The growth of reddit itself could account for a large portion of the growth in job postings on /r/haskell.
987 karma · joined November 16, 2018
I mean if you're fine with blocking I/O then obviously you don't need async, but on the other hand having non-blocking I/O is the whole point of async ^^
[1] https://www.zdnet.com/article/twitter-seeing-record-user-eng...
Why do you exclusively point out "for profit" open source projects here? What about basically any commercial software that leeches on open source software without donating anything?
They want you to sit in an open office, which is even worse imo.
[1] https://adamtooze.substack.com/p/chartbook-200-something-bro...
> Also, I am sorry to be rude, but there is another reason that it is maybe not great to be the Bank of Startups, which is that nobody on Earth is more of a herd animal than Silicon Valley venture capitalists. What you want, as a bank, is a certain amount of diversity among your depositors. If some depositors get spooked and take their money out, and other depositors evaluate your balance sheet and decide things are fine and keep their money in, and lots more depositors keep their money in because they simply don’t pay attention to banking news, then you have a shot at muddling through your problems.
But if all of your depositors are startups with the same handful of venture capitalists on their boards, and all those venture capitalists are competing with each other to Add Value and Be Influencers and Do The Current Thing by calling all their portfolio companies to say “hey, did you hear, everyone’s taking money out of Silicon Valley Bank, you should too,” then all of your depositors will take their money out at the same time. In fact, Bloomberg reported yesterday:
Unease is spreading across the financial world as concerns about the stability of Silicon Valley Bank prompt prominent venture capitalists including Peter Thiel’s Founders Fund to advise startups to withdraw their money. …
Founders Fund asked its portfolio companies to move their money out of SVB, according to a person familiar with the matter who asked not to be identified discussing private information. Coatue Management, Union Square Ventures and Founder Collective also advised startups to pull cash, people with knowledge of the matter said. Canaan, another major VC firm, told firms it invested in to remove funds on an as-needed basis, according to another person.
SVB Financial Group Chief Executive Officer Greg Becker held a conference call on Thursday advising clients of SVB-owned Silicon Valley Bank to “stay calm” amid concern about the bank’s financial position, according to a person familiar with the matter.
Becker held the roughly 10-minute call with investors at about 11:30 a.m. San Francisco time. He asked the bank’s clients, including venture capital investors, to support the bank the way it has supported its customers over the past 40 years, the person said.
Nah, man, you don’t get to be a successful venture capitalist by taking a long view or investing in relationships or being contrarian. I’m sorry, I’m sorry, this is unfair. Of course they were right — Silicon Valley Bank did collapse, and if you got your money out early that was good for you — but that is largely self-fulfilling; if all the VCs hadn’t decided all at once to pull their money, SVB probably would not have collapsed.[6]
[1] https://www.bloomberg.com/opinion/articles/2023-03-10/startu...
The only areas in CS that come close to that in terms of difficulty are discrete math/algorithms related ones and the CS chairs that are involved in such research also tend more towards the applied side of those topics. Really theoretical discrete math stuff resides in math departments most of the time.
Still, having also taken some discrete math classes, in terms of difficulty those somewhat paled in comparison to differential geometry/topology/abstract algebra kind of stuff. I can't even imagine how difficult it must be to be doing research in those areas.
In my (limited) experience it seems to perform even better for typed languages (for example Kotlin/Java/Swift) compared to Python. The Python code it provided often had subtle type issues when working with dates. While the Kotlin date-related code it provided was more accurate and correct in terms of types. Which makes sense since the additional type information likely leads to a much better "internal model of how Kotlin works"
I think another possibility here is that they might have used an execution environment to check whether the code the model came up with actually compiles and used that as additional input during training. Some sort of execution environment seems to me to also be a possible explanation for how they managed the model to emulate a terminal so well.
Almost all of the content that the new book covers, with the exception of the third part on deep learning, is about theory that was almost exclusively invented before 2012. Classical ML (non deep learning) is actually very rigorous compared to modern ML. There exist good theorems (statistical learning theory) for most of the classical models I'm aware of.
Is that really true? That's not my impression at all (though to be fair I haven't been keeping up with current research as much as I used to). My understanding is that there is still hardly any knowledge on what deep learning models (and large language models in particular) actually learn. The loss surfaces are opaque, one still doesn't know why local minima reached by gradient descent tend to generalize fairly well for these models. The latent representations that generative language models learn is, with the exception of the occasional paper that finds some superficial correlations, hardly investigated at all and overall extremely poorly understood.
Very much interested in any references that contradict that sentiment.
"Of the options provided, the best option would be to send another email to the 30 students who did not respond to the initial questionnaire, encouraging them to complete it. This would allow the university to increase the response rate and collect more comprehensive data. Starting over with a new sample of 120 students would be time-consuming and may not necessarily produce better results, since there is no guarantee that the new sample would have a higher response rate. Using the 90 questionnaires that were submitted as the final sample would not provide a representative sample of the entire freshman class, and choosing 30 more students at random and emailing them the questionnaire would not address the low response rate from the initial sample."
Just as a reminder, after we had spectacular results on ImageNet, highly respected AI researchers were predicting the end of the radiologist occupation. Turns out that even when some state-of-the-art CV classification algorithm is used on any kind of scan, you still need a radiologist to look at the image basically in the same way as before.
If you write large-scale applications with the help of a system like ChatGPT you will still need to create accurate test coverage and an understanding of the problem that is essentially equivalent to that of the people writing the code themselves. Whether all of this would in the end actually lead to large enough productivity increases depends on how error-prone the AI generated code will be and given that it takes a lot more time to dig into unfamiliar codebases than those you've written yourself, I think it's anything but obvious that this will have a huge overall impact on the industry. But obviously I might be biased here, since I have a stake in the game.
My original post was aimed as a response to a post implicitly claiming that it was only a certain minority in Germany that sympathized with the regime. Whether that sympathy was due to successful propaganda or not isn't really the point when refuting that original statement, as is the way the Nazis came to power. From what I've read (and you also seem to agree with this) the majority of Germans supported the nazi regime while in power.
It would be interesting if you had any numbers for the number of displaced or killed German citizens. There were about 500 thousand Jews in Germany before 1933, most of which probably either left Germany or were murdered. People from the political opposition who had deeply seated beliefs that wouldn't have been subject to propaganda and who just continued to live in Germany in fear, were probably the largest percentage of those that didn't support the Nazis, but I don't have good numbers here. I would assume those would constitute at least 25% and at most something like 40% of the population, but those numbers are speculative. It does seem reasonable to assume that a majority of those who voted for the other conservative parties in the last election would have come to support the Nazis and those plus the people who actually voted for the Nazis already constituted a majority of the German population.
What I claimed was that the Nazis were popular after having gained power. In fact Nazi ideology even remained popular after the Nazis were defeated. Tony Judt has survey results [1] in his book 'Postwar', e.g. a majority of Germans were of the opinion that Nazism was a good idea but poorly applied. 37% in the US occupied zone said that the extermination of the Jews was warranted.
[1] Richard J. Evans - The Third Reich in Power