Recent submissions

Comment · 6 months ago
Comment · 7 months ago
Engineers, creatives, and other product focused roles tend to be well suited for these types of flexible schedules. Customer or sales focused less so, especially if the team or customers all have a similar preferred working hour. If everyone has flexible schedules, it'd be easier to ask for one. Though, there's the opposite of the spectrum where I've yet to convince anyone to try the 40 hour work weekend...
Comment · 2 years ago
The article does a decent job of explaining of how Chinese characters work, but it falls short of explaining why.

The reason why Chinese continues to use a logographic writing system is due to both tradition and practicality. English has grossly grouped together Chinese as one unified language, when in actuality it is not. In fact, many "dialects" are mutually unintelligible--one speaker cannot understand another speaker. If all of China switched to using a phoenetic writing system, everyone would write everything differently. It'd be very difficult--impossible at some points--to read and write materials from other "dialects". However, with a logographic approach, everyone can understand that the character 工 means "work" even if I pronounce it like [wirk] and someone else pronounces it like [wak], for example. It's one of the reasons why subtitles are so prevalent in Chinese media. Obviously, this problem can be eliminated by eliminating individual "dialects", which is sort of promoted through the adoption of Mandarin Chinese. Many Chinese media is also dubbed in the standard dialect so that actors with regional dialects can be understood.

As for Chinese characters in other languages, Japanese becomes a lot easier to read with the addition of Chinese characters. Kanji allows sentences to be shorter, less ambiguious, and easier to parse. Unlike Chinese, each character is not just a single syllable, and there are many homonyms in Japanese because there's a smaller set of sounds.

https://history.stackexchange.com/questions/46658/did-china-...

Comment · 3 years ago
I'm not aware of any experiments, formal or informal that tries to seek if negative content is promoted more. The article states one experiment with bots saw how quickly the feed would become saturated with majority negative content. But what about the opposite? How quickly does positive content fill the feed if a bot were to engage with inspiring or educational videos? I suspect it'll be slower, though maybe not because of the algorithm itself but because video makers have less incentives to make positive videos, and so there's simply less "good" content to recommend.
Comment · 3 years ago
This touches a little on philosophical difference, but I have found that one major difference is the quantification of uncertainty. Bayesian models, being based naturally around modeling a probability distribution, easily enables the researcher to make claims such as "I am 95% certain the mean lies between x and y, given the data." On the other hand, neural networks, decision trees, and other models more associated with ML than Bayesian statistics do not have this capability built in. (Though there are variations and techniques to build confidence intervals and such with them.)
Comment · 3 years ago
Comment · 3 years ago
I think that it's interesting that it's BS that's the focus. Are there many fields where you can thrive on non-BS alone? You could be a great sports player, but without some amount of BS you might not get noticed or believed.
Comment · 3 years ago
I find it pretty shocking that Teman can advertise its law-circumventing tech so blatantly. No one really objects from cameras in hallways, but jumping from using them as more of an auditing and security tool to actively monitoring and assuming a policing rule becomes problematic.

Fortunately, the industry is still growing, so there's still time to prevent this from becoming the norm in every space, not just those who want to evict their residents.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton