Anyone can learn to code well enough for a corporate environment.

As the repo owner, you can put in place PR guardrails to help you manage the workload it puts on you. You can enforce pre-commit linting and code formatting, mandatory PR templates, size limits on PRs, etc and these can limit the chunks of work you’re sent by this person.

This is part of creating a culture of good code, enforcing code standards and contribution behaviors comes with the territory as you move up the chain in your career.

Another part unfortunately, even if you’re not a supervisor is having sometimes tough conversations with contributors it’s just part of the deal.

“Hey Bob, I just wanted to connect with you. It seems like you’re having kind of a tough time keeping up with our standards (producing code that’s usable for our team, or something said tactfully like that), is there something more that I can do to help you? Or is there something specific you’re having trouble with? I just want to help you be the most successful that you can be, because the more successful you are, the more successful our team as a whole is.”

If you have a discussion or two like this and it’s not working out, then maybe you need to talk to Bob’s supervisor/manager directly about the issue. Sometimes people don’t even realize what’s going on.

“Tech writers learn about steganography, more at eleven.”

Is your upright the one with all the little compartments? That one looked to me like the most efficient upright design I’ve ever seen.

They’re pissy cause it being open source and more efficient means that it’s gonna be more cost effective for people to use. Which is real bad if your company overcommitted to the slop and needs to recover losses.

I’m a Minnesotan, currently at my kids hockey tournament. This is an acceptable scenario.

I bought a protectli awhile back. Mines 4 port 2.5 gbps nics, and it runs opnsense out the box.

You should take a look at their sfp+ model, if I were in your shoes that’s what I’d be looking at. It’s all in one, works nicely, is incredibly customizable, and is lower power usage than basically anything you’ll build yourself.

I use that for my router/firewall, then I use an off lease dell thin client to run my home assistant server, and a standard off the shelf buffalo nas. If you’re into immich, I’ll recommend jellyfin over Plex. I used it for years but they started collecting more data, sticking their own junk in etc. Jellyfin is open source and works great.

I’m glad to hear this, I keep looking at it. I’m a little worried about the hardware requirements for it though. I keep putting off getting hardware for a dedicated jellyfin server (it’s on my wife’s gaming PC right now and it bugs the heck out of her)

To expand on the 3-2-1 rule for the uninitiated:

3 total copies

2 onsite, using dissimilar media

1 offsite, for disaster recovery

Is it not safe to expose externally with ssl yet?

Ubiquiti has had outages in the past that meant you couldn’t manage the equipment right in front of you.

Even discounting the potential security implications of that kind of management, the rage I would feel in that situation is enough that I while my AP is nice, works great, I will never use any of their gateways.

I’m using a 2.5g protectli with OPNSense now, and it’s easy to manage, and all local.

It would be a crying shame if someone were to figure out a way to force those e ink displays to refresh fast enough that it kills the batteries on those things…

So how long till the Supremes rule that CBP is not only allowed to search your phone but also to perform colonoscopies at checkpoints 100 miles inland and sell the resulting videos to extremely wealthy perverts?

Beyond any issues with the owner of the company, these cars have multiple dangerous issues.

You cannot treat a company that makes physical stuff that can endanger lives the same way you treat a software company that makes a leisure activity platform.

Iterative design for a purely software environment is way more forgiving than iterative design for physical hardware or even software that interacts with physical hardware. You can profoundly fuck up the backend for a website and take the whole thing down until you could roll back to last known good production, you won’t kill anyone, but you’ll make the line go down temporarily.

If you profoundly fuck up an iteration on an embedded vehicle system and don’t catch it because you don’t respect safety regulation or existing engineering norms you can and will kill people.

Analogue doesn’t have firmware that can reject a device based on id.

So you can reverse engineer a replacement part if you absolutely have to.

It’s not the iPads themselves, it’s the addition of Bluetooth and/or wifi to support them. I agree that they can alleviate a lot in terms of paperwork reduction etc. My issue is the additional exposed surface.

It doesn’t, that’s just a very common reaction to these types of articles. I recall having some very intense discussions around stuff like iPads in cockpits. I’m on the “not a fan” side, but I’m also not making avionics software anymore either.

Certification is expensive. But updated dbs are pretty huge and seem to only get bigger over time. Stuff like radio firmware tends to be in the hundreds of KBs though, so for that it really wouldn’t be a big deal either way.

These should be USB sticks, but otherwise this is preferable to something like wifi.

You do not want to stop requiring physical access to avionics for updates and reprogramming.

The fewer surfaces for entry into the avionics systems the better and if that means an engineer schlepping a database update on a thumb drive to the cockpit that’s what you want.

I spent the better part of a decade on avionics, and while this as a headline sounds bad it’s one of the few things Boeing shouldn’t be mocked for right now.

It’s a surprisingly good comparison especially when you look at the reactions: frame breaking vs data poisoning.

The problem isn’t progress, the problem is that some of us disagree with the Idea that what’s being touted is actual progress. The things llms are actually good at they’ve being doing for years (language translations) the rest of it is so inexact it can’t be trusted.

I can’t trust any llm generated code because it lies about what it’s doing, so I need to verify everything it generates anyway in which case it’s easier to write it myself. I keep trying it and it looks impressive until it ends up at a way worse version of something I could have already written.

I assume that it’s the same way with everything I’m not an expert in. In which case it’s worse than useless to me, I can’t trust anything it says.

The only thing I can use it for is to tell me things I already know and that basically makes it a toy or a game.

That’s not even getting into the security implications of giving shitty software access to all your sensitive data etc.

I try to not buy any Wi-Fi smart home devices anymore. I try to stick to zwave or zigbee, zwave I have better luck with generally. I even left my nest thermostat at my old house and installed a 10+ year old zwave thermostat at the new one. Way happier. I’m not relying on googles API to be stable anymore for home assistant interaction.