Notices by Ian Z (soviut@hachyderm.io), page 3

@thomasfuchs I'm kind of afraid to ask, but I'm assuming this is something around DHH?

@thomasfuchs and don't feel obligated to use projects started by fascists just because that project is somewhat popular; there's always an alternative that's usually better.

@thomasfuchs Who needs LLMs when this summarizes DHH perfectly!

@thomasfuchs @saxnot I know what ray tracing is. I'm saying that real time ray tracing hasn't been commercially available until very recently, and at great expense.

The Quake Wars demo ran extremely slow on the fastest hardware of the time and was, ironically, CPU bound. It wasn't hardware accelerated.

These were all bespoke implementations, not "general features" of GPUs at the time.

@thomasfuchs @saxnot 20 years is a major exaggeration. Real time hardware ray tracing has only existed in GPUs for 6 years if you're counting the RTX series and those are dedicated graphics cards.

While this isn't necessarily ray tracing, it's still a physically based shader pipeline. From a technology perspective, that's pretty impressive to have in your GUI compositor in an OS.

@thomasfuchs DHH is the same twat who wrote something similar in support of the "trucker convoy" here in Canada. It was started by bunch of racist groups and dressed up to look like a "protest".

It's especially hilarious with him giving his opinions about a young country like Canada. Anyone here with white skin descends from immigrants. Meanwhile, the actual natives were chased away, then put in residential schools, then forced to live on reservations.

Maybe these white guys worried are worried about being a minority because they think they'll be treated the way they treat minorities.

@thomasfuchs Yeah, that's fair.

@thomasfuchs The requirement of a TPM module is a good thing for security, even with Linux. I just wish there was more talk about how to install after market TPM modules on older machines. Instead, most of the chatter was about how to work around the Windows requirement.

@thomasfuchs Directly overlaps with EB Games vs Gamestop map.

@thomasfuchs lol, yeah, debate in bad faith arguments, spout antisemitism and christo-fascism, then hitched his wagon to a right wing grift. Poor him.

@thomasfuchs Not really seeing the "sports" aspect you're talking about unless you're still referring to that "MAX" font they used? That, or the "look, this thing is dust and dirt proof" so you can use it to shoot a movie while debris hits your camera operator's face.

As for other reasons to get a Pro, the Pro Max for the bigger (frankly, better) screen.

As for the camera, yes, that's been the differentiator. The Pro can shoot in raw formats as well as ProRes for video. In the case of 17, each lens is now 48MP, and it has an 8x optical zoom on the extra telephoto lens, etc. And faster USB-C port for transferring footage.

The regular iPhone 17 has a decent 48MP camera, just not telephoto. No ProRes support for video, etc.

@thomasfuchs Anecdotally, I've seen plenty of non-programmers say things like "I'm not a programmer, so I used ChatGPT to help whip up this python script". Dozens of hardware youtubers use it for this specific reason.

Meanwhile, the programmers aren't using it because they yearn for simpler times; we're in the simpler times. They're using them to do drudge work. Porting and decompilation projects have seen dramatic improvements, for example.

My original point was about vibe coders being just like the other art, music and writing slop peddlers. They don't see this as another tool, they want it to be a replacement for those pesky programmers/artists/musicians/writers who dared to charge money for something they couldn't do themselves.

@thomasfuchs I disagree. They want LLMs to generate code because they resent people who took time to learn a craft and they see this as a low effort shortcut.

Art hasn't become too complicated, music hasn't become too complicated, writing hasn't become too complicated. Yet people are generating that too.

@thomasfuchs I generally agree, but I use it mostly when there isn’t a framework. I don’t have the luxury of telling a client to rewrite their site or a backend cloud service only supports micro frameworks. I need to write similar, but all slightly different tests for each endpoint so I just use it as a more capable find and replace.

When I say boilerplate I mean drudge work like writing a mapping between ISO 3 digit country codes to 2 digit country codes.

I checked for a library but the only one close was just 2 digit codes. Funny enough, I contributed my mapping to them as a PR.

@thomasfuchs Were any of those apps a social network? heh.

I agree that it can be made fast, but social graphs can be notoriously difficult to optimize.

I'm mostly just giving them the benefit of the doubt.

@thomasfuchs "no reason" is a bit of an over statement. For example, I'm in Canada and my server is across the ocean, so even getting the initial assets can slow things for me.

Next, these are running on community run instances that could be a laptop hard drive for all we know.

Those instances do more than just serve content, they collate, manage accounts, etc.

Those servers are running a rails app so there's some overhead built in.

That rails app is getting older and maintained by a small team.

The frontend code is starting to get archaic and has had pushback on switching to a framework like React. So fewer people can easily contribute to it.

Things like privacy and moderation tools have been a high priority for a while; performance less so. The small team's time is limited.

etc.

All of these things are solvable, but it's a problem of time and/or money. Something that is is short supply when it comes to Mastodon.

@thomasfuchs move that research to Canada. We love breakthroughs like Insulin and can actually keep our drug costs reasonable.

@thomasfuchs that’s a real bummer.

@thomasfuchs I'm not sure many people are very happy with the summaries. Even non-technical people I know are frustrated with how far down you have to scroll to get to a real result link.

@thomasfuchs "...and we're naming the last one Icarus"