lbrandy

8 ms·

1.
by lbrandy 28d ago
> I've seen many cases where engineers optimize the sad path, but pessimize the wretched path. As said GP, yes this is exactly what I meant and "happy-path" was an unclear choice of words. They optimize for the unhappy pat
2.
by lbrandy 28d ago
Some anecdotes 1. People add retries w/o backoff. Now we have a retry storm. 2. People don't add jitter so we get huge waves of highly correlated retries that cause self-perpetuating overload cycles and failure. 3. People add retr
3.
by lbrandy 28d ago
I cannot even begin to express how many times I've seen engineers working super hard to optimize happy-paths so that we turn 3 nines of availability into 4 nines but introduce unintended emergent behaviors in unhappy-paths that turn 1
4.
by lbrandy 4mo ago
NASDAQ index has a 3x float weighting (and a far, far smaller total capitalization) which makes it far more susceptible. Other indexes do not have these multipliers, and are much larger. The exposure for e.g. VTI is far, far less.
5.
by lbrandy 4mo ago
You should not trust me, but here's my understanding. I wish there was a really good writeup somewhere to explain this authoritatively but I'm not sure there is one. Would also love to see one. Frankly vanguard should do it. VTSAX
6.
by lbrandy 4mo ago
There is a huge amount of misinformation on this topic, including in this thread, at the minute. Some index funds have a very long horizon before they include them (e.g. a year). Others are "fast-tracked" (e.g. notably VTI). Most
7.
by lbrandy 4mo ago
At least part of the answer is this doesn't meet the requirement of the proposed law. You need to actually provide the functionality of the server-side, not just its API.
8.
by lbrandy 11mo ago
> Anyway, I don't think that there's anything that it's like to be an LLM. I don't see how anybody who knows how they actually work could think that. While I have almost zero belief that LLMs are conscious, I just don
9.
by lbrandy 11mo ago
I have no idea how you can assert what is necessary/sufficient for consciousness in this way. Your comment reads like you believe you understand consciousness far more than I believe anyone actually does.
10.
by lbrandy 1y ago
> My point isn't that coordination is easy - it's that treating it as impossible becomes self-fulfilling. While I see what you are getting at, and I think its super important we come up with philosophical frameworks to push bac
11.
by lbrandy 1y ago
I was struck how the argument is also isomorphic to how we talked about computers and chess. We're at the stage where we are arguing the computer isn't _really_ understanding chess, though. It's just doing huge amounts of d
12.
by lbrandy 1y ago
> has a model much as we humans do The premise that an AI needs to do Y "as we do" to be good at X because humans use Y to be good at X needs closer examination. This presumption seems to be omnipresent in these conversations
13.
by lbrandy 1y ago
Suppose this is as good a place to pile-on as any. Though this was not the post I was expecting to show up today, it was super awesome for me to get to have played my tiny part in this big journey. Thanks for everything @je (and qi + david
14.
by lbrandy 1y ago
It's fun to see everyone arguing about what "everyone" thought.. when... we can just... look... https://news.ycombinator.com/item?id=3817840 is a fun thread from 2012. The top reply to the top comment has som
15.
by lbrandy 2y ago
For uninitialized memory reads, which is one of the biggest classes of issues, valgrind can still be invaluable. MSAN is one of the more difficult things to get setup and remove all the false positives. You typically need to transitively co
16.
by lbrandy 3y ago
> most Well... "a few" is probably more accurate. https://twitter.com/tjaltimore/status/1763571057703723344?s=...
17.
by lbrandy 3y ago
I struggle to resonate with what you are saying, as my experience is the opposite. I'm curious where this discrepancy is rooted. Reckless hypothesis: are you working on majority latency or majority throughput sensitive systems? I have
18.
by lbrandy 3y ago
I agree entirely with the premise here save one subtle bit at the start. I think there is grave danger in reducing "vector database" to "vector search" as equivalent domains and/or pieces of software. I would argue
19.
by lbrandy 4y ago
The only reason you and others seem so desperate to redefine bailout so you can call it a bailout is because you want to attach the negative connotations to that word to “hurt” the people you want to be hurt. It’s embarrassing, shortsighted
20.
by lbrandy 4y ago
We should be clear... in these large organizations... HEAD is always broken. But it has the advantage of being broken for everyone, tested by everyone, fixed by everyone, and thus fixed for everyone. And this usually makes it far better th
21.
by lbrandy 5y ago
FWIW, I joined HN 14 years ago (wow) and there's never been a time in those 14 years that HN wasn't convinced that facebook's demise was just around the corner.
22.
by lbrandy 5y ago
It depends what you mean by free will. Non-philosphers almost always mean "libertarian" free will when they say that, and yes, this same argument also basically outlaws libertarian free will as well. There is no physical mechanism
23.
by lbrandy 5y ago
> they seem to be advocating for some sort of dualism. This paper is doing the opposite. It's arguing that other papers/ideas advocating dualism are going the wrong way and purely physicalist explanations are the best path for
24.
by lbrandy 5y ago
Here's my attempt at a tldr. 1. The laws of physics at the time scales, space scales, and energies of the human brain are "known" and we have deep reasons to believe that more will not be discovered. He spends a great deal of
25.
by lbrandy 5y ago
if the team is remote, yes
26.
by lbrandy 6y ago
Software is more like building a machine than either math or art. There've been attempts to make formally-provable programs (so it _is_ like math) but these are not in widespread use. Go watch the Lockpicking Lawyer on youtube pick loc
27.
by lbrandy 6y ago
I used to post quite a bit more. Times have changed.
28.
by lbrandy 6y ago
This is the only good use of a const& shared_ptr which is when you 1) only _might_ want to claim shared ownership of the thing, and 2) the perf of those unnecessary refcounts matter. However, you missed a _super_ important caveat of thi
29.
by lbrandy 7y ago
hey i've worked on this stuff for awhile at fb. One important update is that the primary rationale for us, at the time, to change std::string's implementation to `fbstring` was to get SSO. We got huge perf wins for having SSO on s
30.
by lbrandy 8y ago
All of msvc, gcc, clang, and the isocpp committee have active work ongoing for C++ modules. We'll have them Soon™.

More ›