Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
codeflo
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
9 ms
·
121.
▲
by
codeflo
2y ago
That is enough to distribute malicious code though, at least in certain scenarios. Someone might create a setup where reviewers check/sign one version of the source code, and what gets distributed is another version with the same hash.
122.
▲
by
codeflo
2y ago
That's kind of a trivial observation -- if chess AI were predictable to grand masters, then those GMs could play like the AI and thus wouldn't lose to it.
123.
▲
by
codeflo
2y ago
From a nostalgia viewpoint, adding labels feels a bit "impure". You might as well want to introduce other syntactic sugar, but at some point, you're no longer writing C64 BASIC. But what if you had an IDE with convenient/
124.
▲
by
codeflo
2y ago
It seems that the Rust macro system is inspired by a similar idea: In the first step (the "reader" in this article's terminology), the source is converted into something called a token tree . A token tree is not a full parse
125.
▲
by
codeflo
2y ago
This article invents a new binary operation, calls it "division" and uses the "/" operator to denote it. But the article repeats multiple times that this new operation isn't a multiplicative inverse, so it'
126.
▲
by
codeflo
2y ago
> everyone is wrong! Well, not everyone. I wasn't the only one to mention this, so I'm surprised it didn't show up in the list of theories, but here's e.g. me, seven days ago (source https://news.ycombinato
127.
▲
by
codeflo
2y ago
Pushing a half-abandoned but widely beloved project into the visibility of the bean counters at Google with a birthday announcement like that is a dangerous game. Best of luck.
128.
▲
by
codeflo
2y ago
I'll reserve judgement on its net effect until the moment they kill it.
129.
▲
by
codeflo
2y ago
At this point, we have to assume anything that becomes a published benchmark is specifically targeted during training. That's not something specific to LLMs or OpenAI. Compiler companies have done the same thing for decades, specifical
130.
▲
by
codeflo
2y ago
Euler’s graph theorem.
131.
▲
by
codeflo
2y ago
It’s supposedly a “fallacy” and “arbitrary” and yet, 26 years after this prediction, the growing languages right now are Rust (compiled, statically typed systems language) and Python (dynamically typed glue code written by non-CS mathematic
132.
▲
by
codeflo
2y ago
GP is correct, that’s the definition of “lossy”. We don’t need to invent ever new marketing buzzwords for well-established technical concepts.
133.
▲
by
codeflo
2y ago
How can a reusable function have a fixed return address?
134.
▲
by
codeflo
2y ago
On x86, ret and call are explicit instructions. Ret always predicts the address of the last call, which is (usually) 100% accurate. Your example of `if (condition) jump $panic_recover_address` contains two branches, either of which can be m
135.
▲
by
codeflo
2y ago
> Returning to an alternate address shouldn’t be significantly more expensive than returning to the default address, so this has to be cheap. Modern CPUs add complications to arguments like this. Branches stall the execution pipeline, so
136.
▲
by
codeflo
2y ago
Okay, but narrowly focusing on a "quantization-robust unlearning strategy" as per the abstract might be a red herring, if that strategy doesn't incidentally also address other ways to undo the unlearning.
137.
▲
by
codeflo
2y ago
I think quantization is a red herring. If there's any way to undo the unlearning, this means that the knowledge is still in the weights -- that's basic information theory. I'm sure there are a million other ways to recover
138.
▲
by
codeflo
2y ago
It's not a bad question, these units of measurements are always a bit confusing. You can similarly ask why for humans, rubbing a balloon is harmless, although that builds up 30 kV of static electricity, while touching a 230 V power soc
139.
▲
by
codeflo
2y ago
I'm not informed enough to rebut this, and don't want to be quoted in the follow-up article that suggests HN is still too dumb to get the genius of Ballmer, but here's my take. It's only the footnote of the article that
140.
▲
by
codeflo
2y ago
Showing a picture of Notre Dame photoshopped against unsettling clouds to make a point about the psychological effect of its architecture is borderline fraud. Any actual photo makes the building look a lot more majestic rather than scary:
141.
▲
by
codeflo
2y ago
You’d probably be surprised how often people have to create config files programmatically. With properly specified formats, you can serialize the configuration, which means nesting and string escapes are taken care of automatically. With
142.
▲
by
codeflo
2y ago
And that’s assuming the repository never gets reorganized in any way that doesn’t perfectly preserve history — which over long timespans, is bound to happen at some point.
143.
▲
by
codeflo
2y ago
Can someone clarify whether this is intended as a joke or whether the author is actually confused? I mean, nothing about this makes sense: it's not "scripting"; it claims to introduce "strong typing" while it does n
144.
▲
by
codeflo
2y ago
Unordered simply means the order isn’t exposed, not that the bits have somehow literally fallen out of the computer and are scattered on the floor.
145.
▲
by
codeflo
2y ago
Not the least of which is that people who take stuff like that as gospel instead of as heuristics tend to endlessly argue what is and isn't a "unit".
146.
▲
by
codeflo
2y ago
Someone on Twitter claimed that Svelte 5's runes are basically a nicer (custom) syntax for Solid.js -- is that true? I find the way Solid.js decouples signals from components to be a huge step up from React's state management in t
147.
▲
by
codeflo
2y ago
But the headline is “GM-level performance without search”, not “computer beats human at mouse movement speed contest”.
148.
▲
by
codeflo
2y ago
Exactly, but distinguishing the two that requires an excellent understanding of the problem space, and can’t at all be figured out in the solution space (i.e., by only looking at the code). But less experienced people only look at the code.
149.
▲
by
codeflo
2y ago
There’s also the effect that a certain code structure that’s clearer for a senior dev might be less clear for a junior dev and vice versa.
150.
▲
by
codeflo
2y ago
I'd like a theory to explain why a certain class of jokes makes me feel physical pain, when others find them hilarious. This example from the article, > Q: Why are cats so good at video games? A: They have nine lives. firmly belongs
More ›