HNHacker News
TopNewBestAskShowJobs

jozvolskyef

424 karma · joined January 5, 2018

submissionscomments
jozvolskyef··on Linux Evangelist Turned Dentist, handrolls all IT infra in his dentist practice
OP specifically mentions the sensor, so I'd speculate they use an interface to download data which is different from the control mechanism.
jozvolskyef··on Commonly used idioms in the tech industry
I just tried searching HN for occurrences of "is orthogonal", and it does seem that most uses would be better served with "unrelated". In some cases, however, orthogonal sounds more appropriate, and the reason are the geometric associations that it evokes. For example in the sentence "Having chip/pin is orthogonal to the card being a credit or debit card", orthogonal evokes the image of a contingency table where all the cells are non-zero. The word "unrelated" doesn't have the same effect despite being synonymous.
jozvolskyef··on Bottled water is 3,500 times worse for the environment than tap water
In 2015, approximately 11000 people got sick in Dejvice, Prague 6, after wastewater contaminated the fresh water supply[1]. I was out of town, but when I returned in 2016/2017, the incident was still fresh in people's memories and their water consumption patterns clearly changed in favour of bottled water.

[1]: https://vodnihospodarstvi.cz/dejvicka-havarie-a-epidemie-neb...

jozvolskyef··on What’s it like living with a phenomenal memory and can it be learned?
When people say 'I never said that,' the reason why they're saying it is usually more important than the statement's veracity.
jozvolskyef··on Where does the plastic in oceans come from?
If by "first world" you mean the EU, they now prohibit the export of plastic waste outside OECD[1].

[1]: https://ec.europa.eu/environment/news/plastic-waste-shipment...

jozvolskyef··on New UUID Formats – IETF Draft
I presume you are referring to UUIDv4/1, which has 122 random bits. It is a common misconception that this is insecure. In reality, the UUIDv4/1 is just a way to encode 122 random bits, which is more than enough for most needs. The property of being practically unguessable comes from the random generator, not from the encoding.

Secondly, whether or not being predictable is a problem depends on the use case. The keys used on this forum are predictable and it is not an issue.

jozvolskyef··on New UUID Formats – IETF Draft
What are the advantages of sortable UUIDs with embedded timestamps over random 128 bits and a created_at column?
jozvolskyef··on The rise of never-ending job interviews
If you ever carried a canister of gasoline, you'd notice it to be uncannily light. It may have been a practical experience question.
jozvolskyef··on Math Problems for children from 5 to 15 (2004) [pdf]
Maybe try explaining to him that school and other institutions tend to optimize for the average person, so unless his goal is being average, he'll need to take responsibility.
jozvolskyef··on Analyzing the legal implications of GitHub Copilot
From what I understand about these models, having studied and applied machine learning but not deep learning, they all build abstractions (embeddings) akin to eigenfaces[1][2] where many samples of a feature are turned into an abstract feature. These abstract features are then manipulated by the model. Sure, just like with Fourier transform-based compression, anything can be reconstructed from primitive features, it is called overfitting and the challenge of training these models is building a learning process that achieves generalization rather than overfitting. It is not true that 'it is not clear that there is abstract learning', because without it, generalization is impossible, and without generalization the model wouldn't pass even the most trivial benchmarks.

[1]: https://media.geeksforgeeks.org/wp-content/uploads/202003171...

[2]: https://www.geeksforgeeks.org/ml-face-recognition-using-eige...

jozvolskyef··on Analyzing the legal implications of GitHub Copilot
If a person who's never seen a goat looks at a million copyrighted images of goats and draws a goat, are they committing copyright infringement?

What if an algorithm does the same? The result is 'a recomposition of learned fragments' in either case.

jozvolskyef··on Facebook's “see no evil” strategy
You are making a straw man argument by assuming that GP claims they can infer individual-level attributes from telemetry while all they claim is to infer population-level attributes.
jozvolskyef··on Tact Filters (1996)
One can have social anxiety and a rock-solid self-image as a scientist at the same time. The latter requires no social or synchronous communication skills.
jozvolskyef··on Norman doors are everywhere (2016) [video]
> Doors with a button off to the side to unlock?

It is often power assisted doors that have a button on the side. The button can be difficult to notice, among other reasons because it is low in order to be reachable by wheelchair users.

jozvolskyef··on Tact Filters (1996)
Brilliant article. Explains why I find myself frustrated with people who take my well-intentioned feedback and try to retaliate against it as if it were some kind of a game.
jozvolskyef··on I Stopped Using Emojis
From what I remember, sentiment analysis should reveal the cold and aggressive association to date at least to early 2000s ICQ communication in the Czech-speaking 7-15y/o cohort.

