Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
haimez
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
24 ms
·
211.
▲
by
haimez
9y ago
So, every time, he took a pic in the exec's chair, stole a file from the server room, and also did whatever he wanted? What is the 'N' factor here, because it sounds like your friend is a bold high schooler who achieved N=1,2
212.
▲
by
haimez
9y ago
The part where the execution graph is "linked" at runtime into an actor graph. It's hard to reason about because it's flexible enough to enable large numbers of possible combinations. I'm not saying it's bad, b
213.
▲
by
haimez
9y ago
"Erlang style" supervision error handling is short-hand for opt-in, optional error handling without any compiler support. That's a pretty serious trade-off to not address directly.
214.
▲
by
haimez
9y ago
To your second point, your lisp analogy requires a macro or very sophisticated JIT. These are not viable in swift's primary runtime, not even a little bit. It's a non-starter.
215.
▲
by
haimez
9y ago
Forgive my naivety, but parent article seems to contradict the github post linked in your reply in quite a few ways. What is the time / design iteration relationship between the two, and why (not to put you on the spot)? These are VERY
216.
▲
by
haimez
9y ago
Feels like the elephant in the room for the description is not once mentioning Futures but instead jumping straight to using async / await keywords and the actor model. As evidenced by C#, you can't avoid leaking the type signatur
217.
▲
by
haimez
9y ago
Kind of ironic, given that the whole article is about the divergence of system load from being a queue length metric.
218.
▲
by
haimez
9y ago
I think the problem with "this" is that you are trying to recreate classes instead of passing it as a parameter, or using currying. Arrow functions might be slightly more intuitive to intention, but the problem is still there. Th
219.
▲
by
haimez
9y ago
In what world is 2.8% sufficiently small to be ignored?
220.
▲
by
haimez
9y ago
Please adjust your equations for time.
221.
▲
by
haimez
9y ago
Install the app! It solves these usability problems. No? ... waits 5 seconds... HEY INSTALL THE APP. HEY, HAVE YOU HEARD OF THE APP?
222.
▲
by
haimez
9y ago
And voila, rust.
223.
▲
by
haimez
9y ago
Lol. Parallel data computations across resource (including battery) constrained devices? Good news, the owner of the device is now the product. The device is also the product. Can't wait.
224.
▲
by
haimez
9y ago
Let me summarize: academic language, it's easy for the GC that way, and "if you though that Float boxing would kill performance you wouldn't have started down this path"
225.
▲
by
haimez
9y ago
Devils advocate: they're not doing anything better than anyone else in the market and they don't make a product. They're very invested in their ecosystem which is only as good as they make it because they have a very quiet, w
226.
▲
by
haimez
9y ago
No love for Oracle here (except for all the money they spend having talented engineers work on the JVM. ie fuck their databases). It's still laughable to call IBM and RedHat a community. They're consultants. Government consultants
227.
▲
by
haimez
9y ago
Yes.
228.
▲
by
haimez
10y ago
They're also famous as athletes and not creatives.
229.
▲
by
haimez
10y ago
Every week, some poor soul will sacrifice the entirety of their karma to the horde of unwashed masses. Perhaps these masses will know JavaScript particularly well. Maybe Java. It doesn't matter, we will have no peace.
230.
▲
by
haimez
10y ago
Sounds like you need a managed runtime. Wait... You already have one, it's just a bigger problem than solution. Hrm... Have you considered making a real IDE?
231.
▲
by
haimez
10y ago
No one cares about the difference between 99.9 and 99.999 reliability at the DB layer and then adopts a new open source DB to solve that problem. Especially when that exciting, new, experimental database cripples your ability to back it up.
232.
▲
by
haimez
10y ago
Sounds like crippleware for a distributed database. If you can't incrementally back it up, you can't really afford to run it in production in a cluster that has a large dataset. If you don't have a large dataset, you don'
233.
▲
by
haimez
10y ago
Right... which means that "select" is like "class". It means some shit, but it isn't supposed to mean "KERNEL ASYCIO INTERFACE" the same way that "class" doesn't mean struct.
234.
▲
by
haimez
10y ago
Forgive me if I'm wrong, but I believe this is because the select keyword is a language syntax convenience whereas actual I/O needs a full go routine because that's the unit of parallelism.
235.
▲
by
haimez
10y ago
It's a boring tautology. Arcade games require "clever" (read: not generalizable) solutions to optimize for speed in decision making. "Learning" AI requires the broadest feasible search space for any solution and suc
236.
▲
by
haimez
10y ago
Maybe you should find any of the websites that profit from faster load times and argue the business value. Because that's most of the places.
237.
▲
by
haimez
10y ago
Wet blanket! The rust will sustain them!
238.
▲
by
haimez
10y ago
I'm also not an expert, but I think the bigger cost here is memory latency. Size correlates to, but scales differently than the switching cost of registers because it's zero sum. A register saved is a register waited on, twice. Th
239.
▲
by
haimez
10y ago
No, just means that someone needs to bother to make an AI for Ms. Pacman. Handling random ghost movements is only trivially confounding for AI learning. Low value, low yield.
240.
▲
by
haimez
10y ago
Lol, no. It's just the scale of the use that makes it valuable. Google datacenter cooling problems only exist inside of Google data centers, and the control knobs therein are 1. measurable AND measured 2. deterministic 3. simple. They
More ›