cwt137

I find it interesting the blog author tied PHP and Lua together. PHP uses a JIT from Lua. Is this related to the log issue?

It would be nice to have a link to an example YT video to see what exactly the AI is trying to locate on the screen.

This visualizations reminds me of the 3blue1brown videos.

There is a big difference between an Emperor penguin and a Galapagos penguin. They should of used a different measure.

Everyone talks about Websockets for pushing real time data to the browser. This article highlights some of its drawbacks. I use Server Sent Events (SSE) instead. A lot of the problems the author of the article faced are solved with SSE. Also, SSE scales way better than polling all the time.

I wonder if in the future the author will investigate Anycast. This is the big feature that competes with the big CDN.

I didn't see anything in the article about battery capacity or the density. This (with the other info in the article) would tell us if this is groundbreaking or not.

It is not true that Ollama doesn't use llama.cpp anymore. They built their own library, which is the default, but also really far from being feature complete. If a model is not supported by their library, they fall back to llama.cpp. For example, there is a group of people trying to get the new IBM models working with Ollama [1]. Their quick/short term solution is to bump the version of llama.cpp included with Ollama to a newer version that has support. And then at a later time, add support in Ollama's library.

1) https://github.com/ollama/ollama/issues/10557

This is the first in a series of posts exploring how to build a smartwatch in 2025!

Not always true! Your statement is only true when the running clock's speed is the same as time. Thus, regular time and the clock's time will never meet.

If the clock is running faster than regular time, it will at point catch up to regular time and thus be correct for a split second. If the clock is slower than regular time, regular time will catch up to the clock and the clock will be right for a split second.

A lot of things presented in this article reminded me of the same things that were in Joe Celko's “Trees and Hierarchies in SQL” book https://www.oreilly.com/library/view/joe-celkos-trees/978155...

This product seems similar to Ngrok and some other solutions. How does this product differentiate itself from everyone else?

Just last week I was looking into OptaPlanner [1] and MiniZinc [2]. OptaPlanner even has a real-time/continuous optimization mode. There are a whole bunch of other solutions, but these were the most interesting to me.

I wonder why Google didn't just go with an off the shelf solution and integrate it instead of building their own solution?

[1] https://www.optaplanner.org/ [2] https://www.minizinc.org/

I hate articles that talk negatively about GPS when they really mean navigation software. GPS allows you to determine where you are. Navigation software tells you where to go.

Two examples: GPS does not spy on you and tracks your every movement. But the navigation software might. GPS doesn’t tell you to go down the wrong road. But the navigation software might. Etc.

I was thinking the same thing, that they got their information from Jeff

A few years ago, I worked on an app where some of the state was kept in the session and some was on the url as GET params. We ran into an issue where some browsers would only accept urls that were less than 1024 chars long. Is that still an issue today?

I've been watching his videos too for the last 6mo or so. If you like them, you should watch his friend's videos on SerenityOS as well, Linus Groh [1]. He does a lot of stuff on LibJS and LibWeb libraries. I think he is the lead developer for those because every month when Andreas does the monthly summary video of changes in the OS, he defers to Linus to talk about the browser changes.

1) https://www.youtube.com/channel/UC84u7JhM9EIAYzyjdf6cBbA

> the biggest challenge would be the right of way and litigation

I totally agree. I live in SoCal and there are some infrastructure projects that took forever (ie creating the I105 and finish connecting the I210), or will probably never get completed (ie. finish connecting the I710 and California High Speed Rail) all because of right of way issues and lots of law suits.

The word you are looking for is “oligopoly”. This is when a small set of firms has a monopoly on something.

I think developers using the Arduino platform on the ESP32 are mostly just using one core. Developers using the Espressif's SDK are more likely to use both cores.

They have their machines also connected to another AS, so when their network doesn't/can't route, they can still get to their machines to fix stuff.

Maybe they used Postscript?

We need a lot of new embedded devices. Lots of them have hardcoded when the time changes and have no way of taking it off.

Raspberry Pi Foundation does the "minimum order of 1000" and passes the savings on to you. Try and buy all the individual components of a RPi Zero and see how much it costs. It is way more than $15.

P.S. The RPi Zero does not come with a display and keyboard like the computer in this article does. So, we are talking about two different types of computers.

You are right. Those who got "natural immunity" from Delta had no protection from Omicron. But how Omicron is structured, they get natural immunity from Delta and many other variants.

https://youtu.be/PYLbJ0H8zdc

Best non-rocket scientist explanation of GPS ever!

Most people are relying on arbitrary testing to figure out how many people are infected. I think this totally under reporting number of infections because the whole population is not tested on a regular basis. To get an accurate picture of people are infected, you have to look at whole populations.

One whole population group to study to get a better understanding of how bad this wave is is to look at the cruise line industry. Everyone has to get tested, and thus you are studying a whole population. According to a recent CDC report [1], between Nov 30 - Dec 14 of 2021, 162 cases were reported on cruise ships. Then the next two weeks 5013 cases were reported. This is roughly a 30x increase!

If you look at the NY Times infections chart[2], in the same time period, their 7 day average only went up like 3x. This is grossly under reporting of infections.

The sewage chart looks at a whole population and it is easy to see the increase in infection is more in the 30x scale than it is the 3x scale.

Be safe out there everybody. If you look at whole population groups, Covid is way more infectious than what a lot of news media outlets and their data would tell you.

1. https://www.cnbc.com/2021/12/30/us-cdc-says-people-should-av... 2. https://www.nytimes.com/interactive/2020/us/coronavirus-us-c...

This sewage report is about a whole population and it is clear to see infections many more times higher than what a lot of news agencies are reporting.

The control loop reminds me of how you would create a game ie. get inputs, update states, based on the current state of the environment and inputs, render (write output).

No wonder John Carmack got into rockets!

PDF doesn't encode something. PDF supports a few data formats. They call them "filters"; for example CCITT Fax, JBIG2, DCT, RLE, etc. Its up to the PDF creation program to pick the best data format for the image it would like to put into the PDF.