I run Wireguard in combination with Pi-Hole so I can VPN into my home network to configure anything I need. DuckDNS if you’re on a dynamic DNS provider. It’s also nice to have this since you can get the adblocking when away from home.
> "Another issue with all the 'BS' jobs in large corporations is that it takes profits away from shareholders who are most often the pensioners and retirement accounts of the rest of America," he said. "...they are also taking money away from the rest of the workforce's retirement programs."
That's ... basically every old person and everyone with some savings.
Even if you don't own Google stock (or even any stock) directly, you likely have exposure to it because it's in the S&P 500, which is used as foundational financial product in many contexts.
Yes, I'm mostly making fun of the fact that this extremely wealthy investor is pointing at the elderly (who likely have a higher percentage of their retirement saved in bonds and less-volatile investments) and distracting everyone from the fact that he benefits massively more if Google decides to make shareholders more money.
- that means corporations have a moral duty to do their best
- if an old person entrusted my company with the financial results of their life's work, I'd feel a lot of pressure to make good on the implied promise that I won't just take the money and put my feet up
Agreed but in the case of retired people, that duty hits different. They don't have the ability to go back and earn more money; many are also in cognitive decline.
The growth of LA was also kickstarted by the 1906 SF earthquake, which burnt down ~80% of SF. If you look at the population graph of LA, it tripled between 1900 and 1910, probably due to the relocation of so many people from SF to LA. Here [0] is a paper that discusses the effects of the 1906 earthquake on population growth on the west coast.
Our team went with a similar approach when refactoring Protobuf debug APIs (https://bughunters.google.com/blog/6405366705946624/fixing-d...). People were relying on debug output and trying to parse it, so in the new implementation we threw up big warning flags and made the output unstable so that you couldn't make the mistake.
The key lesson to take away is: if you want something to be an implementation detail, make sure to have multiple differing implementations :)
In defense of the companies that do this - making native apps is extremely expensive compared to just using Electron. If you want to have a native app for Mac and Windows and a website, that's 3 separate frontends that you have to propagate changes across. This triples all the work that you typically have to do (UX changes, platform-specific bugfixes, etc), and getting some fullstack developers are cheaper than hiring Mac/Windows specialists.
I think HN users are in the minority of users who seriously care about speed/performance. Electron is good enough for pretty much everyone else, and it's clearly working well for for Discord and Slack. It makes more sense IMO to invest time in improving Electron efficiency rather than telling companies to just avoid it.
By the point you are using QML I don't think it's actually that much lighter than Electron. QtWidgets is probably justifiably better in this regard. Not all electron apps have to be as bloated as Slack, for example. It's possible to design sane applications.
No, it's still not the same category. QML doesn't pull in a full featured Chrome with all its runtimes, dom, full scripting, multiple levels of JIT, etc. You're still significantly lighter. Even Avalonia with bundled .net is lighter.
QML is very efficient. Much of custom QML code ones write actually compiles to C++[1]. All the code of Qt Quick components are written in C++ as well. I'm now creating a block editor in Qt C++ (model) and QML (view) that is 4x faster than that fastest comparable native app on macOS[2]. So, yes, it's possible to write very efficient code with QML.
For paid software at least, it would be a silly strategy.
The point of buying software on the Mac is that it is tailored for the specific environment. Someone who doesn't care about any of that wouldn't even be a potential customer, since they wouldn't even use Macs often enough to consider buying anything.
At most it would only be attracting customers who are forced to buy that software because there are no alternatives.
Literally the only reason I can’t use my work MacBook for any work that isn’t just testing Safari support. I was even ready to tolerate the terrible ram quantity given it’s so friggin’ insanely fast at javascript.
It's funny how the article entirely skips the possibility of traveling by train, which just shows how terrible the typical intercity transit option in Asia/Europe is entirely ignored due to the lack of infrastructure in the US. Amtrak is such a terrible experience most of the time with regular delays that it's almost never a good option for most travel.
For instance, the bus ticket from Boston to New York City is $30-40ish, while an Amtrak ticket is an insane $115, taking an hour longer. Why is Amtrak so expensive and slow here? IIRC the prevailing explanation is that freight/commercial trains have priority over commuter trains, which causes delays. Why is this a thing? Why is it so hard to create dedicated fast trains between major cities on the east coast, with this clear amount of demand?
This is… not true. CS170 specifically teaches about reducing NP problems to SAT (you can find this in the Algorithms textbook linked in the class syllabus). I recall solving one of the projects by using MiniSat after converting a problem to 3-SAT. FWIW, the textbook is excellent and the course was very useful.
I definitely recall doing reductions from SAT in Algorithms courses. I think that is a common part of most curricula.
I don't recall being taught any practical uses of SAT. It was introduced only in the context of Cook's theorem, as the problem you needed to reduce to other problems in order to show NP-completeness.
I think most people now learn SAT in that theoretical context, not as a tool to solve problems.
To clarify, you're specifically talking about reductions to SAT, not from SAT, right?
Note the former is used as a solution technique for feeding into SAT solvers, where the latter's goal is basically the exact opposite (to show NP-hardness and hence algorithmic intractability). Formal methods courses do the former, but algorithms courses usually use SAT for the latter.
No. Algorithms courses focus on computability and complexity (including NP-completeness); they don't generally focus on SAT solving. Formal methods are the ones that use SAT solving, SMT solving, etc. to formally prove correctness.
Ah, thank you. We had theory classes with automata and reductions and complexity proofs, and then algorithms classes that covered some solving techniques. I think I mixed up Formal Methods with Theory.
https://news.ycombinator.com/item?id=20980837