Clojurians Log v2

that happened to my mac mini a few years ago. I really need to put linux on it

This is a work computer, unfortunately… and the update was mandated by the IT department.

well, at least you know who to complain to? (it sucks having work kit that doesn't work)

I guess

they don’t care, though, but maybe I can convince my boss to buy me a new one

yeah, it sounds like new computer time (I'm a bit surprised they aren't refreshing you around every 3 years)

5+ years is the policy here, though you can get around it, so I guess it’s time anwyay

They will sort out the performance issues eventually, but this tends to happen on some major updates. A bit soon for a mandated update in my opinion.

the solution for my old mac mini was to replace the spinning hdd with nvme or ssd, so it doesn't always get sorted out in software

I've got another older MBP that is pretty slow as well as it only has 8GB of RAM and that has never sped up

@hkjels Yup, but that’s just how the IT department at the University of Copenhagen does things. It also infuriates me. If we don’t update within 5 days of any new release the computer will be randomly possessed and update itself.

@otfrom This is a 2019 Macbook Pro, so it’s the very last version of the intel Macbook Pro to come out (and it has an SSD of course). I guess Apple is just biased in favour of their own silicon now.

Despite the mess that Apple apparently made with the last macOS, silicon is a significant upgrade, so much faster

There's zero justification for any use facing software being slow even on ten year old hardware. Negligence and bad code.

I don't disagree with that

It's going to get worse

an absolute classic water cooler moment though @simongray 👏🏼

it gets worse before it gets better. no, it just gets worse. ;)

It gets worse for 200-1000 years, then maybe it starts getting better, if previous occurrences are any indication

worse and better are concepts that exist in many dimensions rather than a single continuous timeframe but yeah, entropy will always eventually win

anthropy

Just saying it's time to get into CollapseOS

All relativists suddenly believe in objective reality when they stub their toe on a corner

Ironically, inability to agree on what worse is is an indication things are getting worse for those who think there's no argument about what's worse

Wow. Its a thing. I had not seen CollapseOS before 😆

It's a pretty dope project

I 😍 that DuskOS is here for the prelapsarian among us

You have to be eased into it 😉

Forth-play?

So, learning is off the table now?

I’m not really interested in becoming a D3 expert. I know enough.

It’s not like the original code was based on my awesome D3 skills anyway. I hacked it together from some existing example code… like how 95% of D3 implementations in the wild are probably made.

Sure and you're probably right. I still wouldn't trust an AI to improve it cos it's working from the same principles: it has no idea what works.

OTOH you seem to be getting value out of it so it's probably a hard case for me to argue

If it makes a difference to you, Ray, I was a huge skeptic about it before, but after incorporating it into my workflow I have a more balanced view. 🤷 It’s not like it’s a panacea, but as someone who works entirely alone on fairly big projects, I find it really comforting not having to be an expert in every single domain for once.

This is the improvement to the diagram rendering that I implemented yesterday (before vs. after). I had wanted to do something like it a for long time, but the thought of having to read tons of D3 documentation and attempting a bunch of (possibly failed) repositioning calculations was holding me back.

I don’t understand the difference but I see that some things changed and I’ll take your word that it helped. My biggest problem is that we’re building systems that are so complex that we need these tools to - like you say - read the docs and help us to avoid “stabbing in the dark”. It seems from what you say that you’re still stabbing in the dark but somehow hitting more targets. I’ve got an uneasy feeling that this trust can creep up on us and reduce our overall understanding, even as we get things done in the short term.

I share those concerns (and I have many more, e.g, digital sovereignty), but it doesn’t reflect how I am using it at all.

And today it is writing a bunch of dumb code, so you’ll be pleased to know that I have taken matters into m own hands 😛

hehe KISS to you 😉

btw I am discussing my many issues about AI, along with @jmglov and several guests in this (relatively new) podcast https://politechs.dev/ ... and each episode has transcripts if you prefer to read.