grimcompanion

Of course, you have to be well enough to exercise.

My mom was diagnosed with terminal breast cancer 11 years ago, at 59. She wasn't super into fitness when I was growing up (no time with 4 kids to raise). But ~5 years after her diagnosis, still doing regular treatments and on loads of medication, she picked up running as a hobby.

She is absolutely incredible and I'm so proud of her. But she would be the first to tell you about all of the friends she made who died of cancer, including many much younger and fitter women, and I know this is obvious but I'll say it: it wasn't because they didn't exercise, it's because they didn't respond well to treatment. She had to stop going to her support groups because literally everyone she met there had died[1]. She is so fortunate that she's responded well, enough that she was afforded the time to emerge from the shock and depression and realize that she had a lot of life left. We are so lucky that she's here with us and well enough to develop new hobbies, etc.

I do think the sort of information in the study is interesting on a population scale. On an individual level the fact is that some people will respond to treatment and others won't. The ones who do may just feel well enough to start an exercise program. If they can, it may well improve the quality of their life, and may extend your time with them, and that's an amazing gift.

I'm not sure why I felt compelled to share my anecdote. Maybe just to share what the messy human side can look like.

[1] I should note that her support group was specifically for people with a terminal diagnosis. People in any other type of breast cancer support group won't have this experience.

IMO this definitely helps, but only to a degree, because the more quickly I can read/digest output, the more quickly I've got more output to read/digest. The jargon/quirks aren't as dense per message but they still accumulate eventually, and so either way I hit a wall where I need to do something, anything, besides read another line from CC.

Yep, I've heard humans use load-bearing twice in the past week, versus approximately never in a software eng context before that.

> I believed this so strongly that my company built an entire product around this concept. I used to tell folks that "session transcripts were the new oil," that they were more valuable than the code itself.

This is infuriatingly common wrt talking/writing about how to use AI effectively. All of the "this is how you write an AGENTS.md" and "you need to talk to it like X to optimize it". Like sure, you can believe that as much as you want but unless you provide some evidence you can keep your shitty CLAUDE.md to yourself and don't pollute the whole company's git repo, thanks.

Totally agree. I actually got a job off of AngelList back in the day, when I was fairly junior. Was getting messages from hiring managers, founders, and in-house recruiters almost daily. 5 years later, I updated my profile with a much more legit background than before, messaged a few places - it's been 100% crickets.

OTOH it will provide more surveillance of the police themselves. Humans are also bad at gun detection (sometimes willfully so) and this provides another check.

I think there are different solutions for features vs issues.

For features -

Do developers have ownership? Do they write tech specs, proposals, etc? If so, these should include a section on what problem is being solved. If they don't understand the product this is a good way to kick off more conversations about it.

If developers aren't doing this - e.g. if you're leaving it up to management to tell them what to do and how to do it - that's the problem. Companies like this lack incentives for devs to learn about the product if they're just going to be told exactly what to do and aren't given the freedom to think creatively. If this is you and you want to shift the culture, start encouraging developers to propose features and show that you value their contribution to the product. Invite lead devs to meetings and to comment on documents where features are being proposed, from the beginning, instead of handing them a thing to implement.

For issues -

Are PMs communicating issues well? Including why the issue is important? Perhaps something like a loom demonstrating the exact issue would help.

Have a rotation for addressing issues. e.g. if a dev is on call they should work on customer issues when they're not fighting fires.

I absolutely don't endorse it. I just find it counterproductive to say things like "malice is the point".

I do have trouble finding good faith arguments in favor of this policy. It is cruel. But the people who decided to implement it aren't "other". They're humans who think they're good people (aside from a small minority of people who really don't care) and much as we'd like to think so we're not that different than them. If we can understand their justification, that's a step toward actually convincing them there is another way. And yes, I have changed many people's viewpoints with this level of patience, not everyone is too stupid/mean/insult of your choice to change their minds.

It's way too simple to categorize behavior like this into good/evil. It's a worthwhile thought experiment (and habit) to assume that everyone is trying to do the right thing, and try to understand how they might come to a different viewpoint than you.