valle 0–6 jump · t theme

vaughn@valle:~/blog$ less 2026-10-28-mkdir-blog.md

mkdir ~/blog

· 2 min read

#meta#site

This site now has a blog. It will hold the notes I write anyway while working: the number that did not match the paper, the attack that worked on the first try, the build that failed in five configurations before I gave up and wrote it down. Most of it will be hardware security, since that is what I spend my days on: SRAM physical unclonable functions, how much entropy cheap silicon really carries, and whether LLM agents grounded in real EDA tools can find hardware bugs. Some of it will be Linux, because I have been building distros since 2018 and the habit is not going anywhere.

What to expect #

  • Measurements over opinions. If I claim a number, the post will say how it was measured and what would make it wrong.
  • Negative results. A method that did not work is worth a post. Most of what I learn comes from those.
  • Short. A post is done when it says the one thing it is for.

How the pages get made #

The blog is part of the same repository as the rest of the site. Each post is a Markdown file with a few lines of front matter:

markdown

---
title: "mkdir ~/blog"
date: 2026-10-28
tags: [meta, site]
---

This site now has a blog. ...

A build script turns the folder of posts into static pages, an RSS feed, and a sitemap, and Cloudflare runs it on every push. There is no client-side rendering and no third-party request: the pages are plain HTML and CSS, served with the same strict Content Security Policy as the rest of the site, so they work with JavaScript off and nothing here tracks you. Code blocks are highlighted at build time, which is why they follow the colorscheme picker in the status bar.

If you would rather not check back, the feed has every post in full.

Why now #

I am applying to PhD programs this cycle, and the projects I would want to talk about in an interview are the ones with a paper trail. A blog is the cheapest paper trail there is. It also forces me to explain things to someone who was not in the room, which is the test most of my ideas need to pass anyway.

The grid at the top of the home page is real power-up data from one of my ESP32 boards. The first proper post will be about what those cells did after an hour in a freezer.