Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
ATsch
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
58 ms
·
361.
▲
by
ATsch
6y ago
That will depend a lot on one thing: why do you want push?
362.
▲
by
ATsch
6y ago
Also substantially more expensive, especially compared to the likes of the banana pi zero, which gets you four cores at 1.2Ghz, 512MB RAM, Linux and a GPU for 35% cheaper. Pair that with a Blue Pill STM32 for $3 if you need realtime and you
363.
▲
by
ATsch
6y ago
Maybe, but what scenarios are there where you'd want that, be willing to spend 30 dollars on your microcontroller solution and not benefit from a more capable linux SOC? (With an optional $3 M0/M3 for realtime if required)
364.
▲
by
ATsch
6y ago
This is more or less what DLSS is
365.
▲
by
ATsch
6y ago
I think you'll struggle to find any place without a few datacenters nearby, at least in a neighboring country. Of course it's going to be a bit further away if you're in central Africa. Even then, you can colocate in a datace
366.
▲
by
ATsch
6y ago
You wouldn't run your own datacenter, you would colocate. Even if that's not possible, you can still rent servers for significantly cheaper than cloud offerings.
367.
▲
by
ATsch
6y ago
The best part is that every time your geolocation changes, all google products will reset your language.
368.
▲
by
ATsch
6y ago
I like that things might break two times a year, when I can prepare and set aside time for it, instead of randomly. I also like not having to check a website for any issues like one does with arch.
369.
▲
by
ATsch
6y ago
Not to start a flamewar, but I'd like to point out why I think snap is not only inferior to flatpak technically but actually a threat to the linux desktop: Snap is very deliberately centralized, with a single hard-coded repo URL. The s
370.
▲
by
ATsch
6y ago
The bigger porting problem is bytes vs unicode. That can't be done automatically and is a lot of work.
371.
▲
by
ATsch
6y ago
This isn't a worry over device security. It's a worry about cloning. Most of the serious FPGA market is low-volume speciality devices that cost upwards 10-100k per unit, with yearly support contracts of that magnitude on top. Seis
372.
▲
by
ATsch
6y ago
Vendor lock-in is the primary way in which these companies make money
373.
▲
by
ATsch
6y ago
This is another example of what Moxie Marlinspike calls the "cryptographic doom principle". If you do anything, anything with a ciphertext before checking authenticity, doom is inevitable.
374.
▲
by
ATsch
7y ago
Still incredibly disappointed this didn't catch on widely before NAT prevented any new protocols. There'd be so many problems we wouldn't have today.
375.
▲
by
ATsch
7y ago
That can be solved, but more importantly, Ethernet can not scale beyond 10k or so nodes because the addresses are meaningless and there's no way to tell what the next hop for an address you've never seen before should be beyond ju
376.
▲
by
ATsch
7y ago
Well, your ByteBuf is still a pointer. You also now need to dereference it to get the length. It also can't be passed by value, since it's very big. You can also not have multiple ByteBufs pointing at subsections of the same regio
377.
▲
by
ATsch
7y ago
I think the idea of mobile IP is fundamentally flawed and I'm not surprised it didn't catch on. The whole point of IP addresses is to be a hierarchical address describing a location in the network. Once you can move IPs, they beco
378.
▲
by
ATsch
7y ago
Yes, this shouldn't be done at the application layer. This is why it's done in QUIC, which is layer 4.5-ish. HTTP/3 runs on top of that. The "flat address space" idea however is completely ridiculous. That would mea
379.
▲
by
ATsch
7y ago
I'm not sure if it matters. It might be better for some technical reason, such as speeding up double dereferences, because you don't need to add anything to get to the pointer. But to be honest I just copied it out of existing cod
380.
▲
by
ATsch
7y ago
typedef struct {uint8_t *data; size_t len;} ByteBuf; is the first line of code I write in a C project.
381.
▲
by
ATsch
7y ago
The connection does drop though. Range requests just allow you to resume it.
382.
▲
by
ATsch
7y ago
Multipath is different in purpose though, although it does also break up the tuple. It's designed to allow load balancing across multiple addresses/interfaces. You need to explicitly tell the other side "I am also available u
383.
▲
by
ATsch
7y ago
There's something else, performance aside, that's really exciting about HTTP/3: Fixing a decades old layering violation that has made truly mobile internet impossible. In TCP, a connection is uniquely identified by the follow
384.
▲
by
ATsch
7y ago
> There isn't an official LOS Treble build (for various reasons) could someone share what those reasons are?
385.
▲
by
ATsch
7y ago
This is basically what the "steam machine" model was. Standardized controller, hardware specifications, OS, available from multiple vendors. It was unfortunately a bit of a flop.
386.
▲
by
ATsch
7y ago
Even when there aren't, it is very frequent that the cameras happen to be "broken" when something bad happens.
387.
▲
by
ATsch
7y ago
I brought up pollution to point out that, although certainly worth it, society is paying a pricy to let people fly aeroplanes around as they wish. Since people owning planes are not just regular citizens, society should hence have some leve
388.
▲
by
ATsch
7y ago
Do you own an personal aeroplane with ads-b transmitter? Do you think it's likely you're going to own one anytime soon? This isn't the private data of your neighbor. This is either data from corporations which, despite being
389.
▲
by
ATsch
7y ago
Private Crop dusing pilots aren't trying to hide themselves from the airplane maps, so this is irrelevant to the point.
390.
▲
by
ATsch
7y ago
I hope not. Unlike car license plates, if you own a plane what you're doing with it is very much in the publics interest to know, especially if you don't want people to know. They can have their privacy while they aren't fl
More ›