Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
entee
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
32 ms
·
241.
▲
by
entee
8y ago
I'd add that a PhD in the sciences is like getting punched in the face by mother nature every day with a, "hah! try again!". The rigorous thought process that cultivates by the end is hard to replicate without those very hard
242.
▲
by
entee
8y ago
This is such a complicated issue, I'd highly recommend the following as a primer: https://www.forbes.com/sites/matthewherper/2015/10/13/four-r... Also Derek Lowe is invariably awesome in this s
243.
▲
by
entee
8y ago
True, though I think a mixed school where a substantial portion of the day is just held in that language (i.e. math is taught in spanish or something) can bridge it somewhat. I think I'm a bit idealistic in this area though. Also I
244.
▲
by
entee
8y ago
Practically speaking: French is similar enough to latin that most SAT vocab prep was far easier. It also helped when I learned spanish, because again the two are quite similar. Spanish has been useful at several points when I spent some tim
245.
▲
by
entee
8y ago
As someone who grew up bilingual, I think knowing another language has been invaluable to how I approach and understand the world. I disagree (fully acknowledging my bias) that learning another language is a waste of time. However, I think
246.
▲
by
entee
8y ago
Oh yes, of course there are many motorized processes, but the point is there's no inherent logic to them. A myosin grabs onto a filament and goes to the plus end of the actin filament (usually, not some of them, again it gets complicat
247.
▲
by
entee
8y ago
I believe the components are indeed drawn to scale! The cell is a mess, there is virtually no central coordination, just individual molecular signals. A protein gets phosphorylated by a receptor, then its on its merry way till it hits somet
248.
▲
by
entee
8y ago
This is super cool! People underestimate how insanely chaotic a cell is at a molecular level. Often diagrams show blobs cleanly interacting but the reality is more like the images linked here: https://mgl.scripps.edu/people&
249.
▲
by
entee
8y ago
I think you're right. Still, screw it. If nobody ever takes a stand, this sort of thing will keep happening (and good on the lawyer representing the shareholders for suing) EDIT: clarification about shareholder suit
250.
▲
by
entee
8y ago
1.) If a founder got baited into illegal acts by a VC, they should be fired immediately. Not sure how you fire a VC, but that sort of thing should end that VC's career as well. 2.) It's irrelevant that he built a billion dollar bu
251.
▲
by
entee
8y ago
Golden Parachutes are usually in place for termination without just cause. Companies like Google (and many many others) prefer to have a friendly separation with no litigation so they just have them trigger (or give some other similar packa
252.
▲
by
entee
8y ago
Defense contractors, construction occasionally. It's pretty nuts. The east span of the bay bridge was something like 5x the original cost estimate. I think some of this ends up being because of the same reasons that home general contra
253.
▲
by
entee
8y ago
Fair enough! I didn't know some of those numbers, maybe there's more hope :)
254.
▲
by
entee
8y ago
For sure! But cheap energy is hard :( I'm cheering for Solar, renewables and Nuclear Fission to become cheaper and more widespread. Fusion... well it would be nice, but given that it's never been demonstrated in a form suitable fo
255.
▲
by
entee
8y ago
One of the key questions for desalination is how good can it get. It turns out that many desalination plants are probably close to the thermodynamic limit [1, 2], which means that even with new technologies it's likely energy costs are
256.
▲
by
entee
8y ago
Lumiata | Onsite | San Mateo | Software Engineer - Data Platform, Principal Software Engineer - Data Platform, Senior Software Engineer - Web Platform We're a medical technology company using machine learning and massive amounts of pat
257.
▲
by
entee
8y ago
As a very long time Strava user, I think they actually don't deserve any kudos here. They have the "best" app in the sector, but only because they have essentially the only sports social network out there. They have done near
258.
▲
by
entee
8y ago
Speaking personally, the things that give me the greatest satisfaction are those where I feel I've done a good job from a perspective of craft. Even if the code itself doesn't do very much or will never really be seen, I get the g
259.
▲
by
entee
8y ago
Could be a sales gig, commission can push total comp quite high, particularly in enterprise sales gigs.
260.
▲
by
entee
8y ago
Perhaps as a byproduct we'll end up with more value placed on human-curated and generated information. Just because there's a lot of noise in the world doesn't mean there aren't objective truths. If FB and Insta get shun
261.
▲
by
entee
8y ago
Fully agreed. Most questions I got when I started out as a data scientist interviewing for jobs were pretty simple: "use a hash table to count stuff, join stuff, lookup stuff". However not once did I get asked, "why do databa
262.
▲
by
entee
8y ago
Very, very hard. The equipment to do this isn't trivial but not complicated and largely available in biohacker spaces or on ebay even. That assumes you know how to make a super virus though, that's very very hard, we don't re
263.
▲
by
entee
8y ago
This paper appears to be a mix of what you're suggesting, original paper here: http://science.sciencemag.org/content/351/6280/aad6253 They tried paring down a genome to its essential elements and failed.
264.
▲
by
entee
8y ago
It's a simple fact that vaccines are safe and perhaps the best medical tool we have along with antibiotics to prevent mass death. Only 0.0001% of vaccination doses result in significant complications (roughly 3,800 compensated injuries
265.
▲
by
entee
8y ago
Lumiata | Onsite | San Mateo | Software Engineer - Data Platform, Principal Software Engineer - Data Platform, Senior Software Engineer - Web Platform We're a medical technology company using machine learning and massive amounts of pat
266.
▲
by
entee
8y ago
I question what objective standards are here. Human intelligence sucks in terms of consistent long term memory, and possibly in terms of not being influenced by emotions/outside forces. It's pretty good at a number of things compu
267.
▲
by
entee
8y ago
My favorite is that umbrella usage is highly correlated to rain.
268.
▲
by
entee
8y ago
I've been at a company where the 1-1 firings occurred over a morning. I was fortunate to keep my job, but I can tell you it was a pretty awful day for those staying (obviously not as bad as for those leaving :(). There's no good w
269.
▲
by
entee
8y ago
Lumiata | Onsite | San Mateo | Software Engineer - Data Platform, Principal Software Engineer - Data Platform We're a medical technology company using machine learning and massive amounts of patient data to improve how the healthcare s
270.
▲
by
entee
8y ago
I think primary care workers are generally underpaid and specialists are generally overpaid. That said there's a lot of variation in the data. If you look at some studies of why the US spends more than other countries on healthcare wit
More ›