yosefk

9 ms·

91.
by yosefk 2y ago
If Python's dependency management is better than anything, it's better than C++'s. Python has pip and venv. C++ has nothing (you could say less than nothing since you also have ample opportunity for inconsistent build due to
92.
by yosefk 2y ago
Most (all?) organizations are hierarchical above a certain size. I don't think agile ever addressed the resulting problems; some of it was designed for smaller teams. For an "agile" take on organizations of a medium double di
93.
by yosefk 2y ago
You say that good engineering is about achieving organizational objectives. I say that competent management is about setting organizational objectives. Therefore, you're saying that good engineering is about satisfying competent manage
94.
by yosefk 2y ago
An ability to operate reasonably effectively under the constraints of bounded rationality is an ingredient of competence. You seem to define managerial competence as making the right decisions, and thus file bad decisions under incompetence
95.
by yosefk 2y ago
I don't trust someone claiming to have run a heterodox org as much as someone claiming to have worked in one, like Creo sister comment threads. Any incentive turns perverse, and incentives are not how our zillion cells are prompted to
96.
by yosefk 2y ago
Thanks for your kind words! In terms of what I think & meant to say - I'm pretty sure that incompetent management (or incompetent at least in some key areas) can scale to fairly large org sizes, and that eventually competent manage
97.
by yosefk 2y ago
I did not attempt to cover heterodox organizations such as what Creo appears to have been in the article. I think these are extremely interesting, though quite likely to fail, regress to the mean, and be hard or impossible to scale or repli
98.
by yosefk 2y ago
Author here - I'm simultaneously quite pessimistic about, and very interested in heterodox organizational structures and especially real-world stories about them. I feel that failure or regression to the mean are quite likely and scali
99.
by yosefk 2y ago
TFA author - I have worked on low-cost embedded systems. (Tens/hundreds of megs of RAM, not KBs.) The only case where you don't hoard resources is if the system is so small that it's programmed by a small team since the funct
100.
by yosefk 2y ago
Had Mr Brooks proven this belief in a revealed preference sense, by voluntarily shrinking his budget, it would add a lot of weight to what is otherwise a nicely sounding but not very believable passage. But I doubt that anyone could be so i
101.
by yosefk 2y ago
You would want a CDN anyway to cache data across users, on servers close to those users, cutting the time & energy spent to serve the data.
102.
by yosefk 2y ago
In this case, using a CDN in front of the server is the thing which would improve things for people with limited bw/hw more quickly and to a greater extent than counting the fetches by Firefox on iOS and calling it out for fetching too
103.
by yosefk 2y ago
A CDN caches across users, lowers latency by having servers closer to the user, and lowers the bandwidth your server needs to handle. Cloudflare CDN has a free tier and I assume others do, too, so it's fairly easy to provision.
104.
by yosefk 2y ago
404s have got to be very cheap to serve
105.
by yosefk 2y ago
So 500 KLOC of C doing the same thing would be better?..
106.
by yosefk 2y ago
I also wonder how these requests "beat the shit out of the web server." It's requesting the feed and the favicon, both of which could be cached by a CDN. Even if they aren't, how much traffic are you gonna see from this
107.
by yosefk 2y ago
OTOH, hitting Ctrl-C in gdb and looking at a backtrace a few times would show you the leaf function
108.
by yosefk 2y ago
No shared mutable state in an imperative language is common, as are memory safe languages with performance close to C's? Didn't see the latter in the language shootout benchmarks, in fact Rust is much closer to C than the next clo
109.
by yosefk 2y ago
Thank you for your truthful and not unnecessarily nuanced comment!
110.
by yosefk 2y ago
I don't love the political agendas behind many of the attempts at AI safety, but it's not "just like humans." Humans understand what they shouldn't say; "AI" gives you black Nazi images if you ask it for &
111.
by yosefk 2y ago
adrian_b's comments in this thread are the kind of thing I referred to (my point wasn't that ASan specifically is presented as a panacea but rather various sanitizers in general.)
112.
by yosefk 2y ago
It comes up a lot in HN C++-related comment threads, for starters
113.
by yosefk 2y ago
One thing this explains is why ASan has false negatives. It's a great tool, but the typical comment that it fully mitigates memory safety issues is just not true (even assuming your tests actually trigger the memory safety bugs, which
114.
by yosefk 2y ago
CUDA is a moat. What's ineffective about it?
115.
by yosefk 2y ago
I meant register sets visible to software. The fact that there's even more hardware not visible to software that you need for a thread to run fast just means that the cost of adding another thread is even higher
116.
by yosefk 2y ago
The basic architecture still matters. GPUs still lose throughput upon divergence regardless of their increased ability to run more kinds of divergent flows correctly due to having separate PCs, and SIMD still has more trouble with instructi
117.
by yosefk 2y ago
Check out ispc/spmd by Matt Pharr, a very interesting take on this subject
118.
by yosefk 2y ago
It's improtant that GPU threads support locking and control flow divergence and I don't want to minimize that, but threads within a warp diverging still badly loses throughput, so I don't think the situation I'd fundamen
119.
by yosefk 2y ago
Barrel threaded machines like GPUs have easier time hiding the latency of bank conflict resolution when gathering/scattering against local memory/cache than a machine running a single instruction thread. So pretty sure they have a
120.
by yosefk 2y ago
It could say "cloud" instead of server and "your device" instead of cache and laypeople would get it, I think. It's unfortunate that servers are called cloud but at least there's a word that's commonly und

More ›