Not to mention how many features that used to work have been summarily broken with no indication on whether we'll ever get them back, due to the wholesale replacement of previous functions with LLM-driven functions.

I can't even press the "favorite" button for my google photos on my google home device any more. It just says "I don't have access to photos" whether I use the button or voice (both of which obviously used to work).

Bold move leaving out the dash between words a la experts-exchange lol.

thatsthejoke.jpg

expertsexchange.com was a site from the before times.

I wondered if it was intentional, but thought I would double down on it in case people missed it.

Was the dashless domain really a site (or the site) at one point?

It was dashless for many years.

As dashless and pointed as https://penisland.net/ !

always have a backup plan (:

Claudhd

It's a user daemon that runs on my machine and exposes a unix socket, and then a bunch of hooks in claude, zsh, vim, etc, that report directory and commands I've run and all that, pipes it to claude Haiku for summary, and then stores context in sqlite. It also exposes that data as MCP so I can use claude to say "hey what was I doing yesterday," or any arbitrary time range.

I find that in the age of using AI agents, "Wtf was I working on yesterday" is an even harder thing to remember for me, so this helps me kind of track everything with a database that a) has AI summaries already and b) can be accessed by AI as well as a CLI.

How do you do this? (The hooks part)

What does it matter if it was yesterday or last week, and why would git log not suffice?

Because I work across a dozen repos any given week, and I have my own side projects I play with as well. It blurs together. Global git hooks are another data source for the project, which I'd forgotten to add, so that data is in there too.

The "expose it as MCP so AI can query it" pattern is one I keep seeing work really well in practice. We did something similar for business metrics in Databox - the interesting design question was whether to expose raw data or pre-aggregated metrics. We ended up with metrics (with dimension breakdowns) because agents hallucinate less when they don't have to decide how to aggregate themselves. Curious if you ran into anything similar with the sqlite summaries - do you find the AI-generated summaries more useful than raw command history?

Maybe not hand-written, but definitely static, and at least human-reviewed/tested to only allow sending to previously-validated email addresses.

Right, as in, does not accept an email as a parameter. If its anything like my company they are turning out "agents" super fast and just hooking them up to internal APIs usually via a light MCP wrapper. Since MCP doesn't have any security or auth built in, and internal APIs usually are light on security you have issues like this.

Or they can move to NY "full time", if I'm understanding correctly, which will likely also improve the city's tax revenue from more of that person's expenses incurring city taxes.

Wouldn't that mean that they would then also pay NY income tax?

Not just on their expenses - if they become resident in NYC, they'll own 4% NYC income tax (on top of the 10% NY state income tax).

Now that's a fun microcontroller project idea. An analog dashboard for ram/cpu/whatever. I'm sure it's been done.

Edit: https://sasakaranovic.com/projects/diy-analog-resource-monit...

I'm curious whenever I hear takes with your perspective.

Entering the workforce happens at an age where people have built (some more rudimentary than others) a level of understanding and self control regarding delayed gratification and Type II fun.

Did you have the kind of life where you were never really challenged to build that skillset, or is the mental stimulation so strong for you when you use AI that it overcomes executive function?

I have ADHD and have done quite a bit of reading and study on it, so I'm pretty familiar with how dopamine and dopamine disorders work. I've also been in the workforce as a software engineer long enough to have done some really hard things. So my life and career have both been plenty challenging lol.

And I'm not alone here. Like I said, I was discussing this with a bunch of friends who are also quite senior and accomplished in their engineering careers, and the sentiment was familiar for us all.

> Did you have the kind of life where you were never really challenged to build that skillset,

Do you really think phrasing a question like this will ever induce a productive response?

I guess I could have phrased it better, but at some point I'm asking about weak self control vs if the drug is that strong. The life experience thing was meant as laying down a facesaving reason that it's OK to say your willpower sucks. You just weren't forced to cultivate it. Plenty of reasons that can happen in life.

I think it's pretty normal to be able to reflect on the difference in life skills between myself and those I see in others. There are things I've struggled with throughout adulthood because through some happenstance I was able to avoid the class of challenge as a child.

I didn't learn how to study until my 20s. I didn't have will-power over eating and exercise until my body changed around 30 and I suddenly got fat, then I talked with friends that teased me for being less skilled at something than a teenage version of themselves.

What's the saying: someone who's never smoked doesn't have to learn how to quit smoking?

I understand what you're asking and why, but the phrasing reads very dismissively and that's what I was asking about. Generally a friendly tone will get you a lot further.

I'm on board with your gut that this feels more YOLO than careful but to be fair, in the engineering world fly by wire is very much precedented. I'm specifically thinking of the B2 bomber where it's essentially unflyable without a computer between the inputs and the outputs. Partially just keeping the plane from turning into a frisbee by reacting faster than a human possibly could, but also treating the controls inputs as the intent and manipulating the control surfaces programmatically in order to make that work. It's not quite the same thing of course but I think there's some carryover.

Still. Not a huge fan of this announcement or the general ways the landscape is evolving these days.

This is a great analogy to what this time in IT and tech feels like. We are moving up a layer in terms of abstraction, and for those of us who cut our teeth in processes where we had lower level understanding it feels very destabilizing. I've been telling my team for a few years now that becoming an "agent manager" is the path forward - it's more true now with the latest revelation that "managers are out of style", and every role will have some IC component. I've seen it in my work - I can express intent more clearly to Claude and get immediate feedback for any technical tasks, so my team needs to be able to create intent at a higher level and translate that to their agent team, rather than getting directive task alignment. We've been through these pendulum swings before - this should start to stabilize in a few years... That's the industry vet perspective - I'd be lying if I said it didn't feel different this time though...

> We are moving up a layer in terms of abstraction

I’ll believe it when I don’t need to understand the code. Until then it’s just autocompletion with (a lot of) extra steps.

Wake me up when I can actually let agents do the IC work without having to be the accountability sink that gets fired when things go wrong.

One thing you and the OP are not addressing is that most of these modern tactics are also necessitated by the fact that building an air force, navy, or cavalry that can beat modern superpowers is just a complete non-starter.

I'm not so sure the F-35 is built for the wrong war as much as the war would probably call for the F-35 if it didn't already exist.

Honestly I feel sometimes like about the only thing they do successfully is hacking. Not just in the sense of breaking into systems that are assumed to be secure although also in that sense. They're just, highly effective at fumbling around with a hatchet until something works. We just happen to have version control and automated testing that generally makes that approach somewhat viable for the task of programming. But while I've been genuinely impressed at how much it can put features into a workable state, I've never been confident looking at its output that it's going to do more than POC quality at the current state of things. But it's pretty dang effective at that given enough time and a space safe to hack away and reset until the product looks close enough.

"Genius is but the capacity to take infinite pains."

You know, that's also true. I am where I am because I'm stubborn AF and just keep hacking on things until they work. Maybe one of the biggest differences is just ego, lol.