r0uv3n

Grad students do not solve problems such as the existence of non-sofic groups every day.

This is not about preventing crawling entirely, it's about finding a way to prevent crawlers from repeatedly everything way too frequently just because crawling is just very cheap. Of course it will always be worth it to crawl the Linux Kernel mailing list, but maybe with a high enough cost per crawl the crawlers will learn to be fine with only crawling it once per hour for example

Eh, the social features of Komoot were never intrusive to me, and among social features of most apps they were some of the most well designed. Local community, very much focused on actually sharing tracks and trying out other people's routes (and maybe commenting with your experience afterwards).

There was a guy in his 60s regularly doing very nice circular hiking routes of 40 to 60 km in our nearby forests, and apart from that just being kind of awesome and impressive to see when you look at local routes, actually walking his routes was often a very nice experience with diverse landscapes often along nice small, less used paths. It was great seeing nice weather in the morning, and then oftentimes without any pre-planning just walk or bike to the forest and just start along one of this guy's routes within a few minutes, all in an incredibly hassle free manner and with a result which pretty much always beat out just following the official hiking trails shown on signs etc. I don't know if there's another app right now where you can so easily profit from the experience and knowledge of your local community.

Well, Komoot worked quite well for exactly that use case. I have also only very rarely found tools even in the desktop space that were quite as mature as Komoot for that use case.

Also the question remains, what do you navigate the planned routes / gpx traces? What happens if you notice you want to improvise and replan to hit some target on the way you saw in the distance while on the trail? This was (and currently still is) absolutely trivial and intuitive to do on Komoot. The best alternative I can think of is maybe brouter+ osmand, but that's really quite clunky in comparison with Komoot (similar to the experience you probably mean when talking about pulling your hair out)

That is about something entirely different. It more or less just says that energy might be lost if you have a flux towards infinity. It does not in any way claim e.g. that the divergence of the stress energy tensor is non-zero (which would be how I think most people would interpret energy/momentum conservation).

Eh, space inside or outside the horizon is only different in so far as to whether it can reach our timelike infinity. Locally you cannot even tell where any horizon might be (just look at a small patch of a Penrose diagram near a horizon), they are very much something related to global properties of the spacetime. In particular it's not problematic to talk about some extended volume in spacetime occupied by mass, as long as the divergence of the stress energy tensor is 0.

The point where our notions of geometry would break down would be near the singularity, not near the horizon, and we don't even know if a volume enclosed by a horizon (i.e. anything you might call a black hole) necessarily has a singularity inside, it's just that our simple mathematical models all assume one.

I think this is enough of a classic to be widely known even among younger people. I'm 23 (doing math msc) and I think all the CS people that I know would instantly recognize the 500 miles title.

Though I do somewhat envy the possibility of having read the article close to publication and feel in some sense part of the history when it crops up again like this.

2 for example is not in the sequence. Remember that you need the first differences to this sequence to obtain all natural numbers

Have you tried Plasma 6 yet? To me it feels way more modern than Gnome.

If you couple your system to a heat path that is at rest wrt a specific Lorentz frame, you of course lose Lorentz incariance. On the other hand the lagrangian of the standard model itself is to my knowledge fully Lorentz invariant.

Arguably Bayta is the lead in The Mule maybe? But of course the basic plot was very much influenced by Campbell still

AFAIU Firefox Sync is end to end encrypted, right? I interpreted the language here to mean that the data might be sent to Firefox outside of Firefox Sync. Have there been any changes to Firefox Sync that we know about?

This is a fantastic resource, and I used it heavily in 2018 before starting my BSc in Math while still in school to learn the stuff that interested me a bit in advance (and to some extent as a familiar reference for a few years afterwards).

I can only recommend this as a good starting point for anyone without the time for a full scale education, or in preparation for such, especially if you have some experience with math olympiads (as mentioned also in the introduction).

I only wished I had spent some more time with stuff like category theory before starting my studies, and had had the guts to take more advanced courses directly in my first semester.

With just a bit of prep from these notes I'm pretty convinced that it's possible to directly take e.g. Algebraic Topology, Differential Geometry, Category Theory, or Algebra during the first semester (don't know about number or measure theory, or anything requiring lots of functional analysis, have not engaged enough with those topics to know how good these notes would be as prep).