The first time I can remember having to actively remind myself that people who grew up without internet use periods differently was about 2012.

jozvolskyef··on Eigengrau
Another interesting phenomenon is seeing white blood cells swishing through the capillaries in front of the retina[1]. They are most noticeable looking at the blue sky on a sunny day.

[1]: https://en.wikipedia.org/wiki/Blue_field_entoptic_phenomenon

jozvolskyef··on German region to fine Tesla for illegal construction
Are the permits and rejections publicly available? From the article seems that Tesla didn't intend to use the tanks without permission.
jozvolskyef··on China’s gene giant harvests data from millions of women worldwide
When I wrote "since the data can't be linked to people's identities", I was talking about 23andme research partners. They don't get any PII, and I presume neither does BGI as there is no reason for them to have it, unless they offer tests directly to consumers.
jozvolskyef··on China’s gene giant harvests data from millions of women worldwide
> The women, who signed consent forms stating that their genetic data would be stored and used for research, said they did not realize their genetic information could end up in China.

How do people not see the bias?

23andme doesn't specify which countries can do research on their clients' data, and for one I don't mind whichever country it is since the data can't be linked to people's identities. The only way to link it is to collect the person's DNA, at which point there is no point in using the database.

jozvolskyef··on China’s gene giant harvests data from millions of women worldwide
> the company is a pivotal player in a genomics race between China and the United States

The entire reason for this article. Surely if one were to use a similar US-based service, NSA and US military access to the data couldn't be ruled out.

jozvolskyef··on Poisson's Equation
Thank you for the encouragement. I'm still hoping to solve it one day. No spoilers, please :)
jozvolskyef··on Poisson's Equation
The generation growing up today is the first that can broadly learn mathematics from mathematicians rather than from teachers, and we should encourage them to do so.

When I was 12-16, I was interested in physics enough that I would study it in my free time just for fun, solving problems from the Physics Olympiad. I could solve lots of the problems with just intuition, until I stumbled upon the sliding chain problem. I spent about a week pondering over it with no results. I approached my math teacher. He admitted defeat and referred me to the physics teacher. If I remember correctly, the physics teacher avoided admitting he didn't know how to solve it and didn't give me any pointers. That was the last Physics Olympiad problem I tried to solve, after dreaming of attending the competition for years (I should mention that at that time my primary interest already were computers; physics was just a hobby). I wonder whether this could still be a problem today when anything can be found on YouTube. There's a lot of noise, too. How long does it take the average curious kid to find the signal that is 3b1b, MIT OCW, etc?

jozvolskyef··on Leaving Google
That might explain why cuelang got a new home[1], moving from the google-managed github/cuelang to an independent github/cue-lang.

[1]: https://github.com/cue-lang/cue

jozvolskyef··on Leaving Google
You are probably misreading it as referring to a standard medical expense while it probably refers to some advanced treatment that is only available in private healthcare, anywhere in the world.
jozvolskyef··on I do not agree with Github's use of copyrighted code as training for Copilot
Thank you for pointing that out, I was indeed wrong to assume it was used to classify hardware rather than a model.
jozvolskyef··on I do not agree with Github's use of copyrighted code as training for Copilot
How so? The page you link offers three definitions[1], and all of them require an infinite tape.

You could argue that a stack is missing in my simplified model of the human brain, which would be correct. I used the simple model in allusion to the Chinese room thought experiment which doesn't require anything more than a dictionary.

[1]: https://en.wikipedia.org/wiki/Turing_completeness#Formal_def...

jozvolskyef··on Underground fiber optics spy on humans moving above
> Because the speed of light is a known quantity, the Penn State researchers could shine a laser through a single fiber optic strand and measure vibrations at different lengths of the cable by calculating the time it took the scattered light to travel. The technique is known in geoscience as distributed acoustic sensing, or DAS.

It turns out Marauder's Map was enabled by an underground array of optical cables.

jozvolskyef··on I do not agree with Github's use of copyrighted code as training for Copilot
The human brain isn't Turing-complete as that would require infinite memory. I'm not saying that GPT-3 is even close, but it is in the same category. I tried playing chess against it. According to chess.com, move 10 was its first mistake, move 16 was its first blunder, and past move 20 it tried to make illegal moves. Try playing chess without a chessboard and not making an illegal move. It is difficult. Clearly it does understand chess enough not to make illegal moves as long as its working memory allows it to remember the game state.
jozvolskyef··on I do not agree with Github's use of copyrighted code as training for Copilot
Imagine for a second a model of the human brain that consists of three parts. 1) a vector of trillion inputs, 2) a black box, and 3) a vector of trillion outputs. At this level of abstraction, the human brain "pattern matches and replicates" just the same, except it is better at it.
← PreviousPage 3 of 5Next →