HNHacker News
TopNewBestAskShowJobs

tokinonagare

581 karma · joined May 29, 2022

submissionscomments
tokinonagare··on Thousands of scientists publish a paper every five days (2018)
Thanks for putting some realism in the thread. I've been a PhD student and the involvement of the lab head ranged from actually giving some research direction to doing jack shit depending on the paper and the people leading them. It was very obvious once when he green-lighted one of my paper for submission yet wanted to changed half of it when I made an internal presentation about it a week later.

I challenged the thing one time to an intermediate supervisor (who where at least really contributing a bit) and I got bullshit excuses like "he's securing funding", "running the lab"... well at least not for my project since it costed nothing and I came with my own funding. At this point the toilet cleaning lady was contributing more to the paper because of how she allowed me to work in a clean environment.

So I'm not surprised by Japan being over-represented like mentioned in the article. My lab explicit policy was "quantity over quality" written in a presentation given to new student. I know have a mental, logarithmic curve of contributions from authors: first made 90% of the work, second 9% and the following authors share the last 1%.

tokinonagare··on Rethinking Virtualization for Backends
True. A static executable doesn't even need Linux in the first place, that can be done with a unikernel. It's a tradeoff about familiarity, size, performance, documentation and technical level.
tokinonagare··on HN: The Good Parts (2016)
Well said, save for the fact that HN is a link aggregator. I miss the old style fora, with their sections and content from members, not just a discussion based on links to an external website.
tokinonagare··on Rethinking Virtualization for Backends
It's exactly what it is, except there is also a browser implementation without plugin. Which also means that WASM/WASI is not a total replacement for containers: just like there as JVM image there will be WASI images. The argument of size given in the article isn't very convincing, when images like Alpine exist.

The problem addressed in the post is more about getting around Rust slow build time than phasing out containers. There's a lot of buzzwords (WASM, Rust, Cloud, etc.) but at the end of the day nothing that isn't easy to do with a modern "boring" stack like .NET.

tokinonagare··on Most buildings at MIT campus are remaining closed to the public
> In France, the schools, universities and pretty much any other establishment are closed to the public. [...] In business hours, if you are a member of the university, you must present ID and pass through security or welcome desk

Which damn universities are you talking about? I sure have no issue roaming on the campus of the local university, going to libraries and administration and visiting former professors without any security check. That would be hard in the first place given how vast and open is the campus. The only place some sort of check are mandatory (half of the time) is one of the student association asking to show the membership card when buying beers.

At some point after Charlie Hebdo there was annoying security checks in the parisians univs I was attending but that stopped within a year or so.

There is usually one vigil in stores, but certainly not "patrols". For the record, countries with no immigration issue like Japan or Czech Republic don't have this kind of security in stores. Univ cafeteria in Japan is also open to all, and there are occasionally some hobo going in.

tokinonagare··on Show HN: Bolt.css – Another classless CSS library
The range/output that display 0 or 100 but don't touch the ends of the slide rubs me in a very wrong way. Besides, nice style.
tokinonagare··on Available Today: Windows Dev Kit 2023 a.k.a. Project Volterra
> the point of this machine is to give Developers

At 700€ price tag what a gift! It's 3-400€ too expensive for a very dispensable toy, especially given that the managed stack (.net) of Microsoft development tools can be tested on Raspberry Pi or M1, which are both very popular with developers.

tokinonagare··on AI translates Hokkien, an unwritten language, for the first time
> Even Japanese, a language used by 125 million, has similar issues, my Japanese coworkers frequently switch to English during technical discussions.

Are you Japanese yourself? If not I don't think it's strange that they would adapt their way of speaking with a foreigner, especially since most technical words in IT are coming from English anyway. For other fields, health for instance, it's totally possible to never heard English in months/years. Japanese is well alive, and English is more a social marker than anything else. Most Japanese have very bad command of English if at all and can live their whole life never using it.

tokinonagare··on When is short tenure a red flag?
> It comes down to this: I hate being an employee. I want to work on my own projects on my own terms. This is the only way I can find work that is worth doing.

Sounds familiar: after mild insuccess in academia I found a permanent job but quitted after 4 months. Didn't help that I was a not very well paid consultant with long commutation time. The job was easy but boring. I don't get how many people are able to endure this.

tokinonagare··on AI translates Hokkien, an unwritten language, for the first time
I disagree. It makes a big difference: creating a writing system is not a simple as it seems and it's a task that require quite a lot of work, especially when there is huge dialectal variations in-between dialects. When a system already exist it means there is already a bunch of tools available such as dictionaries, input method[1], a corpus of texts even if small or medium, including text books. That can be used both for AI and teaching. This make the task way less daunting than doing everything from scratch.

