Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
wwright
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
17 ms
·
121.
▲
by
wwright
6y ago
We have a similar setup, but I've been generating UUIDs for each event at logging time to avoid any computational overhead for deduping.
122.
▲
by
wwright
6y ago
I think 11 to 200, but it’s not surjective
123.
▲
by
wwright
6y ago
At work we have a program that uses the exit code to encode three enums. Each digit of the decimal representation of the exit code corresponds to one of these enums. It’s some of the most cursed code I’ve ever worked with. (God bless anyone
124.
▲
by
wwright
6y ago
I disagree. It’s Boolean and flat (there is neither sequential nor nested logic around the state). It does recurse for each sub thread, but only in the case where there is no state. There’s also no business logic built around it, only displ
125.
▲
by
wwright
6y ago
I would define it as a file that is distributed for use among many decoupled hosts and which is generated from some sort of source input. Packages/libraries such as those used by NPM and Pip are some examples. It may also be an entire
126.
▲
by
wwright
6y ago
This site also has almost no state at all per-user.
127.
▲
by
wwright
6y ago
Even if it was, that’s definitely not significant. Each engineer is probably paid more per day than that. (I’m not a fan of the flat 30% charge, but developer program fees are not a challenge for any group with a revenue stream if any kin
128.
▲
by
wwright
6y ago
AFAIK this is false, though I could be wrong. I believe cancelling the account is separate from revoking the apps. I think Apple has only deleted installed apps off of user’s phones in rare, exceptional cases, such as actively malicious app
129.
▲
by
wwright
6y ago
To reiterate the point of my comment above: how big is the difference when using Svelte, really? Is it simpler nicer , or does it let you model the problem more closely? I don’t have enough experience with it to say (though I do like to se
130.
▲
by
wwright
6y ago
I think there is value in looking for boring solutions, but I think that's really a secondary type of problem ("how do we do this well?") versus the primary type of problem ("what are we trying to do?"). There are s
131.
▲
by
wwright
6y ago
The format could still be "hash of everything else + everything else"
132.
▲
by
wwright
6y ago
I think it’s worth noting that no engineer at Mozilla owes anything to us as the public, especially not this week.
133.
▲
by
wwright
6y ago
This problem certainly exists without containers. - How do you provide easy visibility into running applications? - How do you prevent inspecting an application from affecting the behavior of an application?
134.
▲
by
wwright
6y ago
I believe that if you do "Get Info" on an individual application, you can disable Retina rendering for it. The OS will render it with the full non-retina code path into a non-Retina-sized buffer and then upscale it to fit on your
135.
▲
by
wwright
6y ago
Never spend an afternoon programming something that you may have to waste future time debugging and supporting when a dependency already handles it perfectly.
136.
▲
by
wwright
6y ago
I disagree for the exact reason the thread here started: it’s part of your public API. If you don’t want keyword arguments in your public API contract, you absolutely should omit them from the contract.
137.
▲
by
wwright
6y ago
I can’t deny that, but I can certainly point out that billionaires coinciding with the rise in quality does not mean that they were even produced by the same system, or that the presence of the billionaires was in any way necessary. To refr
138.
▲
by
wwright
6y ago
Did billionaires accelerate that, or did they just profit from it disproportionately while slowing it down?
139.
▲
by
wwright
6y ago
Modern Python lets you explicitly opt out of this by putting “/“ in your argument list iirc correctly: def only_positional(a, b, /): Pass Similarly, you can prevent keyword args from being called as positional (wh
140.
▲
by
wwright
6y ago
I don’t mean to start an off-topic debate on this, but I just have to say that personally, I run into issues caused by using old software on our EL7 boxes all the time . For example, cgroup memory limits are broken on Red Hat 7 kernels as
141.
▲
by
wwright
6y ago
requirements.txt has a lot of deficiencies for package management (compare to the various features Yarn or Cargo offer). The biggest weakness is transitive dependency locking: even if you specify the exact version of the dependencies you wa
142.
▲
by
wwright
6y ago
They also try to have aggressive defaults on integer wrapping and so on. I work with a junior who is writing a Go program who would have benefited from the early detection ;)
143.
▲
by
wwright
6y ago
The “anyhow” and “thiserror” crates remove a lot of boilerplate, too. I hope something like that gets merged into std.
144.
▲
by
wwright
6y ago
This argument is predicated on value creation being the primary determining factor in salary, and I think that’s probably worth a second thought
145.
▲
by
wwright
6y ago
Sharing your salary is absolutely not a “revolt” in any sense; this is literally a national worker’s right as an American. Bloomberg framing it as a “revolt” is a not very subtle attempt at casting workers using their rights as the agents o
146.
▲
by
wwright
6y ago
In aggregate yes, but many individuals still have perfectly valid reasons for not wanting to share their salaries locally. No one should _have_ to share that info in most roles (important leadership and financial roles notwithstanding), but
147.
▲
by
wwright
6y ago
Not giving employees benefits does not necessarily lead to a better service
148.
▲
by
wwright
6y ago
He says at the start that he just kept some notes for the last month and was publishing them all
149.
▲
by
wwright
6y ago
IMO, abstractions aren’t easy, but programming with the wrong abstraction is significantly worse than with the right abstraction. I think that because everyone starts with few visible abstractions, they get the sense that adding abstraction
150.
▲
by
wwright
6y ago
This doesn’t seem true to me. You can definitely work with the traditional kernel and C model, but we have eBPF, GPU computing, unikernels, C++20 and Rust… and I’m not a low-level guy, that’s just off the top of my head. Things do evolve.
More ›