mjhay

32 ms·

481.
by mjhay 4y ago
The redesign does the important job of making the app look better in comparison, as you are reminded every time you load a page (if it loads properly).
482.
by mjhay 4y ago
There's a lot of interesting philosophical/metaphysical implications of quantum mechanics, not the least of which is the particular observationally-equivalent interpretation (e.g. Copenhagen or Von Neumann-Wigner[0]). This essenti
483.
by mjhay 4y ago
Our CO2 output is increasing . Under the world's current political/economic regime, we can't keep our CO2 output stable, let alone decrease it. If we lack the seriousness to do that, we definitely lack the ability to do the
484.
by mjhay 4y ago
Those are in Nevada and Idaho, not Arizona.
485.
by mjhay 4y ago
The algorithm for linear PDEs usually boils down to some kind of meshing/discretization (often the hard part) to make a (usually sparse) linear system to solve using standard numerical methods. In the basic 1d, first-order case, it win
486.
by mjhay 4y ago
The open, standards-compliant documentation wasn't just nice for devs, but it promoted web standards that are meant to foster an open, better-functioning internet that's better for users and Firefox's market share. The Servo
487.
by mjhay 4y ago
Why hasn't LibreSSL taken off? I thought for sure it would after heartbleed. I assume it's mostly network effects/laziness, despite being fairly compatible (at least when it originally forked) and everyone already using OpenS
488.
by mjhay 4y ago
And tobacco industry money
489.
by mjhay 4y ago
A broken clock is right twice a day.
490.
by mjhay 4y ago
I'm excited to see where spiking NNs go. Something like that is needed to progress now that conventional NNs on GPUs are pretty much tapped out (in terms of non-incremental advances) from their power consumption and the end of Moore&#x
491.
by mjhay 4y ago
Are you familiar with the term "negative externalities?"
492.
by mjhay 4y ago
Not practically. We are talking about huge scales, such that the industrial production, transport, etc., would be as much of an environmental disaster as what it's trying to solve.
493.
by mjhay 4y ago
Or we could just stop burning coal.
494.
by mjhay 4y ago
What makes you think it would be significantly easier to filter out the carbolic acid from the ocean? The same difficulties apply, and water is much more viscous than air.
495.
by mjhay 4y ago
> Remember, those hiring managers and HR are the same people as you -- got the paycheck, try to do as little as you can to just slip under the radar. Sounds like these companies should hire better HR people, because they aren't doin
496.
by mjhay 4y ago
There's money companies I could work for without this attitude. I shouldn't need to "jump through hoops" to fill out an application. That sort of thing is just a red flag that shows me I wouldn't actually want to wo
497.
by mjhay 4y ago
Getting a job these days sans personal collections requires sending out many, many applications. Each job one applies to only has a small chance of getting a callback. Given that, I'm not going to spend 30 minutes filling out pages of
498.
by mjhay 4y ago
The real secret to Greek fire was the Byzantine's unusually advanced "siphon," which heated the mixture in a pressure vessel to allow it to spray fairly long distances. The main use was on ships of course, but there was also
499.
by mjhay 4y ago
Also, a huge portion of food production is dependent on bees and other pollinators.
500.
by mjhay 4y ago
The highest quality education I received was at a community college. The undergraduate education I got after transferring to a well-regarded research shool was lower quality, with larger classes and professors who weren't incentivized
501.
by mjhay 4y ago
> I’ve since moved to Seattle and it’s crazy to see shit can actually get done here. I've never heard Seattle described as a place where "shit can actually get done," but I suppose its not quite as bad as SF at this point.
502.
by mjhay 4y ago
It's really the "more advanced" way to do it. There are a lot of people, myself included, that are actually much better at thinking in abstractions - I suspect the author might be the same, despite her thinking of over-abstra
503.
by mjhay 4y ago
The existential crisis you are talking about is happening right here, right now - and nothing meaningful is being done about it. The correct course of action is to reduce further damage, and not count on our ability to tech our way out of t
504.
by mjhay 4y ago
Crushing many cubic kilometers of rock, transporting it, and then spreading over large areas takes quite a bit of energy, you'll find. It's not easy as you say, and only specific types of silicate rocks work. You seem to keep thin
505.
by mjhay 4y ago
I don't think you are grasping the full, physical scale here. This is a lot of CO2, and you need a lot of crushed rock. This is not like integrated circuits or anything - it would be an enormous heavy industrial operation the likes of
506.
by mjhay 4y ago
You're describing small systems being scaled up millions of times. It reminds me of the doomed and misguided Ocean Cleanup Project. Where would you grow all this algae? It would take a great deal of area (easily multiple thousands of s
507.
by mjhay 4y ago
"Negative-emissions" carbon-capture technologies have essentially intractable problems scaling up, including basic thermodynamics; that isn't even talking about the massive resource inputs to build a system capable of meaning
508.
by mjhay 4y ago
The second one. It's important to have a clear picture of biases and how your model could fail to generalize. In the case of social data from a deeply biased society, these issues can be seen a mile away.
509.
by mjhay 4y ago
Social media managers form an important work program for executive's fail-nephews, so I don't think what you are suggesting is realistic.
510.
by mjhay 4y ago
Once these systems get complicated enough, it gets really hard to do that. Unless you are doing linear models or tree ensembles, it can be really difficult to get a clear picture of variable importance/effects. It's one thing to e

More ›