vinc

I noticed the same thing at work, the code is very defensive, and often the case could not happen in prod, so my solution at the moment is to make it prove that an issue can happen and try to measure how frequently it would happen with real data. That's often enough to realize that a particular defensive measure is not needed and the code complexity decrease as a result. And it's something we can automate.

I think it's an important point because the more complex the code become the harder it will be to maintain it, so hopefully this process will improve.

I use claude at work, I have to read its wall of text all day long, it's instantly recognizable. Like many others I'm developing an aversion toward it. The OP's project is pretty cool but I just couldn't read the readme.

I think it's an issue for hobby projects. The other day I had a neat idea, looked into it and realized I would not have time to do it but still wanted to use it so I vibe coded it over an evening and there it was. But the code is not mine, I feel bad about it, it's completely hollow, I wish I had the time to do it properly. I'm glad I'm still hand coding my main hobby projects to avoid that.

You can visit only 25 pages on the website with a free account, that's a bit too limiting. I'd suggest limiting the number of searches maybe but not the pagination over the results.

I got curious again to find my old posts and figured I could download and grep the mbox files from Internet Archive instead.

It becomes boring very quickly for me too, the planets are the same, the stations are exactly the same, every alien you meet is the same as the previous ones. You build a base on one planet, then another base on another planet, and it's just the same.

The game is also too unrealistic for me: the planets are too close, you shoot asteroids just like you could 47 years ago, you can't dig very deep at all, your ship follow you wherever you jump to.

The trailer was beautiful, the actual game is really nice for a short while, but it's too simplistic and repetitive.

You should plan the task before implementing it to make sure that it will do the right thing.

Brittany was a lot more behind a few years ago, and we had a big storm in the region that caused a lot of damage and slowed down the work but looking at the map today the issue looks almost resolved. I live in a forest in the middle of nowhere, the poles got installed at some point and then I had to wait for years for the fiber to finally arrive, but now it's there and it's great!

In the meantime we also got 5G and Starlink so plenty of good options for fast internet.

Same in Northwest France: I've never seen the grass so yellow, and now the ground is even becoming bare brown. Some tree species are yellowing/browning. With the prolonged lack of rain and extreme heat for the region, it's a summer like I've never seen before. And it's probably one of the coolest of the rest of my life.

A good part of the article feels like it was written by Claude indeed:

- "The reallocations were real, but they were never the bottleneck."

- "Note that the villain is not the branch itself. It is the branch that [..]"

- "Same million floats. Same threshold. Same function."

- "Notice the price we paid though."

Yes completely!

I started drinking one small but dense cup of coffee every work day just after lunch to avoid being sleepy after the meal two years ago, and this year I noticed that I wasn't feeling good during the weekend, then I noticed that if I drank a coffee after lunch I'd be normal. Skip forward a few months, I notice now that if for some reason I forget to drink my coffee after lunch I wake up the next day with a migraine, each time.

I like the ritual of grinding my fresh grains and brewing my cup, but I think I'll move to decaf and a quick nap like I was doing before. The migraine is the proof of an addiction.

I've been using decimal time for years and I really like it. I revived an obsolete unit, "dimi", to go with "centi" so that 1 centiday is 0.01 days, and 1 dimiday is 0.01 centidays.

Yes thanks!

I remember all the things I needed to implement before reaching userspace, it's a lot indeed. After that it gets easier because every new change is motivated by a specific smaller goal like supporting the network card of a new laptop, or adding sounds to the emulator, or adding one more syscall or device file. You get rewarded by new capabilities after each addition.

Impressive! I like the goal of making a hobby OS viable as a daily driver.

I've been working on my own hobby OS for half a decade. It does a lot less, but it has helped me realize that we can remove much of the complexity of a generic mainstream OS while still meeting our personal computing needs. I know I'm just poorly reinventing something between DOS and Unix/Plan 9 in an extremely limited fashion, but it's absolutely perfect for experimentation!

We started using LLMs heavily at work this year and I switched from Vim to Zed to help with that. I now spend more time writing to the chat than editing code, and what I quickly learned to avoid frustration when I don't like the result was to git stash or reset the code and edit what I last wrote instead of trying to argue with the LLM. The chat doesn't have to be linear, it can branch off. Too bad we can't currently edit previous messages with Claude in Zed.

Repetitive issues are fixed by updating the memory or the prompt file, they can learn this way.

Also lately I noticed that Claude forget too much when compacting, so I just start a new session and it's easy when you spend a lot of time in plan mode to produce a written spec before implementation.

I assembled a solar server with those parts laying around last year:

- Victron Monocrystalline Panel 90W 12V

- Victron Gel Battery 12V 60Ah

- Victron MPPT Charge Controller 75V 15A

- Raspberry Pi Zero W

- Witty Pi 5

- Sixfab 4G/LTE Base HAT

- Quectel EC25 Mini PCle 4G/LTE Module

Almost 100% uptime except for a few days after a bad winter storm, pretty neat!