HN user
FWIW, you're comparing a training-specialized chip to an inference-specialized chip. It'd be more apples to apples to compare to TPU v4 lite, but I can't find that chip's details anywhere beyond some mentions in the TPU v4 paper: https://arxiv.org/abs/2304.01433
Very cool! Here's a tutorial from Sebastian Lague on implementing something similar in Unity, which may have served as inspiration for this project:
Can you share more about what you learned about moderation?
I want to believe this but I can't find a source, can you share where you've read this?
The bad thing is is that it’s too late.
Too late for what exactly? Majority of climate scientists agree the situation is dire, but there are many actions that can be taken to reduce the short-term and long-term impact, via both adaptation and reducing carbon emissions.
We're up against massive loss of biodiversity, loss of habitable space on the planet, and an enormous amount of human suffering and loss of human life due to the current and future effects of climate change. A smaller population will not solve these problems alone, but it will buy us more time to solve them.
FWIW, I'm not trying to promote anything extreme like population control policies, just pointing out that the current trend of population leveling off is generally a good thing.
Until per capita carbon emissions are below zero, falling population is a good thing.
I feel like this happened to Gitlab / GitHub as well. Gitlab was gaining a ton of popularity and momentum with their free private repositories and built-in CI testing infrastructure. Then GitHub swooped in and offered all the same things, taking the wind out of Gitlab's sails.
Do you expect these optimizations making their way into std::sort eventually?
Thanks Isaac. How do I reach that page? I expected it on one of the informational pages such as the home page or benchmark programs page, but had no luck.
I can't find the documentation for it, but you can see here that they measure the size of the source file after gzip compression, which reduces advantage of code-golf solutions:
https://salsa.debian.org/benchmarksgame-team/benchmarksgame/...
Not exactly what you're looking for, but here are some 2D plots of code size vs. execution time with geometric means of fastest entries and smallest code size entries of each language:
https://twitter.com/ChapelLanguage/status/152442889069266944...
Strict immigration policies during the previous presidency combined with the pandemic have led to record low immigration. Census projections generally expected immigration levels to remain constant. However, immigration plummeted from a net gain of ~1000k/year in 2016 to ~250k/year in 2020[1].
[1] https://www.brookings.edu/research/u-s-population-growth-has...
Tangentially, I am amazed at the data in your source about how fast population growth is leveling off in the US. In 2020, the census projected US population to continue growing and cross 400 million by 2058 [1]. This projection accounts for the observed trends in falling fertility rate and assumes relatively constant rate of immigration:
By 2030, immigration is projected to become the primary driver of population growth: more people are projected to be added to the population through net international migration than from natural increase. The projected shift to net international immigration as the primary driver of population growth is the result of falling fertility rates and the rising number of deaths in an aging population, not because of a projected increase in international migration.
I guess the pandemic broke a lot of those assumptions. We will see if they hold in the long run though.
[1] https://www.census.gov/content/dam/Census/library/publicatio...
But the population of the hottest parts could easily move to cooler parts of the world, most of which are sparsely populated.
I would not describe relocating billions of people hundreds to thousands of miles north, often across country borders, as easy.
This article contains some stories of a few climate refugees of today:
https://www.nytimes.com/interactive/2020/07/23/magazine/clim...
The post focuses on countering the strawman-like argument that resource shortages are the primary threat with overpopulation. However, it is overwhelmingly clear that climate change is the biggest threat related to overpopulation, unless we are able to get carbon emissions per capita below 0 globally.
Definitely. With the right training set, video game assets too. Imagine being able to generate N variations of any asset in any video game style...
Where did you find this feature? As a beta user, I have a hard time finding documentation for the features in Kagi. Ironically, searching about Kagi in Kagi is not helpful.
Any idea why this happens? I don't understand why WiFi vs. mobile data would impact how they want to personalize the results.
I've been relatively happy with Kagi so far. Happy to post some search results if you want to propose some queries to compare against Google.
I have been trying Kagi for a few months and have found it better than Google in most cases, though not a silver bullet.
Here are the queries from the original article in Kagi:
how to fix a flat tire
https://www.wikihow.com/Fix-a-Flat-Tire
Also, it predicts a video is best to answer this question and gives 3 YouTube thumbnails listed at the top of the results. They all seem generally helpful based on the title.
how are pearls made?
https://www.thepearlsource.com/blog/facts-about-pearls/how-a...
Also see the global warming tracker: https://www.climatechangetracker.org/
I thought this was a nice way to present the data and serves as a great reference to look up when trying to reason about the different solutions that aim to tackle climate change.
Our ever increasing population and consumption will cause the price of energy and materials to increase as the low-hanging fruit is picked.
Not only are we picking the low-hanging fruit in energy production, but we're picking it at a temporarily discounted cost (with carbon emissions not being priced in) at the expense of our future selves and future generations.
Knowledge growth is exponential. Rather than focusing on specific topics/tools, focus on trying to gain some form of knowledge, insight, or skill from whatever task you are working on.
For example, if you are working on implementing a regex expression for your job, spend the additional time and effort (if you can spare it) to learn a little more about regular expressions and/or the tools associated with them. These small incremental improvements to your knowledge base over time will have a significant impact on your overall trajectory.
Weather Spark used to be my favorite weather website. It had an interactive plot displaying a bunch weather data, historical, forecasted, historical min/max, and historical standard deviations. This was all in a side-by-side pane with a map view that showed radar data and temperature measurements at individual stations, and you could click different stations to see all of their historical data. It was the best interface to view forecasts and get a sense for how the recent weather stacks up against past years, IMO.
I have not seen any other weather site that comes close to it. The WeatherSpark team made a post to explain why they had to drop the old interface, but I forget the reasons now. The new interface is nothing like the old one -- static plots with buttons to switch to past dates / years, similar to Wunderground.
This is succinctly captured by a phrase from the video game industry: "Show locked doors before you show a key"
You may even split up (1) into omicron-recovered and other-variants-recovered, since recovering from omicron strain seems to give a stronger resistance against reinfection for both omicron and delta, whereas recovering from delta offers little reinfection resistance to omicron[1].
[1] https://www.nytimes.com/2021/12/28/health/covid-omicron-anti...
I feel similarly to the other commenters that xkcd references tend to get overused on HN, but it seemed too perfect to not reference in this case.
Obligatory related xkcd: https://xkcd.com/927/
This talk was from April 2021. 2009 is the year he got his PhD[2].