Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
torben-friis
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
7 ms
·
61.
▲
by
torben-friis
4mo ago
>I think that you'd have a hard time convincing regulators that this cause really matters. "A foreign power could potentially deny access to the OS" sounds like a compelling argument.
62.
▲
by
torben-friis
4mo ago
Sure, there's a place for disposable, ultra custom apps that AI can cover. I vibecoded one to help me clean up my local media with a tinder-like UI for discarding/keeping photos, for example. That's not going to substitute ma
63.
▲
by
torben-friis
4mo ago
>We already collectively produce more software than anyone really wants, and the bulk of it is anywhere between garbage and outright fraud, if not actively malicious. I think you're right in the second part of the quote, but the fir
64.
▲
by
torben-friis
4mo ago
Thanks for the suggestion! For the moment I'm dual booting, since the m1 MacBook is still a decent piece of hardware for my needs and asahi Linux offers good support. Given the current market I intend to hold on as long as possible, bu
65.
▲
by
torben-friis
4mo ago
Generation and maintenance are very different beasts. There are engineers that work making a 2 weeks app that is never revisited again, I guess, but I don't know anyone that makes a living from that. Maybe the "WordPress site for
66.
▲
by
torben-friis
4mo ago
It was death by a thousand cuts. I am not a designer and Tahoe bothered me enough that, after trying it at work, I've been actively stopping my personal Mac to update, and it was a factor in moving to android.
67.
▲
by
torben-friis
4mo ago
The decimals of 1/3 are infinite as well and they don't contain a better-than-wikipedia article. And even if they did, it would be useless if it's buried in useless data and your chances or pulling it are effectively zero. Th
68.
▲
by
torben-friis
4mo ago
>The syllabary was widely lauded, as its phonetic accuracy and simplicity made it far easier to grasp than English. I mean, that feels like it's bound to happen when an alphabet is built to represent current language or pronunciatio
69.
▲
by
torben-friis
4mo ago
They have a silent nerfing system for their models and say so openly. The obvious question is how much it is being used already. Competitor companies being nerfed? Non Americans getting worse code? Punishing and rewarding users to maximize
70.
▲
by
torben-friis
4mo ago
OPPO just added a great feature into my phone. You point your camera at a foreign restaurant menu and it generates a translated version including a picture so you can get a general idea of what it usually looks like. The genius part is that
71.
▲
by
torben-friis
4mo ago
LLMs will probably get to a point where anyone who provides a well explained, fully detailed account of what they want can get it. My job is safe.
72.
▲
by
torben-friis
4mo ago
It's still nice to see a company not double down on a fall! They seem to have been on a full year of tech debt and optimisation. I still would have liked a more genuine walk back (they sold it as "iterations and adjustments"
73.
▲
by
torben-friis
4mo ago
If you're on Android, you can use revanced to patch social network apps, to, among other things, remove content from non-friends (and ads). It's scary how empty the feed is once you do this. It can be full days with the same post
74.
▲
by
torben-friis
4mo ago
Anecdotally, I've only seen them done in northern European companies, but every northern European company I've interviewed for had them. It seems to be a regional-ish thing.
75.
▲
by
torben-friis
4mo ago
>If you boil it down to it, you're quite literally saying the problem is the LLMs dont have access to a bunch of facts. Well yeah, And our problem with mortality is not having access to a bunch of medical facts :) Kidding aside, it&
76.
▲
by
torben-friis
4mo ago
>This is domain expertise - software engineers are not needed for that. Ofc often senior sws are expert in it, but they aren't necessary. Our engineers frequently need to be on the loop with product and stakeholders: Due to real wor
77.
▲
by
torben-friis
4mo ago
>Eg last I checked Anthropics one-shot performance on Clojure was about the same as Python or Go despite almost certainly being less represented in training data. The combination of density and simple primitives might be easier for an LL
78.
▲
by
torben-friis
4mo ago
>you take a spec and create tests, every little thing A spec detailed enough and unambiguous enough to be translated into machine execution deterministically is called code. Unlike a compiler, AI can build with a spec that is not detaile
79.
▲
by
torben-friis
4mo ago
My career path is suprisingly similar to the author's. Weirdly enough, what he takes as the first pillar to fall is the one I see most undamaged currently. LLMs routinely fail at our business specifics: Local tax regulations, particula
80.
▲
by
torben-friis
4mo ago
Here's a fun one: firefox lists its current count at about 2.5M LOC, from roughly 1M commits during the years. You end up with about 3 lines added per commit, which is not ridiculous when you consider that most would be editions rather
81.
▲
by
torben-friis
4mo ago
>We're still at the beginning of figuring out what these tools even are, let alone how best to use them. That's a fair point, and I imagine it's going to take a long while for the dust to clear. People are operating on sem
82.
▲
by
torben-friis
4mo ago
Also, "AI good" and "AI bad" are very silly camps to describe a topic. You might have seen some comments by me receiving votes lately that some might classify as anti-AI. I'm not for banning it, I use it at work and
83.
▲
by
torben-friis
4mo ago
In seeing very conflicting reports here. No trust that the job numbers aren't cooked? Random panic? Not American so I'm out of the loop.
84.
▲
by
torben-friis
4mo ago
They're those guys that spend hundreds of hours in pickup artist courses rather than being a person people would like to be with, yeah.
85.
▲
by
torben-friis
4mo ago
Funnily enough, I live for the page. I would love to do my whole job with pen and paper and ignore the computer even.
86.
▲
by
torben-friis
4mo ago
>and I can imagine an AI powered person writing 25,000 lines of code per week. A year of that is 1.3M code: the size of systemd, or postgres. Can you imagine a single person writing systemd (not a POC, the current version feature complet
87.
▲
by
torben-friis
4mo ago
I'm not say it's pure hype. Not in the sense of previous hype waves like the metaverse or NFCs. It clearly has uses. I do think it is hype as a killer of knowledge work. It can certainly remove a lot of friction in the kind of bor
88.
▲
by
torben-friis
4mo ago
>But all now seems like a transition phase. Transition to f-ing what though? It feels like being in the middle of a tornado. But I think it helps to turn off screens, sit in a desk, and calmly remember first principles and consider them
89.
▲
by
torben-friis
4mo ago
There is certainly a certain... entitlement? (It's not the perfect word, but I fail to find a proper term) from some of the vibe crowd. Like an attachment to the output and refusal to accept that most of the work was not theirs. It is
90.
▲
by
torben-friis
4mo ago
I mean, you're not creating the api from thin air as you write code right? Usually you'd have some larger doc about the project with the design you can point to.
More ›