I ran into a weird problem where I couldn’t access the OPNsense web UI using an OPT1 interface. The fix is easy, but it can be very confusing and it’s all caused by an unintuitive default. tl;dr If you can’t access the web UI via an OPT1 interface, try disabling the “reply-to” feature: https://docs.opnsense.org/manual/interfaces.html I’ve been playing around with OPNsense inside a VM as I prepare to replace my current pfSense router which is no longer fast enough.

The title of this post isn’t supposed to be provocative. After all, it’s simply the textbook definition of what git is. So why bother stating it? Well, I’ve worked with a fair few junior developers now and I’m starting to see a pattern. Many of these developers have never programmed without git and they see git simply as “the way to get new code into a repository”. A glorified copy, essentially—but an annoying one that is prone to going wrong.

The Norfolk Coast Path (NCP) is a waymarked, long-distance footpath and National Trail in England. It stretches from Hunstanton all the way along the Norfolk coastline to Hopton. I walked it with my partner Vicki over the course of seven days in September 2022. Planning The NCP is around 135 km (83 miles) long and quite flat. The way is mostly on good, well trodden paths with some slower sections on sandy and shingle beaches (but you can often choose to walk inland on firmer ground instead).

At first glance, undo seems like a simple thing expected of most software these days and hardly worth writing about. Indeed, when I say Emacs has a very powerful undo system—probably more so than any other text editor—you may wonder what could make an undo system powerful. So let’s start by considering two big problems most undo systems have: If you undo something, make some changes, then change your mind, what you undid is now lost and unrecoverable, If you make changes in two parts of the same file you cannot undo changes in the first part without undoing changes in the second part too.

When you work a lot on the command line, history can be invaluable. I’ve lost count of the number of times I’ve forgotten how I ran some earlier command and used my bash history to find out what it was. This is one of the big advantages of using CLIs over GUIs. Accessing history The main interface I use to my history is ^P (Ctrl-P). This recalls the previous command from history.

For as long as I’ve been a Firefox user—almost 20 years at this point—it has featured a “quick find” bound to the venerable / (forward slash) key. Following a pattern established by other software like less, man, and vi the slash key was simple: it finds stuff in the current page. Common patterns like this are great. It’s why everyone has agreed on what Ctrl-C and Ctrl-V (or very similar) should do across all operating systems.

Org mode supports icons in its agenda views. The icons can be given as either file paths to images (like SVGs), as image data or as a display property. I use a Nerd Font along with the nerd-icons package in my Emacs config, so I thought I might as well enable icons in my org agenda views. The nice thing about using nerd fonts is this works perfectly in text mode too (assuming you have a nerd font configured for your terminal emulator).

This blog is a static site built with Hugo. Being static means it can be served from a basic, standard (you might say stupid) web server with no server-side scripting at all. In fact, this blog is currently hosted on Github Pages, but it could be anywhere. Up until now, if you wanted to include an interactive map on a static site you were limited to using an external service like Google Maps or Mapbox and embedding their JS into your page.

Docker Compose is a brilliant tool for bringing up local development environments for web projects. But working with multiple projects can be a pain due to clashes. For example, all projects want to listen to port 80 (or perhaps one of the super common higher ones like 8000 etc.). This forces developers to only bring one project up at a time, or hack the compose files to change the port numbers.

I use the Unbound DNS resolver built in to pfSense. By default the resolver filters out any results that are private IP addresses. Normally this makes sense: no public domain should have a private address. But sometimes it does make sense. For example there are some useful services like sslip.io that will resolve to any IP address that you like. So 127.0.0.1.sslip.io resolves to to 127.0.0.1. This is can be useful for local development, especially when working with containers and reverse proxies and the like.