Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
ncruces
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
39 ms
·
781.
▲
by
ncruces
2y ago
Because it's much better for the consumer to search for a city, get an interactive map right there on the search results, but no link to jump to Google Maps so they can get an estimate on how long it takes to get there, because that&#x
782.
▲
by
ncruces
2y ago
#define DEBUG i<num+100 The example is silly, but you should get the point. DEBUG can be anything.
783.
▲
by
ncruces
2y ago
It can be statically compiled into the official SQLite Wasm builds: https://alexgarcia.xyz/sqlite-vec/wasm.html It can also be statically compiled into the Wasm used by my Go bindings.
784.
▲
by
ncruces
2y ago
Where would you expect to see this drama unfold if it were private? Nowhere? That'd be incredible opsec. Code review happens in public, not on GitHub but on Gerrit. The main issue tracker is public, on GitHub. Proposals are largely han
785.
▲
by
ncruces
2y ago
If I understood you, it doesn't help much, no, but neither does what you suggested. You're suggesting compiling Go to Wasm (presumably using the wasip1 target?), then converting that to C using wabt, then using Cosmopolitan to cre
786.
▲
by
ncruces
2y ago
Author here. Faster was a by product. Maintainability was the goal. API coverage, ergonomics, extensibility all rank higher in my book than performance. An example I'd like to cite is sqlite-vec. Alex was able to build a Cgo-free vers
787.
▲
by
ncruces
2y ago
https://xkcd.com/1053/
788.
▲
by
ncruces
2y ago
That too. Nor do we get to complain when searching for a place on Google shows a map, but no way to go from that map to Google Maps because that'd be a tie in.
789.
▲
by
ncruces
2y ago
An implementation in Go, that's (hopefully) simple enough to be understandable, yet minimally practical: https://github.com/ncruces/sort/blob/main/quick/quick.go
790.
▲
by
ncruces
2y ago
The dot-com bubble? Which ballooned in the late 90s, then burst, with investment not returning to 1999 levels for about a decade (also thanks to the 2008 financial crisis). Their point is that we've been stuck in 1999 for 15 years, and
791.
▲
by
ncruces
2y ago
What I'd find really interesting in this space, would be coalescing around a "standard" storage format and API. This definitely fills a need [1], and I'd easily port something like this to my Go bindings as needed (no Ru
792.
▲
by
ncruces
2y ago
Whoosh… that's the entire point you missed right there.
793.
▲
by
ncruces
2y ago
Related (for the hash based bisecting): https://research.swtch.com/bisect
794.
▲
by
ncruces
2y ago
Generic instance methods are not going to happen. The only point of methods in Go is to implement interfaces, and that doesn't really work out. This has been known since the generics proposal.
795.
▲
by
ncruces
2y ago
Honestly, I don't care about feature upgrades in a device like this, which, invariably, degrade experience, rather than improve it. I would much rather the industry moved to focus on security upgrades. Keep the device reasonably update
796.
▲
by
ncruces
2y ago
Yes: https://labs.leaningtech.com/cheerpx#cheerpx-versus-v86jslin... Also, a little below: CheerpX currently focuses on user mode (ring 3), and does not fully emulate the kernel (ring 0). We do, however, implement a subset
797.
▲
by
ncruces
2y ago
If this really was your problem you'd lower your speed for a few seconds, gain a safe 500 meters between you and the truck, and finally arrive at your destination about a minute later than you would if you were just in front of it.
798.
▲
by
ncruces
2y ago
So the entire world should rather be dependent on SpaceX, and be done with it? Arianespace had the lead and lost it, because it kept the old way of doing things. Or are you arguing Ariane 5 was technically behind and wasn't competitive
799.
▲
by
ncruces
2y ago
Yeah, this. Geofencing countries would work for highway speeds (except for Germany, but that's their problem to fix). But it won't do much to reduce pedestrian fatalities, unless you allow cities to (roughly) geofence lower limits
800.
▲
by
ncruces
2y ago
What selfish purpose is there to download GBs of images from Stripe docs? Even if we're purely selfish, wasting resources just to stick it to others is not really a productive use of our time.
801.
▲
by
ncruces
2y ago
The testing/quick package is frozen and is not accepting new features. Also: https://news.ycombinator.com/item?id=30385195 But, it occurs to me I could try and build on the same API, and compare them for effectivenes
802.
▲
by
ncruces
2y ago
I take it you haven't heard about: https://charm.sh/
803.
▲
by
ncruces
2y ago
Interesting read. I guess I'm mostly interested in on number 2 here: Under this point of view, a property-based testing library is really two parts: 1. A fuzzer. 2. A library of tools for making it easy to construct prope
804.
▲
by
ncruces
2y ago
I agree, it can be. My example above matches yours (fuzzing a data structure). I coded some tests that ensure the data structure is useful, many of them test examples from papers describing the data structure, but that don't necessaril
805.
▲
by
ncruces
2y ago
The Go fuzzer, when it finds a failure, will also walk back and try to shrink inputs (and still trigger the same failure). Not sure how effective it is compared to other options, but I'm not totally missing out there. It also builds a
806.
▲
by
ncruces
2y ago
With the advent of coverage based fuzzing, and how well supported it is in Go, what am I missing from not using one of the property based testing libraries? https://www.tedinski.com/2018/12/11/fuzzing-and-prop
807.
▲
by
ncruces
2y ago
If you go with that, just move it in the correct direction every year, regardless if you make the absolute delta bigger (you'll fix it in a year by going backwards). If this is not enough to ping-pong around the correct result, because
808.
▲
by
ncruces
2y ago
I actually have no beef with either “money will change hands in the future” or “AGPL because we want to advance FOSS.” The problem is that AGPL is a total no starter at “day job,” and even at “side gig,” I'm obviously not going to rele
809.
▲
by
ncruces
2y ago
I actually agree. The AGPL, for something like this, is a total non-starter, IMO, and just changes the conversation. It'd be one thing to choose the AGPL because you believe in the FOSS movement. Or if you're releasing something e
810.
▲
by
ncruces
2y ago
Go makes heavy use interior pointers, which WasmGC doesn't support.
More ›