usaar333

46 ms·

301.
by usaar333 5y ago
It's so nice to be in tech where comparatively we let the free market figure out how to properly evaluate candidates; many companies have moved away from pure algorithm problems (the equivalent of testing) because relying solely on the
302.
by usaar333 5y ago
Right, this makes little sense unless there is little market competition. Alternatively, this might just be a morale thing and remote workers in the Bay Area don't actually make financial sense due to productivity losses. Google will p
303.
by usaar333 5y ago
In effect though, this is location pay discrimination which only makes sense if you believe the remote people have viable onsite alternatives where they are that drives their market salary. If they are only going to work remotely (i.e. in a
304.
by usaar333 5y ago
Exactly, so why have disparate pay by location? I'd expect all remote workers to get the same salary, perhaps a bit lower than onsite workers to account for productivity loss.
305.
by usaar333 5y ago
It's a weird policy though. If I'm remote, why does my location matter to Google?
306.
by usaar333 5y ago
The problem with COL adjustments is that they are very crude. Different workers consume more (and consume different baskets), so see wildly different impacts. Being "fair" then becomes risky from the point of view of everyone as t
307.
by usaar333 5y ago
> Had the US continued to vaccinate in the summer at the same rate as in the spring, the US would be near herd immunity by now. This obsession in the media with herd immunity is strange. It's not even clear herd immunity with Delta
308.
by usaar333 5y ago
Because it's the land that drives value in the Bay Area. I suppose you can use some weighted combination of lot and house, but I haven't seen common and simple valuation metrics there.
309.
by usaar333 5y ago
> Housing is actually more expensive outside of San Francisco, at least suburbs close to the city. Not per lot SQ feet
310.
by usaar333 5y ago
That's not answering though about capacity. UK's went up 8x, but they were nowhere close to hitting capacity. NYC's hospitalizations dipped to ~18 admitted/day citywide in June and are up to ~50 now. But that's a
311.
by usaar333 5y ago
That's before a considerable amount of population gained some level of immunity. UK experienced relatively low hospitalization in their last wave.
312.
by usaar333 5y ago
Is there evidence NYC's hospital system could be overloaded? UK just peaked ( https://coronavirus.data.gov.uk/); Does NYC have significantly worse vaccine coverage in vulnerable populations or worse capacity?
313.
by usaar333 5y ago
Upgraded by some specific UN agency. They don't have a monopoly on labeling countries developed and probably lag years behind
314.
by usaar333 5y ago
And Taiwan is ignored because the UN doesn't even consider it a country.
315.
by usaar333 5y ago
Alibaba and Tencent are hardly purely "get people addicted" companies. Together, they power nearly the entire retail transaction market in China. Alibaba heavily drives commercial transactions and Tencent powers the vast amount of
316.
by usaar333 5y ago
Unfortunately, the best and brightest often also like money. This incentivizes said people to emigrate.
317.
by usaar333 5y ago
I'm still confused on the semantics here. My data in this case is broadly representative of the real world input - so what's wrong? This is where I'm confused on the general idea of "algorithmic bias". Any maximally
318.
by usaar333 5y ago
I'm not following why cross validation helps here. One group (sunglasses) is simply much rarer so my algorithm may globally perform better on the overall set by accepting lower recall on the rarer subgroup (allowing for higher precisio
319.
by usaar333 5y ago
Maybe? I've always found the ideas on algorithmic bias lacking formalization. I'm not really sure what data or algorithmic bias really means. I think "data bias" means non-representative data and "algorithm bias&quo
320.
by usaar333 5y ago
You don't need to charge an EV every day... On a model 3 range, you maybe change 3x as often as fill up. Gas is a 10 minute overhead generally. So 90s vs 10 minutes per 400 miles or so.
321.
by usaar333 5y ago
> Charging still takes a long time Barring roadtrips, charging at home costs about 30s. Even factoring the longer time supercharging on roadtrips (relative to gas), I'm confident I spend far less total time charging my EV than fill
322.
by usaar333 5y ago
They note though that the follow up population that had caught covid after testing didn't differ significantly from those that didn't. That seems to add credibility here to a causal mechanism as opposed to a background correlation
323.
by usaar333 5y ago
It was a legislature introduced constitutional amendment; the legislature was heavily involved by virtue of voting to put it on the ballot.
324.
by usaar333 5y ago
Do schools compete on quality of education or on self-selecting their students? I suspect it's mostly the latter, given how parents evaluate public schools as well. You end up failing at your job not by virtue of teaching worse (you mi
325.
by usaar333 5y ago
ELC ( https://www.ucop.edu/enrollment-services/programs-and-initia... ). It doesn't get you into the top UC though (which Texas' program does), just any. There is wide variance in UC schools.
326.
by usaar333 5y ago
> So, by using AAPI students as a proxy argument for their weird screed at the end while leaving off factors like what percentage of that UG population is in-state v. out-of-state v. international does a disservice to that over/unde
327.
by usaar333 5y ago
We already know what happened as 2021 data is out: Applications: https://ucop.edu/institutional-research-academic-planning/co... Admissions: https://ucop.edu/institutional-research-academic-planning&#x2
328.
by usaar333 5y ago
CA does equalize funding at the state level and has since the 1970s ( https://en.wikipedia.org/wiki/Serrano_v._Priest ). There's only a small number of wealthy areas (1% of the population perhaps) that have excess f
329.
by usaar333 5y ago
Agreed, but that argument is more relevant to argue along the lines of how many in-state vs. out-of-state/international students you can educate. (Even then it's murky -- I can argue the international students both bring additiona
330.
by usaar333 5y ago
This is not a particularly strong article. Here's a much more data-backed one arguing similar points (though still with some bias): https://freddiedeboer.substack.com/p/you-arent-actually-mad-... Issues with this

More ›