In case of Taiwanese, there is a small but existing corpus of literature, as a 5 seconds Google/Wikipedia search shows[2][3].

[1] https://aclanthology.org/O16-1026.pdf

[2] https://zh.wikipedia.org/wiki/%E8%87%BA%E8%AA%9E%E7%99%BD%E8...

[3] https://zh.wikipedia.org/wiki/%E5%8F%B0%E6%96%87%E5%B0%8F%E8...

tokinonagare··on AI translates Hokkien, an unwritten language, for the first time
They won't. It will only help accelerate the move towards concentration to national languages (Standard Chinese in Taiwan and China, French in France, etc.) Why would anyone put the effort to learn a "small" language when it can be translated automatically for understanding?

The way to preserve a language is putting humans in the loop. Creating content in that language; interestingly more and more shows are produced in Taiwan in or using non-Mandarin languages as a political way to mark a difference with the big neighbour. And having government support, notably at school (at young age) by allowing partial or total teaching in the language to be preserved.

tokinonagare··on AI translates Hokkien, an unwritten language, for the first time
The AI community would be taken more seriously if they didn't over-exaggerate or even straight up lie about their achievements like it is done in that article title.

Hokkien, specifically the version mentioned in the article (Taiwanese Southern Min) has two main writing systems: Chinese characters and Latin letters based romanization. Heck, that's actually more than most languages in the world! The Tâi-lô[1] system, a slight modification of Pe̍h-ōe-jī was recently standardized by the government. On the other hand, Chinese characters can be used to write Taiwanese, not just Classical Chinese or Standard Chinese (Mandarin), save for a few words that are often written using Latin or bopomofo letters. It's true that a large number of speakers have not been taught either of those writing systems, but it does not mean they don't exist. I've already met someone which taught me few words in Hokkien (the infamous chia̍h-pá--bōe 食飽未) and write it in Chinese characters for me.

I'm working with various Sinitic languages, and Taiwanese is far from being the worst when it comes to available resources in written form (for my use cases dictionnaries).

[1] https://en.wikipedia.org/wiki/T%C3%A2i-u%C3%A2n_L%C3%B4-m%C3...

tokinonagare··on Covid Vaccine messenger RNA found in breast milk
It's mandatory to get published. "Science" is unquestionable and don't suffer contradiction.

Incidentally this is exactly what a few so-called conspiracy theories (aka hypothesis in the old world) predicated a few years ago.

tokinonagare··on The UX of Porn Tube Sites Are Designed for the Ultimate Money Shot
Yes, like xvideos.com, for those who want a URL. Their video buffering tech is pretty amazing.
tokinonagare··on Who cares about diversity?
I've seen it for a university position in Sweden. It has a self-criticism and cult like feel that I really don't like. Position in universities should be considered on scientific and educational grounds, not political.
tokinonagare··on Build your career on dirty work
>One way I try to spend time doing the dirty work is by over-estimating. It's taken many years to really learn this, but I now over estimate tasks by quite a lot.

I'm only 4 months into the industry and I felt like a slacker for figuring out this trick. Now I feel better about it thanks to your comment :)

tokinonagare··on Why was western printing superior to Asian printing?
And then we would need to use two very different spellings for American and British English. Be careful with what you wish.
tokinonagare··on Why was western printing superior to Asian printing?
> There’s also the fact that Hangul is phonetic

Hangul is not phonetic, it's a morphophonemic script. The Korean language has a lot of rules that applies at different boundaries, which make the pronunciation of words different from what is written. The best example of this is the "Korean wave" /han/ + /lyu/ => [hallyu], written has "hanlyu" (한류) which is conforming to the phonemic decomposition but certainly not to how the word is pronounced.

tokinonagare··on Use TouchID to Authenticate Sudo on macOS
No. I only use sudo in "sudo su" once in a while when I need a root shell (sometimes not even on the local machine), then does everything within that shell. I don't understand people bothering themselves with that command. I also find egregious tutorials in which each command is prefixed with sudo.
tokinonagare··on The culture of rejection in computer science publications
> Comments often show the reviewers never read the paper.

