The Threads logo looks like a parasitic whipworm.

Apparently its called “Innovis”. I tried to find a few articles but they all read like press releases or AI slop.

There is a blurb on Wikipedia fortunately:

https://en.wikipedia.org/wiki/Credit_bureau#Consumer_reporting_agency

https://en.wikipedia.org/wiki/Innovis

Yeah, I’ve been fortunate enough to be offered those multiple times as well. I froze my credit with the big three agencies after the third or fourth breach. Recently learned there’s apparently a fourth agency now? Cool. And there’s hundreds of data broker sites…

As a settlement for the wrongful death of your parents you are entitled to 12 months of LifeLock’s DataScrub™ service!

I had a tricky time getting hardware encoding to work and it ultimately ended up being I needed to expose the GPU to the Docker container. The yaml config needed:

    devices:
      - /dev/dri/renderD128:/dev/dri/renderD128
      - /dev/dri/card0:/dev/dri/card0

Note this was on a low-end Synology NAS with some sort of crappy intel GPU, but it actually works now, I was surprised. I only mention because before this I spent lots of time messing around with the Jellyfin settings and only the logs tipped me off. Jellyfin loves to fallback silently to CPU transcoding it seems, which I guess is good, but make troubleshooting unintuitive. Searching for log errors online gave me this solution.

It looks horrific to me. Like a film prop from Cronenberg or Lynch. I think it’s the mix of mechanical motion, a material that reminds me of Jean Jacket’s stomach from Nope, and a structure like a severely prolapsed rectum. No way could I get off in this thing.

Maya and Motionbuilder run on Linux, but that happened before they were hoovered up by the monster. Autodesk just ignores that part of their portfolio. I know a few people who work/have worked on the Maya team and they’re talented, passionate devs, but management just doesn’t give a fuck about Media & Entertainment when Autocad and Revit are making so much money.

If you can find it, I keep a small bag of straight-up wheat gluten and I add a spoonful or two when I want to make stronger flour. A small bag lasts forever and a little goes a long way.

It’s not just tech. Gardening, DIY, cooking, and similar popular subjects have been completely destroyed by this crap. If I see an AI generated header image or thumbnail I immediately backpedal now because I assume that means the text is bullshit too.

The example stuck in my memory now is when I was trying to read about watermelon growing times and the article said they flower a week after germination.There’s now frequently this, “oh GOD DAMN IT *close tab*” moment when you realize it’s actually total slop. Like, “oh so this article is BULLSHIT bullshit.”

I see it as the continuation of a very old problem. Old school engineering didn’t have any standards until a bunch of people died over and over and the public demanded change. The railroads, construction tycoons, factory owners, mine operators etc all bitterly fought, and still fight, engineering safety requirements. Computer industries have continued this. They all oppose public action, hide negative information, and try to pin blame for conspicuous failures on individuals rather than systemic rot.

I think also because of the relatively less visceral nature of software catastrophes we don’t have a culture of safety. That’s not to say software errors can’t cause horrific accidents but the power grid going down and causing a dozen people in the service area to die is less traumatic than a bridge collapsing and sending a dozen people into an icy river. That’s an extreme example but my point is that humans undervalue harms that are seen as less acutely, physically brutal and software just seems more abstract.

Most of us aren’t working on power grid either, so when you start trying to quantify our software’s risks you have to speak to “harms” rather than just crimes like negligence, and then you expose this huge contradiction about how responsibility is allocated socially. Like, not only should engineers, pilots, and doctors have higher responsibility to prevent harm, but so should cops, journalists, politicians, billionaires, etc.

So the risks are undervalued and both intentionally and unconsciously minimized. The result is most of us who’ve seen the inside are quietly horrified and that’s the end of it.

I don’t know what the answer is except unignorable tragedies because that seems to be the only thing powerful enough to build regulations which are constantly being eroded.

I’m not really into audiobooks, but my mom is, and she’s lent/given me a couple. I think, for her, having a good voice actor is at least half of the experience, at least when she describes her favorite books half of her praise is for the actor.

