Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
zbentley
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
15 ms
·
211.
▲
by
zbentley
3mo ago
There’s a big difference between executing something from your package manager—and bluez is packaged by literally every major distribution—and executing: a random binary from a brand new GitHub repo (which you didn’t build and can be swap
212.
▲
by
zbentley
3mo ago
A community script is fine. A community script which redistributes opaque binaries that are actually renamed builds of an unrelated project with an incompatible license via curl | bash is not. At best it indicates a lack of awareness of
213.
▲
by
zbentley
3mo ago
Don’t download this. It’s a curl | bash installer for a binary whose sources appear to just … be the popular FOSS bluez btproxy tool, which it downloads and builds, and whose binaries are checked into the repo. But those binaries could just
214.
▲
by
zbentley
3mo ago
I'd argue that the latter's a degenerate case as well--for the somewhat nitpicky/minor reason that package managers' errors are extremely poor at indicating the difference between "the version you asked for doesn&
215.
▲
by
zbentley
3mo ago
Genuine question: why not? There's pretty consistent reporting that rental building owners are earning considerable returns from rent. Rent freezes are one (of several) ways to encourage building owners to reinvest those returns to b
216.
▲
Magic codec – preprocessing for the Python interpreter (2024)
(github.com)
1 points
by
zbentley
3mo ago
|
0 comments
217.
▲
by
zbentley
3mo ago
This code seems to be fairly pervasively AI-written (both structurally/syntacticly, with many additional AI verbosity tells in the documentation, and going by the commit history). That's a problem for me, given what this library d
218.
▲
by
zbentley
3mo ago
Because (even if you could decorate for-loops, or instead decorated functions containing a loop and nothing else) the resulting behavior would be completely unlike the behavior of all other @decorators in Python. It would also be nearly imp
219.
▲
by
zbentley
3mo ago
I mean, you made a good metaphorical case for controlled handlebar slop/headset resistance in bicycles/motorcycles. Which are things that exist to mitigate crashes from oversteering on both of those vehicles!
220.
▲
by
zbentley
3mo ago
I suspect this is in bad faith, but assuming not: how does that follow? “Large numbers of people will uncritically follow sketchy instructions and get hacked” doesn’t in any way lead to “and therefore they cannot be trusted with devices”. “
221.
▲
by
zbentley
3mo ago
Eh, this is a tough conversation for engineering-minded folks because the right answer is probably somewhere in the middle of a few different variables. Too far towards trying to make mistakes impossible (which is easy for corporations to t
222.
▲
by
zbentley
3mo ago
> You irrigate those with water that's fit to drink To be able to irrigate even a fraction of the US’s crops with potable water, we’d have to build some massive infrastructure to purify, store, and deliver that water. Existing nonpo
223.
▲
by
zbentley
3mo ago
> 5 orders of magnitude too small for another. Nitpick on an otherwise good post, but I don’t think there are very many 6billion RPS systems out there, and those that do exist are almost certainly using bespoke, purpose-built tools
224.
▲
by
zbentley
3mo ago
I mix up "throw down the gauntlet" and "throw in the towel" at least monthly, don't worry.
225.
▲
by
zbentley
3mo ago
That's often true, but less true than it was a decade ago. Private, commercial space companies are a lot more numerous now. Sure, most of them have military contracts/oversight as well, but there are more opportunities with them t
226.
▲
by
zbentley
3mo ago
Somewhat, but rocket engineering has some upward cost pressures that offset the savings from being in India: expertise is expensive even adjusting for the cost of labor (many staff are likely competing on salary with the rest of the world--
227.
▲
by
zbentley
3mo ago
And with $160MM in funding at a $1.1B valuation? I don't know about their debt, but to get to LEO in 8 years on that little money is extremely impressive.
228.
▲
by
zbentley
3mo ago
I think that they meant “throw in the towel”? Not positive though.
229.
▲
by
zbentley
3mo ago
I'm sure many/most don't know that specifically. But as a sibling poster pointed out, people do have a strong sense for "bullshit"--defined here as fairly crass and/or greedy, low-effort marketing or design.
230.
▲
by
zbentley
3mo ago
That is indeed baffling and regrettable. Perhaps they believe that users will just hard-cast what .parse() returns to JsonObject/JsonArray/whatever, and that the resulting ClassCastException will be uglier and harder to debug than
231.
▲
by
zbentley
3mo ago
Eh, I think it's a good tradeoff. If it losslessly supported deserializing comments/JSONC, or trailing commas, or JSON-lines, or whatever, the serialization APIs would get more complicated. Every time you serialize you'd
232.
▲
by
zbentley
3mo ago
> You insinuate that codeberg's opinion is due to hosting Zig. I didn’t interpret GP that way. Rather, they were saying that like-minded projects are appropriately homed.
233.
▲
by
zbentley
3mo ago
That’s pretty silly whataboutism. The two aren’t mutually exclusive, don’t trade off, have both risks and benefits, and can be used or prescribed responsibly or irresponsibly. What are you getting at?
234.
▲
by
zbentley
3mo ago
Firmly agree. A lot of the lessons you mentioned have been discussed previously on HN threads for hardware startups. This is one of my favorites: https://www.simonberens.com/p/lessons-learned-shipping-500-u... HN threa
235.
▲
by
zbentley
3mo ago
I guess, but like ... long context/context rot/contextual memory retrieval are still very unsolved problems. If an AI's getting input from multiple family members for years, I have a lot of trouble believing that current LL
236.
▲
by
zbentley
3mo ago
Row ID preloading is an extremely effective technique—and not just for SQLite. I’ve also used it to great effect on massive Aurora MySQL or Postgres clusters since I could send the SELECT to a replica, and the whole point of deletions was t
237.
▲
by
zbentley
3mo ago
> The White House electrician and kennel manager, Traphes Bryant, trained Pushinka with peanuts to climb up a ladder to Caroline Kennedy's playhouse and slide down the other side. "White House electrician and kennel manager&quo
238.
▲
by
zbentley
3mo ago
Doing generics in a new Rust project’s types by “dyn”-ing everything you can and then falling back to monomorphized “true” generics after benchmarking is (or will be, when dyn compatibility for more layouts is added) a pretty okay way to go
239.
▲
by
zbentley
3mo ago
> Thanks for high quality reply. Likewise; thanks for engaging constructively as well. > I traditionally have treated this as a state in a database. Which is a super appropriate tool many (most) times! I've been using "MQ&qu
240.
▲
by
zbentley
3mo ago
> The answer is usually a performance problem for which this is a bandaid. I agree that if that's the problem you're trying to solve, MQs are a poor stop-gap and not a general-purpose solution. But many (most?) MQs aren't
More ›