And when they do, it's not sure they understood it or even put the slightest towards understanding. I've a rejected paper where one of the comment was that the header of a table featuring 4 columns named N, V, ADJ, ADV was "hard to understand". The table was between two paragraphs each mentioning nouns, verbs, adjectives and adverbs, in a paper mostly about dictionary...

tokinonagare··on Ask HN: Anyone working 4 day week here, as an employee?
Awesome! I joined the industry temporarily but the 5 days week (with 3 hours commutation) is soul crushing. Four days makes it fine to cope with. I hope I can find something using your website so I can work more on my personal projects.
tokinonagare··on Silent crisis of soaring excess deaths in Britain is only tip of the iceberg
Thanks for trying to do good science. It's a shame that asking questions (the basis of our job) has become forbidden on some topics.

> So my guess with this paper is that they have data, they've thought of the vaccine as an issue. But they have to go with a more politically correct reason to get the data out in the open and then they can follow up later once their data is in the literature. > [article] So what is going on? Experts believe decisions taken by the Government in the earliest stages of the pandemic may now be coming back to bite.

If one reads between the lines, what is the largest "decision taken by the government" that could impact directly health? Vaccines. Of course, since media, big corporations and governments have close ties the truth will take time to get out, if it ever do.

tokinonagare··on Why the Learning Styles myth spreads
> Because the predominant ideas of pedagogy consist of what education academics would like to believe is true

This, times 10. I worked in a lab on education-related subject at a top university and it was wishful thinking all the way down: from Ministry bullshit like "education 5.0" and grants to whatever buzzword heavy proposal was thrown, to researchers ignoring obvious flaws in their dumb but cool things that are set to fail from the start, to arranging the reality of collected data (from weak protocol) and their interpretation, etc. It was full of people trying to make a name for themselves but very few solid research. Most of the papers were flawed, irrelevant or kinda stretched reality.

tokinonagare··on Show HN: Japanese Complete Book 1 Released
> the point of the book is to build symbolic recognition

So basically totally pointless and a waste of time and money because this ability comes really quickly when learning how to write kanji. This can be useful when the phonetic value is learnt with the corresponding graphical part, but since Heisig skips pronunciations the whole method doesn't not teach anything really useful.

tokinonagare··on Show HN: Japanese Complete Book 1 Released
How many kanji can you read? Even after finishing RTK it will still be zero if not using real learning materials, at which point having RTK is a waste of time anyway. That "method" is bad because it skip the main difficulty of learning kanji, which is memorising the ton of readings each character has and the actual words using them.

Feel free to waste your time on a snake oil method however, I'm not the one impacted.

tokinonagare··on Show HN: Japanese Complete Book 1 Released
> I like having kanji etymology explained

Then I recommend this dictionary, which has an edition focused on kanji too: https://www.outlier-linguistics.com/ The team who build that knows what it talks about, and it contains real etymologies, not fake ones based on the current graphical shapes of characters.

tokinonagare··on Show HN: Japanese Complete Book 1 Released
That's on purposes for few reasons ranging from legal, mistakes in the linguistic modelling, approximations in the data used for algorithmic comparison, lack of pagination in the indexes, etc.

Send me an email from the address found on the paper on the website if you want, so I can keep you notified of the real release.

tokinonagare··on Show HN: Japanese Complete Book 1 Released
The "kanji situated in an English context" thing is the most brain-dead weeby thing I've seen since Heisig's RTK, and that was a very high bar to pass. Same with the grammar mixing English and Japanese.
tokinonagare··on Show HN: Japanese Complete Book 1 Released
I'd say learn Hokkien first, which as a Min dialect has a more conservative phonology, which will help more than Mandarin when learning Japanese and Korean. The rest is a matter of test.

Also, I recommend using a website like the one I made (in alpha), which helps comparing the pronunciation of vocabulary of Sinitic origin: http://144.24.197.67/entry/e5b585e5-aba3-0000-0000-000000000...

tokinonagare··on Ask HN: What is something you built but never marketed?
That was a complete shitshow thanks to cultural differences, my personality and covid. I had interest in Japanese/Chinese while my lab only focussed on English. Basically whatever I proposed for 3 years was thrashed to the bin without any explanation, and when I found another lab more appreciative of my work it was already too late to switch. The only "guidance" I got during this time was to "make a quizz" (multiple choice) as if it was something revolutionary and doctoral level.

In the end, I had to back back in my country, where I was able to reconnect with someone that did his PhD on a very close topic (albeit not using computers), so I plan to finish my degree here using things I worked alone on for years. It's infinitely easier to work with people able and willing to understand ideas and who focus on quality instead of quantity. I'm indeed sitting on some stuff albeit it would need a lot more development for some to reach the light. I'm also finishing writing some papers but it's progressing slowly due to having a 9-5 job with 3 hours half commutation.

← PreviousPage 5 of 7Next →