Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
joshvm
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
84 ms
·
841.
▲
by
joshvm
7y ago
They're not necessarily fake, but they may be incentivised. I've had offers of a £5 voucher for a photo review or £10 for a video (the product was a £15 solar panel). No idea what happens if you slate the product, but amazon would
842.
▲
by
joshvm
7y ago
No mention of rail? Not a solution in the US, but virtually all of Europe and Asia is connected by a dense, largely inexpensive, rail network which is increasingly being electrified. That puts the onus on the grid, but there is a trend towa
843.
▲
by
joshvm
7y ago
You can use Eulerian image magnification for that. Just needs a normal camera: http://people.csail.mit.edu/mrub/evm/
844.
▲
by
joshvm
7y ago
Brits can look at https://www.objectsofuse.com/ (there's an actual store in Oxford).
845.
▲
by
joshvm
7y ago
Aside from the runtime difference between training and inference, having on-device ML makes a lot of sense for other reasons. There can be more guarantees over data privacy, since your data can stay on-device. It also reduces bandwidth as t
846.
▲
by
joshvm
7y ago
Always worth looking at a point cloud versus a disparity map. Grayscale disparity/depth maps are somewhat misleading - the large regions of constant intensity suggest that the algorithm is good at segmenting areas of constant depth. Ho
847.
▲
by
joshvm
7y ago
Something I found out, after participating in HashCode was that Google have a pretty comprehensive optimisation library called OR Tools. There is a module specifically for routing. I don't know what Google uses in production, but presu
848.
▲
by
joshvm
7y ago
Spaced repetition aids are fairly common. If you can store your curriculum as flash cards, then apps like Memrise and Anki are designed specifically for this. Spaced repetition is not the same as interleaving, but the apps will force you to
849.
▲
by
joshvm
7y ago
Implicitly, when you print your visitor badge you consent to the NDA. Though that may be a generic site NDA, not an interview one. I never had to provide an actual signature, aside from a verbal agreement with the recruiter to not share int
850.
▲
by
joshvm
7y ago
It sounds believable to me. I've had some interactions with big companies and you have to readjust what you think is a reasonable price for something. I work with hardware and you'd be amazed how much people will spend without bat
851.
▲
by
joshvm
7y ago
In real life, eg aviation, you would formally verify the code and the firmware that interacts with hardware. https://en.wikipedia.org/wiki/Formal_verification
852.
▲
by
joshvm
7y ago
If you have no experience in digital electronics at all, I would actually start with a book like Code by Charles Petzold. It's a pop-science book about electronics and goes from light bulbs and switches to simple logic blocks like ad
853.
▲
by
joshvm
7y ago
Matt Walker makes this point in his book (at the risk of sounding like a shill). One group of people who sleep abnormally long hours are those suffering from various illnesses - they sleep more either as a result of the illness or while in
854.
▲
by
joshvm
7y ago
Thanks for that, nothing has changed in academic review! Reviewer 2, 3: 9/10 great. Reviewer 1: "The paper seems a bit preliminary and unfinished." Authors go on to win best paper with their submission.
855.
▲
by
joshvm
7y ago
Conda gives you two real advantages: the first is straightforward virtual environments (you could also use venv) and secondly better dependency management than pip in some cases. There are certain packages which aren't available on pip
856.
▲
by
joshvm
7y ago
Reading Why We Sleep has made me somewhat skeptical about people claiming to not need sleep. A quote from the book: > We have, however, discovered a very rare collection of individuals who appear to be able to survive on six hours of s
857.
▲
by
joshvm
7y ago
In my case, I guess I am? Obviously the site owner will get some logging results, and maybe they will try to optimise for that. I'm fine with a site owner knowing their demographic. That's potentially useful for other things, eg s
858.
▲
by
joshvm
7y ago
One grey area that I approve of is referral links, if they're obviously presented as such. For example if you read a tutorial on how to build a weather station and the author provides referral links to a component on Amazon, I'm h
859.
▲
by
joshvm
7y ago
I work with thermal imaging cameras. The rules are very specific on what can be exported, but it's not a blanket ban. I've asked FLIR for CAD documents before and they just sent them. The actual technology inside the systems is ve
860.
▲
by
joshvm
8y ago
Part of the study involved decoding images from memory/recall. I did say _limited_. It depends how tightly you want to define "thought", but being able to think of an object and having a computer display a reasonable represen
861.
▲
by
joshvm
8y ago
Actually you can, in limited cases. If you have prior knowledge of the person's brain activity when looking at an image, it's possible to predict (reconstruct an image of) what they're looking at. https://journals.
862.
▲
by
joshvm
8y ago
Fantastic! What extra steps would be required to support the GBC? Do you have to provide dual palettes? Or does the GB guess? (are there other differences aside from colour? I guess for the transparent paks)
863.
▲
by
joshvm
8y ago
So you want to figure out how often you need to sample products for QA? A model is two things: a description of what's in the black box (could be a linear model, a neural network architecture, etc) and some weights which uniquely defin
864.
▲
by
joshvm
8y ago
PyImageSearch [1] is one of the best resources for doing actual stuff with opencv. I personally find the writing style a little patronising sometimes, but the content is excellent and kept up to date. [1] https://www.pyimagesearc
865.
▲
by
joshvm
8y ago
> If you have any suggestions, please, I am all ears. (I am in Europe) What was your MS program like (research/taught/etc)? Happy to give some advice, but a bit more background would be useful.
866.
▲
by
joshvm
8y ago
It's supposedly possible to predict to a reasonable degree of accuracy where a roulette ball will land if you know its speed (ie time it hits the wheel, and time to pass a predefined marker). https://www.insidescience.org&#x
867.
▲
by
joshvm
8y ago
Yup, dump on a piece of paper and segment by contours, bright colours so super easy. Though you'd lose the nice skittle graphs which I think is half the fun of the process.
868.
▲
by
joshvm
8y ago
It's not just pointing (2D) but also getting the baseline between telescopes. This is also being investigated for the LISA system https://en.m.wikipedia.org/wiki/Laser_Interferometer_Space_A...
869.
▲
by
joshvm
8y ago
Previous HN discussion about seeing round corners: https://news.ycombinator.com/item?id=17902177
870.
▲
by
joshvm
8y ago
Yes, in theory. The challenges are likely to be dish size and getting accurate enough positioning. Not to mention the downlink capacity required (and also onboard storage).
More ›