vitus

64 ms·

751.
by vitus 6y ago
Are you not proving my point by bringing the conversation back to "how will unionization impact engineers"? > Before going all-in on unions, consider if you’re ready to start making it harder for Googlers to get fired or make i
752.
by vitus 6y ago
The prevailing fixation seems to be on how unionization would help the highly-compensated engineers, when the people who need and would benefit most from unionization are the TVCs, who aren't considered Alphabet employees, don't g
753.
by vitus 6y ago
> I suspect that there are many at Google that would happily join a union, just look at how high the attendance of Google's walk outs have been. I'm hopeful that you're right, but I also suspect that a lot of engineers wil
754.
by vitus 6y ago
> Nobody really thinks Rust is faster than C now. The submission title would beg to differ. (I know, it explicitly calls out "benchmarks" as the context.) I think languages like Rust or Swift have significant advantages around
755.
by vitus 6y ago
You raise a good point. I've always been fascinating by PyPy's performance, personally -- anecdotally, I've achieved ~10x speedups from just running a script with `pypy` instead of `python`. I always attributed that to better
756.
by vitus 6y ago
Why bother to write the vtable in the first place? My experience is that C++ that's written for performance will often prefer the use of templates over inheritance since the cost is then paid upfront by the compiler. What's stoppi
757.
by vitus 6y ago
I recall an anecdote about how Haskell actually outperformed C on various tree benchmarks because it was using a better implementation. At some point, the C programmers got fed up with the airs of superiority from Haskell programmers, porte
758.
by vitus 6y ago
At the same time, a lot of those mechanics involve additional overhead (e.g. vtable lookups for dynamic dispatch per inheritance). But yes, some of these do provide hints for the compiler, e.g. constexpr, ownership semantics per unique_ptr.
759.
by vitus 6y ago
Tinyurl actually has a preview feature, which you can enable by default. https://preview.tinyurl.com/examplezoom Curiously, this specific tracking behavior (both the redirect and the cookie) goes away when turning on previe
760.
by vitus 6y ago
> Not going to get into terroir other than to say that to reduce it solely to the soil type is the wrong way to look at it. We're agreed on that front. My comment about limestone was to present one scientific aspect of terroir as _I
761.
by vitus 6y ago
I would be fascinated to see a scientific approach to mimicking the effects of terroir. I don't believe for one minute that it's magic, but I also don't think it's fully understood. Some aspects are explainable, e.g. lim
762.
by vitus 6y ago
should have been != was I don't know what the security situation is like at Godaddy, but I'm sure there's some amount of investment needed to roll that out broadly without accidentally breaking existing employee workflows. An
763.
by vitus 6y ago
Absolutely, in the context of a physical item being shipped to you (especially if they can't just distribute it at the office), if it's not through payroll. (e.g. we had site-specific fun events in lieu of the annual holiday party
764.
by vitus 6y ago
> You shouldn’t fail by clicking a link sent by an internal email address. I disagree in making this broad of a claim -- insider threats are certainly an issue. And as a sibling commenter points out, email headers are easily spoofed. I&#
765.
by vitus 6y ago
I went to a public 4-year university and transferred in community college credits that I'd previously used to satisfy high school graduation requirements. Requiring courses from a university is assuredly too high a bar, given that it w
766.
by vitus 6y ago
Chrome web store has no rollback strategy, there is only roll forward :(
767.
by vitus 6y ago
Depending on the company, "major" can be used to mean a number of things, such as visible and painful customer impact, or significant revenue impact. Based on the description of the outage (issues serving new ads to Messenger), it
768.
by vitus 6y ago
> I'm pretty surprised that BBR has such rapid uptake - the technical situation remains a little unclear. As I mentioned in a sibling comment, BBR has high uptake among Google properties, but not necessarily much outside; many of th
769.
by vitus 6y ago
Actually, the author seems to be a grad student in the field. From downloading the PDF and skimming the raw bytes, I see: <pdf:Producer>Microsoft® PowerPoint® 2016</pdf:Producer></rdf:Description> That s
770.
by vitus 6y ago
To understand congestion control, we need to first understand congestion and congestive collapse. In a naive world, we might have every computer try to send all the packets at once (up to some buffer if you're working with a reliable p
771.
by vitus 6y ago
C++20 designated initializers must be specified in definition order, whereas they don't in C99. In your example, Point point = {.z = 2.0, .x = 1.0}; produces a compilation error. This is annoying, but workable. And at least it
772.
by vitus 6y ago
(I've flipped the order of some statements, since I think it reads better this way.) > there is transmission as the article documents to Europe The article cites opportunity (Jesuit missionaries) and motive (Gregorian calendar refor
773.
by vitus 6y ago
> Dr George Gheverghese Joseph from The University of Manchester says the 'Kerala School' identified the 'infinite series'- one of the basic components of calculus - in about 1350. > The discovery is currently - an
774.
by vitus 6y ago
At Berkeley, for instance, there were multiple Physics sequences: engineering + physics + chem majors took the 7 series ("Physics for Scientists and Engineers"), which used calculus; for most other majors (e.g. premed), the 8 seri
775.
by vitus 6y ago
I'd assume not, given that Tim Berners-Lee hadn't invented HTTP yet. But, the website looks rather similar to how it did back in 2001 (with the recognizable two-column list of bullet points): https://web.archive.org&#x2
776.
by vitus 6y ago
Incident reports such as this one are incredibly valuable for understanding what went wrong and deficiencies that need to be addressed, but also what went well (e.g. point-in-time rollback as effective mitigation). The full timeline is valu
777.
by vitus 6y ago
Actually, price / earnings (P/E ratio) is typically 10-20 for _any_ company in the S&P 500. When you look at big tech, the numbers are drastically higher: - AMZN: 92 - GOOG: 34 - FB: 33 - NFLX: 76 - AAPL: 35 - MSFT: 35 Compare
778.
by vitus 6y ago
One thing to note: 538 uses a t-distribution (and calls out on regularly in their podcast that this yields much heavier tails than a normal distribution). Even 40,000 samples is not enough to characterize the tails. So, it seems to me that
779.
by vitus 6y ago
> Not all bugs can be reproduced by an automated test While that may be true, I'd say in this case, order-dependent tests are evil. This case study pretty much fits the definition -- the test result changes depending on what other t
780.
by vitus 6y ago
> In the 1950s, physicist Léon Brillouin published a landmark paper refuting the idea that adding a single diode, a one-way electrical gate, to a circuit is the solution to harvesting energy from Brownian motion. Knowing this, Thibado’s

More ›