There are good reasons to hide what’s going on inside. Some are simply not prepared: https://www.instagram.com/reel/DIKJ6n-vSKt/

None that have come up in that time. I’m in my mid 40s now.

I’m personally on a combination of Adderall and Adderall XR and have had the exact same problem (except the insomnia, which I’ve always had). For me it’s about every 5 or 6 months, unless I skip it on the weekends then it’s about every 10 months.

What I do to counteract it is take occasional two week breaks from my meds, just whenever my workload becomes less critical or the side effects get unbearable. I’ve been doing that for about 18 years now, and it works for me.

This seems like one of the areas where LLM generative AI could actually be a good thing. It would not be difficult to have such a system desensationalize the media we consume. It’d still be propaganda, but it would largely be toothless. Propaganda relies heavily on appeals to our fears and other emotions. Put it through a sieve that filters most of that out and it becomes much less engaging.

I hadn’t thought about it until reading your comment here but I accidentally raised my GenZ kids this way. If I hadn’t done this I never would have seen my eldest heading down the misogynist edgelord path and intervened.

The only decent computer in the house was in the living room, and they had clunky retired elementary school computers in their rooms running Linux and Open Office so they could work on homework (no internet, USB drive to save homework and send from family computer). It wasn’t because I sought to keep them from using the internet privately, it was because we couldn’t really afford better.

It’s apparently PowerAutomate Adaptive cards.

It’s MS Teams with their PowerAutomate flows from Fabric. The limitation might not exist in the direct rest API, which I could have used through Python; but it’s a hackathon, and my other team mates know PowerAutomate. Faster if we each coordinate using what we’re good with.

Honestly no idea. It’s funny though. The API allows us to either read it directly from our lakehouse with the 28Kb limit, or allows us to encode it in a json object. It actually recommends using the json method if we want to send larger files… but then complains it’s too large if it’s over 28Kb 🤷‍♂️

I think it was probably originally only intended to allow attaching icons.

Literally just today solved a problem of delivering analytics plots over our internal chat system. The file size limit is 28Kb and I was just getting ready to say screw it, can’t be done.

Lo and behold our chat system that doesn’t support svg does support webp. Even visually complicated charts come in just below the size limit with webp.

I think the proliferation of videos as primary information sources is a huge part of how propaganda and disinformation became so effective and powerful. It’s why we’ve done a collective nosedive into regressive politics and can no longer agree on the objective facts regarding… well… anything!

Information delivered by video tends to be trusted on the way it’s delivered rather than the content itself. So we’re thinking less critically about what we choose to believe.

I love seeing comments like yours where people change some of their thinking as a result of online discussion.

I was starting to feel like no one listens to understand and only listen to respond, and comments like this help lift me out of that perspective.

Half of an ortholinear split keyboard and a trackball that’s missing the ball.

I showed this meme to my wife and she showed me this in response. It’s the second result in a search for Trump Hair Shoes

I looked into it a little using TinEye and the image predates generative AI images of this quality. I actually think it’s a photoshopped normal cat; possibly as part of that weird trend of making some small detail of a normal photo look freakish.

Full context to that panel:

First panel of that comic here