Having listened to her favorites I can confirm the actors are really good. They are true professionals, far beyond what AI can do. AI can do commercial voiceovers, where there is purposely a single-note, unevocative tone. How can it do a shift in emotion across a line of dialogue as a character has a revelation? Or a slow change in personality as a character goes insane? Or slightly modify their voice as an angry, drunk father finally realizes he is pushing his daughter away, or his voice cracks when he knows the treatment is hopeless, or drops his guard when he remembers his old friend he didn’t recognize? Etc. Even the pauses can communicate volumes.

This is the emotional landscape actors excel at navigating but tech bros aren’t even aware of because of their terrible media literacy. So even if some “prompt engineer” was babysitting the AI it wouldn’t be nearly as good. Basically, just saying the words is only half the actors skill, they are great at analysis also.

Seems like basically every company is covering up crimes that happen on their properties, and lots of those are sex crimes. I have no data, just anecdotally it’s been almost every company I’ve ever worked for and the experience of virtually every woman I’ve known well enough to talk candidly about this shit. I’m not talking about “nice ass” comments either, I’m talking like, “blow me or you’re fired” type shit.

Not an excuse for Ubisoft, but it’s kind of like how Covid is now endemic so we’re like “oh well”. This disease is so common we apparently don’t give a shit. There was a brief window of hope with “Me Too” but then reactionaries shut that down.

I found the original blog post more educational.

Looks like these may be typosquats, or at least “namespace obfuscation”, imitating more popular packages. So hopefully not too widespread. I think it’s easy to just search for a package name and copy/paste the first .git files, but it’s important to look at forks/stars/issue numbers too. Maybe I’m just paranoid but I always creep on the owners of git repos a little before I include their stuff, but I can’t say I do that for their includes and those includes etc. Like if this was included in hugo or something huge I would just be fucked.

I worked in a place where you dialed in to the PA system, and NOT using your finger to hang up was a rookie move, since the rattling of the receiver was deafening over the speakers. Definitely worse to use a sensor.

I’m starting to think it’s something super specific to the particular hugo theme I’m using and how it wants users to insert custom js/css to get it all baked down into the right place in the final output. I’ll keep bashing on it, thanks for your help!

Edit: OK this is kind of hilarious considering the community I posted to, but I actually think it works fine but something about my Librewolf setup is breaking it. It works fine in Firefox and Chrome, and since I jump around between them as I work I just happened to test in Librewolf right as I made this change. Not to get too far into the weeds but I think I’m going to just go ahead with not linking cloudflare. Thanks again.

Thanks fixed. Interesting jerboa and the web version of lemmy are developed by the same person but using the “code” button in the web frontend only uses one backtick. That might be worth a bug report.

I’m actually trying to get away from github also, so maybe codeberg pages instead? This is a part of the process I haven’t done enough research into, I wanted to get the static site working locally first then “shop around” for hosts.

OK, looks like the image paths are correct. It’s something about the JS that fades them in. If I toggle the opacity property on/off then suddenly it works fine, until I refresh, so something funky is going on there. At least I know the structure is correct hugo-wise so it’s just a matter of tracking down the fade-in issue.

The issue seems to be with how hugo renders everything down into a /public directory. Somehow this is breaking the static images Lightbox uses to do prev/next/close. It’s a small issue and I’m sure the fix is something dumb, it just wasn’t obvious to me (the images appear to be correct). But sounds like it’s worth just debugging it…

Something about how hugo is cooking everything down into a /public directory is breaking the overlay images (like the next/prev arrow). I’m sure I can track it down but since I’m pretty inexperienced this will take me some time (at cursory glance all the paths seemed good, so I’m not sure why it’s broken).

I would also prefer to host it myself so maybe I should just do this…

You’re fucking incredible, Cowbee. I’ve watched you spend literally days patiently and politely responding to dozens of confrontational, probably bad-faith posters in thread after thread with nothing but solid information. I really admire it.