Designing ShareVitalSigns for iOS

For usage notes, see the LNhealth wiki page on SVS for iOS. This post is cross-posted to the LNhealth wiki at Designing ShareVitalSigns for iOS.

Bridges of Vancouver

For reasons yet fathomable, I’ve decided to cross by foot (almost) all major pedestrian-accessible bridges crossing a body of water in Greater Vancouver. These are, courtesy of this list, as labelled on the following map, and listed as follows, in alphabetical order:

Thulium Part 3: Ghost, monitoring, backups, and a summary

This is part 3 of the Thulium series. Go back to part 2 or jump back to part 1.

Exam season has ended, and so too must this story. There are a lot more things I could self-host, but I’ve come to a point where I’m comfortable with the services I’ve set up for myself, and other ideas have larger scales and likely would deserve their own posts (setting up a mail server, for instance).

Thulium Part 2: Nextcloud (formerly Syncthing) and Gitea

This is part 2 of the Thulium series. Visit part 3 or go back to part 1.

We are now in the middle of the exam season. What better time than now to set up file syncing and a personal Git host?

Thulim Part 1: SSH and Web Server

This is part 1 of the Thulium series. Go to part 2 or jump to part 3.

Exam season is coming up now, so naturally I’ve decided to spend my time setting up a home server. I’m hoping to eventually be able to replace Google Drive with a self-hosted instance of perhaps NextCloud or SyncThing, but we’ll start small first. I’ve installed Ubuntu Server 16.04.4 LTS (Xenial Xerus), which was sufficiently straightforward that I won’t elaborate on it except to say that using an LVM caused me to be unable to boot into the OS, so don’t do that.

Advent of Code 2017, Day 23

As with Day 18, today’s problem involved running a custom assembly program. However, as stated in part b, the program run with a = 1 is much too inefficient to run directly. Whereas with 18 you could simulate a machine in whichever language you choose and finish running the program in a reasonable amount of time, this problem requires deciphering what the program actually does, then optimizing it. We begin with the input (whose real values I won’t bother with hiding):

Advent of Code 2017, Day 3

My solution for star 2 can be found here, but without the four pages of diagramming I did to get there, it’s largely indecipherable, so here’s some explanation on the mental process.

Making Swagger play nice with ProtoBufs

My previous workplace’s frontend is mostly in TypeScript and the backend mostly in Scala, and to share data back and forth, DTOs written as ProtoBufs were implemented some time ago. There’s a script that generates from these ProtoBuf files Java classes using protobuf-java and TypeScript classes using protobufjs. The Java classes can then be used directly in Scala code.

From Mercurial to Git

A few months ago I had to figure out how to migrate a 4 GB repository from Mercurial to Git, and trim the size down along the way. Luckily, I wasn’t the first one to have to do that, so there were a number of resources I could reference, namely these two. But of course, every specific case has its own specific problems.

BunsenLabs on VirtualBox on Windows 10 on a Laptop

Why: Because.

How: With great effort and time.

I’ve gone through so, so many iterations of this installation process because of various things that have gone irreversibly wrong. Luckily, this has all been on a VM, so nothing is truly irreversible (save for the overall entropy of the universe), but many mistakes were made, then remade (“testing”, they call it) to be sure of their causes. To note: