Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
Osmium
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
11 ms
·
91.
▲
by
Osmium
9y ago
> I think Apple is so many years behind Google in AI that they have little hope of catching up. We are so early on in the AI revolution, it seems very premature to make this claim. I would wager we can't even imagine the scope of
92.
▲
by
Osmium
9y ago
Yes, to be clear, I'm certainly grateful I find (some aspects) rewarding -- if I didn't, it would be very irrational to continue, because it has very little else going for it. Academia is so dysfunctional right now. It's para
93.
▲
by
Osmium
9y ago
> A lot of hard work, very little financial certainty. Yes. It's exhausting. I'm several years past post-grad, have had a succession of one-year contracts, doing work I love and believe in and I genuinely believe has a lasting
94.
▲
by
Osmium
9y ago
> The cohort of young people that used AIM/MSN/Y! (depending on your region) aged up into Facebook Chat [1]. Speak for yourself, most of my social group is split between iMessage and WhatsApp. Even so, what Facebook has that My
95.
▲
by
Osmium
9y ago
> Those pesky rebels and their lasers blasters would just bounce off. They're not meant to be lasers in Star Wars (: They're packets of ionized gas (a gas Lando and friends mine at Cloud City).
96.
▲
by
Osmium
9y ago
> These companies also frequently market with 3D models: > - apple.com Actually, to my knowledge, Apple doesn't do this. All their product photography are photos of real objects.
97.
▲
by
Osmium
9y ago
Any alert is better than no alert, but if you're sitting right on top of a fault it won't give you much time to act. I'd be surprised if it's even enough time to get under a desk. Would any expert reading this like to co
98.
▲
by
Osmium
9y ago
> The bit about compressibility tickles my mind. One of my favourites is 'electride' bonding, where you compress a crystal so much that the electrons get squeezed into the gaps (interstices) between atoms instead, essentially m
99.
▲
by
Osmium
9y ago
Big fan btw :) Work a lot in crystallography, your work definitely inspires me to give it a go myself one day.
100.
▲
by
Osmium
9y ago
Perhaps a stupid question, but where does the $1bn. valuation come from? Amazon already make hardware, including cloud cameras, would it have been so hard for them to make a doorbell in-house?
101.
▲
by
Osmium
9y ago
The reviewer seemed to be very frank about listing their biases in fairness (see the big section titled 'Bias' in the review). And they shared their original data, so it's up to you what you want to make of it.
102.
▲
by
Osmium
9y ago
That's part of the point: it has a 7-tweeter 360º array that is used to add directionality to the sound. Obviously won't compare to a true stereo pair, but it's not nothing either. Personally speaking, I'd take great mon
103.
▲
by
Osmium
9y ago
iOS does too, along with a number of other display accommodations. It has adjustable filters for specific forms of color blindness, along with an Ishihara-like test pattern to help tune the filter to your specific needs.
104.
▲
by
Osmium
9y ago
> How can you be so sure of that? ... Do you think we could figure out a material designed by a being a million years in advance of us? It's a fair question, but I'm pretty sure we could. We have constraints: we know the objec
105.
▲
by
Osmium
9y ago
I read it as saying it was impossible that there were materials being stored that we couldn't identify[0], not that the existence of the materials themselves was impossible. The point being that if these materials exist it should be qu
106.
▲
by
Osmium
9y ago
> based on their current understanding rather than testing the claim. The point is that there is no claim. It is very very straight forward to identify an unknown material (figuring out how it was made, what it's for, etc., is the d
107.
▲
by
Osmium
9y ago
Agreed! I think I've seen it before as a theme, but I'm not sure what it's called. Wish there was a CSS version available somewhere. (Edit: thought it looks like it's not using small-caps variants for numerals, which is
108.
▲
by
Osmium
9y ago
Likewise. Social engineering is a big concern. I understand the risks of getting locked out of my account, but would much prefer a stricter system -- along with published guidelines on Fastmail's process for handling these cases. Being
109.
▲
by
Osmium
9y ago
Haha, cool! Hi :) Given the small size of the laser projector, I imagine natural movement from the phone being hand-held would result in significant displacement of the projected dots over a 1s interval? Have you tried integrating the 15 fr
110.
▲
by
Osmium
9y ago
> 2) The "TrueDepth" data is not really all that granular. It seems similar to the depth map you get from the rear-facing cameras on the "plus" sized models. Here's what the sensor data spits out: https:/&#
111.
▲
by
Osmium
9y ago
I wouldn't say no benefit ... instant directory size information, instant copies are both pretty cool.
112.
▲
by
Osmium
9y ago
It's also the case that often there aren't spaces between words. One of the surprising things when you first start doing audio analysis is that the pauses between words that you perceive aren't actually there in reality --
113.
▲
by
Osmium
9y ago
Submitted due to interesting discussion from Chris Lattner on adding Python interoperability into Swift (something I would love to see, and seems like it could work quite elegantly).
114.
▲
Swift for Data Science / ML / Big Data Analytics
(lists.swift.org)
4 points
by
Osmium
9y ago
|
2 comments
115.
▲
by
Osmium
9y ago
Thanks, this looks really useful too! I had no idea it was there.
116.
▲
by
Osmium
9y ago
Draco looks exactly what I need actually -- thank you!
117.
▲
by
Osmium
9y ago
Slight tangent, but does anyone here have recommendations for a good serialization format for 3D models for the web? Think for loading into Three.js. Ideally something that can be converted to/from Collada. It seems like a lot of the s
118.
▲
by
Osmium
9y ago
Excellent! Good to hear :) Also saw in your documentation that you're concentrating on lossless compression right now, which makes complete sense. However, as a scientist, I just want to put a vote in for lossy compression too: it'
119.
▲
by
Osmium
9y ago
Looks exciting. What's the easiest way to get started with this via Python/NumPy? Looks like this is a design goal, but not currently supported.
120.
▲
by
Osmium
9y ago
> A "mole" is a unitless quantity. Literally it's just a number. True, but it's defined relative to the kg. It's the quantity of Carbon-12 atoms in 12g of Carbon-12. As for why it isn't an 'simpler'
More ›