Benjaben

Lol yep sounds a lot like my process! Took time to get it down and settle on tools (though those always changed anyway) but once you did, could make a buncha money for sure. With KVMs I could do a lotta volume on those kinda jobs and get some of my engineering homework done in between. Hardware repairs were more fun but way more time consuming and hit or miss depending on overall condition.

Not a bad gig overall but certainly did come with some downsides. Like, desktop computer filled with insect carcasses, brown everywhere with tar from cigarette smoke, stinking up the shop, customer somehow oblivious to the gnar-bomb that is their daily life intersecting with "ordinary" society.

Certainly seems to lend itself to automation way more than actual opinions. Set a bunch of measurable conditions tied to generic article prompts that relate to that condition ("late viewership surge" in this case or similar? Didn't read it lol), and then just run a routine that watches the metrics for all big IPs, checking each for your list of conditions, and let it fire away.

Devil is in the details and I'm not claiming that's an afternoon's worth of work for something convincing/ sophisticated, but what I'm describing is ultimately just quantifiable inputs and outputs with some "LLM window dressing" so it feels natural to readers. And of course the articles end up feeling thin and cheap as a result.

Edit: I should add, this is just in reference to discussion on metric-centric articles in general, not the one in the OP (which doesn't look AI-y at a glance)

Definitely not wrong! Especially once you've dialed in your routine of anti-malware utilities to run on pretty much everything. It's like an antibiotic cocktail, lol. Or did you prefer the "back up and nuke on sight" approach?

Oof, couldn't help but notice my subconscious response to that line as well.

Yep, I did similar around the time. Can't blame people for being mad that the thing they bought is damn near unusable (and was destined to be, but they didn't understand that part). If someone buys a new bike, even if it's cheap, it shouldn't roll like you're on gravel after a couple weeks and become impossible to pedal within months. But damn, there were a lot of horrible machines sold in those days.

And then of course, the least fun part of that era, the guys who would bring their machines back weekly despite very stern warnings to stop visiting "those sites".

For some reason your comment showed up to me initially as "Deleted by creator". Which was kinda hilariously apt.

Ah, gotcha. Yeah I haven't had great luck with that one for dev work either.

Which did?

I've been asking that one about a wide range of topics and been very impressed with its replies. It's mixed on software dev, which is to be expected. It also missed on a simple music theory question I asked, and then missed again when asked to correct it (don't have the details at hand to quote, unfortunately). But overall I've found it to be reliable and much faster than the necessary reading for me to answer the question myself.

How've you found Claude?

Awesome, thank you! This largely matches my own experience, I've found it (Claude in my case) most useful in areas where I'm weakest. I haven't tried this scaffolding-via-comments approach though, it sounds cool.

Any experience with Cursor or other IDEs or agents? Was co-pilot a choice or just kinda a natural default?

I felt the same, just flat out terrible looking. Just completely foolish, juvenile.

But an almost instant second thought was about how impossible it'd be, in that time and place especially, to keep those big white...stretches...continuing to look anything close to white. Civil War being the messy thing it was, Confederacy in particular not exactly known for their sharp discipline and order, lol...

Bet those things* looked like total ass, flown.

Almost makes for a funny metaphor, somewhere. Can't quite get there myself, but something to do with their big, badly designed white spaces getting browner all the time 🤷‍♂️

Would you mind sharing a bit more about the workflow you're describing? I'm on a "ask people how they're using AI to help them dev" kick.

Sounds like you're using an agent integrated with your IDE, would you be willing to give specifics? And you're talking about writing some comments that describe some code you haven't yet written, letting the AI take a stab at writing the code based on your comments, and then working from there? Did I get that right?

Happy for literally any elaboration you feel like giving :)