2018 July Assorted Links

Photo taken in July 2016 at Hoover Dam in Boulder City, Nevada with a Sony Alpha α7II, FE 28mm F/2 and 21mm F2.8 Ultra-wide angle conversion.

Stumbled across Chris Stucchio a while back and saw that he would dump interesting links into a round up style blog post, have always felt like that was a good way to keep tabs on interesting articles. I’m lazily attempting to shoehorn everything I find into packet of late, but attempting to write down why it’s interesting will likely force me to critically think a bit more about what I’m attempting to hoard. Many of these I’ve just found in July, despite likely being published at an earlier date.

Culture

Technology

  • This article on Factoring the Noise protocol matrix was well written for someone who isn’t a crypto expert. If you’re using wireguard then you’re building on top of the Noise Protocol.
  • Feels weird to link to a guide for a hosting provider, but this article on netfilter was succinct and well written. Before reading I’d never visualized the tables x chains matrix.
  • Well written article titled Enter the Matrix about the “open network for secure and decentralized communication”. I am still using matrix despite synapse attempt to turn all available electricity into heat, hoping that as more people become interested the governance issues around the specification can be sorted out and more implementations will come forward.

Analysis

  • A post on stochastic gradient descent from a physics perspective was really accessible. It led to an article on how physics conneccts to machine learning. I’ve not grok’ed the entirely, and likely never will, but it’s proven an exciting jumping off point to head down the rabbit hole.
    • Using a relatively simple unsupervised machine learning approach to author identification in the federalist papers, an accessible article and also quite interesting history/background realizing that some of the artifacts don’t have author attribution.
    • A publication titled How the proof of the strong perfect graph conjecture was found was a delightful read into the tribulations of performing basic research in mathematics. It was humanizing to read about the application process, award, and struggle towards results in a way that is often not reported in final publications (indeed this was a supplementary artifact to the publication in 2008).
    • An article titled Szeged in 1934 edited by Reuben Hersh from two manuscripts written by Edgar R. Lorch about his time in Szeged with Frigyes Riesz. This was an absolutely heartwarming and delightful read, I had to gush about it to anyone who was interested. The depiction of Lorchs’ upcoming responsibilities as an assistant to John Von Neumann as well as the other social and professional obligations at the time were spectacularly humanizing. I feel like leaving with a quote from the text is appropriate:

Here I had the opportunity to work next to the most brilliant mathematician of his generation. But the job entailed such onerous duties that only someone with an iron constitution would survive. My constitution, it so happened, was not made of iron. It was made of reeds and bamboo sticks, very satisfactory under moderate pressures, but completely incapable of standing a huge overload.

Photo taken in September 2016 at Speyside Cooperidge in Craigellachie Scotland