienced when I ported my number from Sprint years ago for example.
2. The author says that 2FA is overhyped, which is maybe true, but why don't more services allow physical devices (e.g. Yubikeys) to be used for 2FA? Often the phone number is the only choice offered for 2FA.
Example:
task add project: support_customer_migration priority: H -- support#XYZ - analyze dependencies for customer X - github.com/repo/issue/number
I'm young and already excuse myself by saying that I don't mind having just one or two close friends. But this is how it starts: we tend to think we can handle being mostly friendless, and yet, as we age, it seems that many of us can't.
Not sure what the solution is, but you've definitely underscored a big problem that I've (till now) underestimated. And I'm sure others have too.
EDIT: similarly, does it support MMS in addition to SMS?
Also, his argument that computers have been a force for conservatism is new to me. I'm not sure how true that is today.
What is the advantage of the source map?
Seeing a real-world example of this fantasy come true is fascinating. The article was also surprisingly well-written; most mention of higher mathematics in the media is oversimplified to death, but this was an honest and yet approachable presentation of the Rota conjecture (now theorem).
By the way, here's another result on chromatic polynomials (proved first by I don't know, but re-discovered by my combinatorics class):
Define a "gluing" operation by taking two graphs and connecting them along a common vertex.
The chromatic polynomial, h(x), of the new graph, is the product of the chromatic polynomials of the subgraphs over x: h(x) = f(x)*g(x) / x.