Can someone explain to me what is the difference between whatever they added new and creating several PRs and "stacking" them by pointing one branch to another via the Github UI? I've always done that. Is it just a nice UI on top of that?
Sure merging with one single button is nice but I rarely do that, if I'm stacking PRs it's because I probably want to merge gradually.
Maybe it matters if you do not have the right to push to the main repo but still want both your PRs to live there. Is there a way to do that without stacked PRs?
Most people crave change and novelty and flat design provided just that. I'm now anti-flat design but I remember vividly thinking Windows Vista looked awesome and then some years after that thinking Windows 8 was pretty cool.
Now I long for skeuomorphism, but if that becomes the norm for 10 years... after that who knows?
I've no idea on the state of research about this topic but I've always wondered if this really has to do with the locomotion side of it or if it's more about the fact that writing by hand takes a bit more effort and time.
For example, for taking math notes it's easier for me to write by hand rather than take notes in my laptop (because writing LaTeX takes time). If I spend more time on my notes in my laptop, would that improve retention?
I don't. Never understood how people need more than 10. Usually you have a few websites always open like gmail etc. and whatever stack of 1-6 tabs for whatever you're doing or researching right now.
If you're holding tabs more than a few hours those should be bookmarks (I don't use those either, I just send myself links via Telegram or keep in notes app).
I have a strong feeling that with the advent of AI these kind of games are going to come back in style. Many programmers myself included aren't doing that much "coding" in the workplace anymore.
For me, it’s less that I’m burned out on coding and more of a feeling that if I’m going to be doing puzzles in assembly I mine as well just do it in real assembly.
Usually you get both kind of assignments, short exercises in paper without any help whatsoever and more open, larger take-home assignments where you can usually use external help.
I very much prefer to work in Rust but it does force you to design things in an counter-intuitive way in certain situations (linked lists for the canonical example). Specially for lower level software I find C or C++ to allow for much more flexibility and thus the most straightforward design possible.
I really don't think Rust does, but that's if you're outright refusing to use `unsafe` anywhere. Might be a little noisier if you're using things like ManuallyDropped instead of just using raw pointers, but the language doesn't stop you from doing things in ways you would with C or C++, it encourages you to encapsulate the hard parts so you can worry less about correctness of 98% of your codebase, instead of fearing incorrect behaviors can silently sneak in.
That's absolutely correct. Wild to see it downvoted without explanation, while you preemptively mentioned the tradeoff that you do need to worry about correctness in that case (just as much as in C/C++), but that only applies to a small part of your codebase, so it's still a huge benefit.
The only downside is that unsafe rust is more verbose than C/C++.
Hey this is pretty cool, I think probably the best iteration of these I´ve seen so far.
I´m surprised to see so many regional channels in my country I´d never guess existed.
I think a nice addition would be a filter to easily see the most relevant/national channels of a country, so I don't have to scroll through many regional ones.
Sure merging with one single button is nice but I rarely do that, if I'm stacking PRs it's because I probably want to merge gradually.