Compsci

This wee note is about physics results I noticed suggesting it may be possible to greatly reduce the power and water required by datacentres. This is in a field of physics I am not remotely able to verify but I’m keeping an eye on what seems to be happening. Large Language Models are subject to the laws of physics in a bad way, because they use so much electricity and make so much heat. Landauer’s principle ↗ identifies the thermodynamic cost of information erasure, and some later work builds on this to show that computation can avoid erasure through logical reversibility. ...

Not-forking ↗ exists to solve a difficult problem presented in the LumoSQL project: how could we graft two very active source trees together in an ongoing way to produce a third, independent tree? I designed and tested the first phase of Not-forking, and Claudio Calvelli did most of the coding. These days I’m the maintainer, and it turns out Not-forking is a reproducibility tool which addresses the computer science problem of developer intent. ...

Reversible execution creates computers that seem to run backwards, applying time shifting techniques with simulation/virtualisation to address software unreliability and complexity. I stand by my excited comments from way back in 2005: Reversibility is the biggest advance in debugging since source code debugging — Me, on the GDB developers list ↗ in 2005 In 2026, reversibility still isn’t seen as an ubiquitous must-have for software development, but awareness is increasing. 💡 Confusingly similar names I also have an article on the totally different but similar-sounding topics of logical reversibility and thermodynamic reversibility . If you’re interested in the problems power-hungry AI datacentres present the world you might find it interesting. What is checkpoint-based reversible execution? Reversible execution is about giving the appearance of a program executing backwards in time. If you’ve not seen it before, it is just as strange and impressive as it sounds. ...

These exercises assume a CS graduate-level background and familiarity with the tools mentioned; they are for mentors to adapt for their students. I have either created or been subjected to all of them over the years, and I have mentored students through them on many occasions. The general theme here is that most of the systems and stacks that are taken for granted often don’t work very well, and often don’t seem to have a very bright future. This is even the case for famous codebases relied on by billions of people. There are no absolutes and no immediate fixes, but it is food for thought if we can demonstrate immense waste of human effort amid poor quality computing systems, even when impressive modern computer science is applied. ...

Random humans and computers Humans are terrible at randomness. If you ask people to write down a list of random numbers the result can usually be shown to not be random at all. Stage magicians and marketing experts exploit our inability to assess how random an event is. But computers, surely they should be random? It sure feels like it when your printer jams. But no, computers are often worse than humans at being random, and that’s a problem. Randomness is exceedingly important to making computers and networks work. ...