Blog Directory
Directory Blog

Aphyr

Kyle Kingsbury on distributed systems, Jepsen testing, and database correctness.

aphyr.com

Aphyr is an independent blog covering systems & infrastructure. It publishes a few times per month, with 12 posts in its archive and 7 readers following along on Blogs Are Back.

About This Blog
Activity

Occasional

Publishes a few times per month

Followers

7

Category

Independent Blog

Languages

English

Feed Accessibility

How this blog's content is accessed through Blogs Are Back.

Full Content

RSS feed includes complete post content for reading in-app

Proxy Required

Feed is fetched through our proxy for browser compatibility

Proxy Post Links

Post pages are loaded through our proxy for compatibility

Embeddable

Posts can be displayed inline in the reader view

Collections

This blog appears in the following curated collections.

Latest Posts

Recent posts from Aphyr's RSS feed.

Basic Letters with LaTeX

Every so often I find myself cracking open LibreOffice to write a mildly-formal letter—perhaps a thank-you note to an author, or a letter to members of Congress—and going “Gosh, I wish I had LaTeX here”. I used to have a good template for this but lost it years ago; I’ve recently spent some time recreating it using KOMA-Script’s scrlttr2 class. KOMA’s docs are excellent, but there’s a lot to configure, and I hope this example might save others some time. Here is the TeX file. You should be able...

How to Unsubscribe from Modern Luxury

A few years ago I started getting issues of Modern Luxury in the mail. I had no idea why they started coming, and I tried to get them to stop. This should have been easy, and was instead hard. Here’s my process, in case anyone else is in the same boat. First, if you use it, try to unsubscribe via PaperKarma. This is convenient and works for a decent number of companies. PaperKarma kept reporting they’d successfully unsubscribed me, but Modern Luxury kept coming. Second, write to subscriptions@mo...

Trudging Through Nonsense

Last week Anthropic released a report on disempowerment patterns in real-world AI usage which finds that roughly one in 1,000 to one in 10,000 conversations with their LLM, Claude, fundamentally compromises the user’s beliefs, values, or actions. They note that the prevalence of moderate to severe “disempowerment” is increasing over time, and conclude that the problem of LLMs distorting a user’s sense of reality is likely unfixable so long as users keep holding them wrong: However, model-side i...

Durastar Heat Pump Hysteresis

In which I discover that lying to HVAC manufacturers is an important life skill, and share a closely guarded secret: Durastar heat pumps like the DRADH24F2A / DRA1H24S2A with the DR24VINT2 24-volt control interface will infer the set point based on a 24-volt thermostat’s discrete heating and cooling calls, smoothing out the motor speed. Modern heat pumps often use continuously variable inverters, so their compressors and fans can run at a broad variety of speeds. To support this feature, they us...

Blocking Claude

Claude, a popular Large Language Model (LLM), has a magic string which is used to test the model’s “this conversation violates our policies and has to stop” behavior. You can embed this string into files and web pages, and Claude will terminate conversations where it reads their contents. Two quick notes for anyone else experimenting with this behavior: Although Claude will say it’s downloading a web page in a conversation, it often isn’t. For obvious reasons, it often consults an internal cac...

Follow Aphyr

Add this blog to your reading list on Blogs Are Back, or visit the blog directly.

https://aphyr.com/posts.atom