abbefaria27

This is very different and doesn’t relate to what the linked repo does.

In my experience having the agent explain the code doesn’t work very well for real world apps, even the parts written by humans. For example we tried using to generate diagrams, class hierarchies, etc as part of documentation. If you don’t know the code it looks great. If you do, it’s focusing on all the wrong things, missing the mental model, and ignoring lots of important bits. And Claude tends to be extremely verbose to the point of muddling things.

All these trackers (rings, watches, wrist bands) seem so… useless. Do you really need to know your heartbeat every minute of the day? What does sleep tracking tell you, can’t you just tell if you’re tired? For athletes there are garmin type watches, which I understand during a workout. But the lifestyle trackers I don’t get the point.

What’s the story for Rust-C++ interoperability these days? That’s what kills adoption. Most C++ devs I know like the idea of Rust, but no one is going to go rewrite 30 years of working code. It needs to be something you can incorporate gradually.

If you learn to play, over time your ear will improve and you won’t be tone deaf anymore.

To be fair to the committee, upgrading C++ compilers is already a chore, even with the changes as small as they are. Would you rather spend time updating your lambdas because they changed how capture works (a real breakage not long ago), or doing something of actual value? The only reason we finally updated from Python 2.7 to 3 was that the OS vendors dropped it, but it was effectively a waste of time.

That aside, I don’t understand why ABI breakage is a big deal. There’s probably a good reason, but ABIs are already so fragile, you can’t mix and match different compiler versions anyway, or even the same compiler with different flags. Why does it matter if setting -std29 or whatever breaks ABI when -fno-rtti does the same?

The C++ committee should really stop adding esoteric features for library writers and focus on stuff for normies. Tooling is one as the article says, or how about finally getting Networking TS? This is probably the only mainstream language left that doesn’t have even basic networking support.

I spent quite a while trying to learn Common Lisp and I really wanted to like it, but I’m not sure I’d recommend it. The first problem is finding good learning materials, as most are out of print. On the language side people complain about the parens, which turns out is no big deal. There’s a lot of other weird or needlessly tricky stuff though. The function names are convoluted (e.g. there’s map, but also mapc, mapcan, mapl, maplist). Image based development was interesting, but you might waste an hour debugging because your image diverged from the actual code. Building an actual executable is more complicated than you’d expect, and there’s a lot of magic with packages. There’s some elegance there if you squint, but other lisps are probably better.

Setting aside the “why” that everyone is so focused on, I want to know, how? I can’t get Claude to do anything even a fraction of this complexity. How are people setting up their agents, Claude.md, etc to do such big projects? There are a lot of lessons to be learned.

I didn’t realize IEEE had courses. I’m curious if anyone can comment on the general quality and if they have any good ones.

I don’t understand why the Steam Machine is getting so much hate. The form factor is amazing and a huge part of the value proposition. It’s slightly bigger than a box of Kleenex and you can put it next to the TV, on your desk, etc. I’m tempted to get one just to free up the floor space used by my ATX case. The graphics capability seems good enough, most people can live without 4k gaming.

It’s covered in the risk factors section at the end.

It does get easy to read, but then you unlock a deeper level of misery which is trying to work out the semantics. Stuff like implicit type conversions, remembering the rule of 3 or 5 to avoid your std::moves secretly becoming a copy, unwittingly breaking code because you added a template specialization that matches more than you realized, and a million others.

I tried setting up mu4e once. It wasn’t worth it. It took me literally a few hours of reading random blog posts to figure out the configuration, and that was only to download email. Never got around to setting up sending them, which is a totally separate process. Even then, there were lots of issues. First, it’s slow. Loading an email had a noticeable pause and was slower than GMail. Also, you can’t avoid HTML email nowadays. There’s a very basic render, but expect all the formatting to be wrong. I also ran into rate limits from Google because we get way too much email at work. That’s not mu4e’s fault, but just another obstacle. Can’t really have my inbox be one hour behind real time.

The biggest tip is to read fun books. Pick up an easy dumb page turner like The Da Vinci Code, and reading won’t be a chore. Save the non-fiction and literary fiction for when you get more in the habit. Also, it’s ok to give up on a book if it’s not interesting.

I was on the same boat as you. The “Learn Fusion 360 in 30 Days” series on YouTube is awesome. In the first video you make a working Lego brick, and after a couple of hours I could make my own simple parts. Complex objects still seem hard, but simple household 3D prints turned out to be easier than expected. I started with OnShape but switched to Fusion just so I could follow this tutorial. I think (hope) the skills will be pretty transferable to any CAD.

https://www.youtube.com/watch?v=d3qGQ2utl2A

Amazing write up! Everyone at work is itching to try Rust, but I think what’s killing adoption is that it’s not very clear how to gradually transition a code base. We have a few million lines of C++, some of it written 25 years ago. A full rewrite is just out of the question, at best we could use it for new sections. This is super common in the c++ world, so it’s a pity that porting wasn’t a first class concern in rust considering C++ devs are the target audience. It sounds like it was a challenge even at 57k LOC. Congrats to the fish team though, great accomplishment!

There was some article about it on HN a while ago. If I remember right the problem was that its bioavailability is super low. You can take all you want, but only a tiny percent makes it through to get absorbed. In theory you could increase the the dose a lot but I imagine that might have other issues.

Many negative comments here, but the notion that food might be less nutritious is certainly interesting and worth exploring, even if this article isn’t the end all be all. As mentioned in there, apples are sold up to a year old, and I know the same is done with potatoes. We have global issues with soil quality and depth, produce is selected for transport instead of taste (which is why grocery store tomatoes are gross but garden tomatoes are amazing) and other practices like breeding chicken that grow twice as fast as they used to.

I discovered Rory Sutherland this year and he’s awesome. I recommend to anyone checking out his TED talk or book.

In the past 3 weeks I blocked, no exaggeration, about 350 of these same text messages. They’re almost all for republican fundraisers, and almost always either unscrupulous or borderline fraudulent. Stuff like “50X match if you donate now!”, or “our records indicate you’re voting for Kamala. Click now to set the record straight!”. Sending “Stop” or marking as spam does nothing of course, so I had to get a blocker app. I did finally click on a couple out of curiosity (how much worse could the spam get) and they use every dark pattern there is to try to force a recurring donation. I don’t really get how it’s worth it, but I guess some people must go for it.

Just for another data point, I haven’t gotten Wayland screen sharing to work reliably at all. I tried it in Fedora 35 and Ubuntu 22, where Wayland is the default. On Fedora it just didn’t work, and on Ubuntu, Firefox would crash every time. On Chrome it sort of worked after some updates. For Slack you needed to launch from the terminal with special environment variables. There were also weird bugs around docking my laptop, requiring hard resets. All in all, for any sort of modern remote work it wasn’t usable. I switched back to X11 and it’s been smooth sailing. I do miss the fancy trackpad gestures.

The author just needed to get a second opinion. I showed this to a veterinarian I know, and she said this is pretty obvious from a quick glance at the bloodwork. In fact she's skeptical that the first vet didn't see it, and rather thinks there may have been some miscommunication. You can look up anemia in a reference and it lists the same differentials. Though according to her, ChatGPT didn't present them in the most likely order (e.g. babesiosis should be lower), and there are many other factors to consider.

This is still very impressive for a computer program, but not as mind-blowing as I first thought when reading the thread. ChatGPT didn't find some obscure disease like in a medical TV show. Rather, it correctly read the low blood cell count, and pulled up the differentials for anemia from a reference book.

On a side note, considering how often ChatGPT will lie with full confidence, personally I can't imagine using it for anything medically related.