I'm not convinced, yet, that eliminating LLM-generated comments is the right path for me. I do review everything written by an LLM and some comments are actually pretty good, but sometimes I'm just too tired to try to figure out how to reword an oddly worded one.
I just added Sanglard's rules to my ~/.claude/CLAUDE.md file, did another code review, and found some LLM-generated comments were really hard to understand. I think they're due to invented metaphors and flowery language instead of using standard terms, so I've added this:
- Comments must be literal. Don't invent figurative language for what a plain technical term already says — write "rows still reference it," not "rows still wear it."
Improving LLM code generation is an iterative process. I'm glad people share their efforts to improve it.
I worked for a company that measures developer's output in number of commits, PRs created and approved, and Jira tickets per sprint. Management doesn't even insist on that the tickets have any kind of an effort estimate.
It can be pretty depressing, until you learn how to game the system - create tickets for yourself that are tiny amounts of work. I hear that it's getting harder to do that, because management is looking more carefully at tickets generated and it looks like they'll start having developers assign points to their tickets before the tickets are added to the sprint
I too have written C programs for decades. I encountered reference counting when I learned how to write Windows kernel drivers. It was very liberating to see that there were so fewer opportunities for memory leaks when reference counts were applied liberally.
> Well, we know that it's impossible to remember everything.
Yet it's possible to remember a lot. Those who pass "The Knowledge" test are truly inspirational. See this 60 Minutes report on London's Black Cab drivers [0]
Remembering and learning are very different activities, however. It is pretty impressive what people can learn! And some of us do have superior knowledge of what appear to be "useless" trivia as well. (Ever seen Jeopardy?) But it is very well-researched and understood that _most_ of us have very bad memory and recall compared to a computer. One popular hypothesis is that every time a particular memory is recalled, it is slightly changed until it bears little resemblance to the original event, or even the earlier memories of it.
...yes but apparently while the posterior hippocampus actually grows in these people the anterior hippocampus shrinks... so there may be tradeoffs required to get this level of spatial knowledge... https://www.pnas.org/doi/10.1073/pnas.070039597
I'm curious as to how low a temperature your heat pump will operate. I live in New England and replaced a whole-house air conditioner with a heat pump, but the heat pump works only to 35F. Much colder than that, and an auxiliary electric heater kicked in. The first Winter cost me about $800 over my gas-fired forced hot water heating system. I had the contractor disable the electric heat in the Spring and rewire the thermostats to start the (high efficiency) furnace when the outdoor temp got too low.
Keep in mind that heat pumps have a limit how much they can pump (it also depends on temp., there is less heat in 35F air). If your house is not well insulated, at a lower temperature it would be loosing more energy and eventually it would reach the threshold where it's performance is not enough to keep up.
reply