Google is genuinely bad now. I switched to Ecosia which is just Bing with a simpler front end and they use their profits to plant trees. I don’t think Ecosia is particularly special though. Duck Duck Go, Bing whatever, they’re all better than Google.

Whenever I set up a new computer then search for something, I’m always surprised at first seeing the awful layout and quality of the search results before I realize that I haven’t changed the default search from Google. It’s awful now. Seriously, how are people using it?

My new favorite way to search is perplexity.ai. It’s an AI search tool that summarizes the loads of crap out there so you don’t need to read through the junk that people write. It provides sources, unlike using ChatGPT, which is incredibly valuable. All AIs make shit up, so having links to double check it is a must. Unlike Bing Chat, or whatever Microsoft calls it this week, you can ask follow up questions to home in on what you want.

I think you’ve confused vegan with strawman.

It’s easy enough to record using Zoom. Start a meeting with only yourself. I’ve used OBS too and it’s fine, but not necessary if you’re just looking to record something simple.

You’ll almost certainly want something to edit the videos. Rather than trying for a perfect take, if i flubbed something I would start over from a convenient place and edit out the bad part. I hated Adobe Premiere. Reaper isn’t really ideal for this stuff, but at least it isn’t Premiere. I dont remember if OBS had anything. LightWorks was OK.

You’ll want a good mic. Although lapel mics are ideal for your use, Bluetooth tends to suck. This webcam has an OK mic. https://www.logitech.com/en-us/products/webcams/c920s-pro-hd-webcam.960-001257.html

For unrelated reasons, I had an audio interface (Scarlett 4i4) and a cardiod mic (MXL 770). The Scarlett is very nice. I don’t love the 770, but it’s better tham a webcam. These are not critical but if you plan to do this a lot, these would make it sound noticably better, and maybe someone has a better mic recommendation, and again lapel would be perfect for lectures. Somewhere I got a baffle like this for $20 to $30. https://www.bhphotovideo.com/c/product/1116780-REG/primacoustic_p300_0101_voxguard_vu_vocal.html

It makes a difference and for a low price it’s nice to have if you aren’t using a lapel mic. I wouldn’t pay $100 for one.

You might need to update. Occasionally Google changes how YouTube works which requires NewPipe to change. You can get the latest version faster by adding the newpipe repository to fdroid. https://newpipe.net/FAQ/tutorials/install-add-fdroid-repo/

It sounds neat. Offline maps. If Google hadn’t banned it I would have never heard of it. It’s on F-droid too for those who don’t want to download and install an apk.

On my phone this can be set by going to settings, default apps, opening links, wait unusually long for the list to load, scroll to NewPipe, choose the links you want it to handle.

I’m not an MD but I would say having one now and then isn’t going to hurt you, but they are not at all a tool for weight loss even without sugar. Weight loss is primarily achieved by reducing calorie intake, and drinking a zero calorie drink doesn’t reduce calorie intake. The caffeine will not do enough to suppress appetite to be effective. Other people with more credentials say a similar thing: https://goaskalice.columbia.edu/answered-questions/can-energy-drinks-help-me-lose-weight.

Whenever I have issues with YouTube refusing to do things it used to do, I stop using it for a while and eventually they put it back. If you’re not willing to do that, I find that the NewPipe app is better than the native YouTube app. But be warned that occasionally Google makes changes that break New Pipe and you need to wait a couple days for the devs to catch up to the change.

I thought it was going to be something that used vision to monitor the roadways and dynamically make decisions about the lights, which would be a very different way of controlling traffic. It’s a predictive model of traffic that they used to adjust light timings, which is the same way that traffic has always been controlled, informed by a model. It’s nice that it helps, but they probably had no shortage of decent ways to model traffic that would have led to this result. My guess is no one looked into it and Google found an easy place to showcase an AI model.

Don’t forget to burn.

Yeah, it’s ironic that one of Google’s selling points was that Chrome didn’t have a lot of clutter. It’s even where the name comes from. Now it looks messy. It’s no Microsoft product yet, but it’s definitely one of the ways it used to be better.

By the reasoning given for why Mercury is the closest for each planet, the Sun is the closest object for each planet, on average, excluding satellites of the planets.

I used Firefox when it first came out. Google and Mozzila got into a hot race to make the best browser and they both did well. Somehow I ended up using Chrome a lot more even though I thought that by the time the race ended they were pretty even. Both were very fast and had great plugin libraries. Chrome looked nicer out of the box, but Firefox is highly customizable. Since the end of that race, Chrome has gotten worse and Firefox is about the same. I’ve switched back fully to Firefox, and the only thing I miss is the “Piss off publisher frames” plugin, that I haven’t found a replacement for. It’s a nice browser.

I just bought a new laptop for a family member. It wasn’t very expensive, but hardware now is generally amazing. It has Windows 11. My 12 year old laptop running Windows 7 is faster for most tasks, despite far inferior hardware. Plus search actually works in 7, it’s better organized, it doesn’t come with a ton of junk you need to disable or remove (good god the default start menu on 10 is a mess), and it doesn’t look like they designed the UI over the weekend. I kept waiting for the typical MS move of fixing the dumb crap they added, but with 11 it’s clear that they’re doubling down.

I don’t know that it’s technically harmonization, but sometimes when wolves howl at the same time, they will each choose a different pitch. Presumably it’s so the group sounds like it has as many individuals as possible.

There is something called the Power Delivery protocol for use with USB C. If your charger and device both support USB C PD, then everything is fine. The wattage of the charger will limit how fast it can charge, but the PD protocol sorts out the correct voltage. Most laptops, tablets and phones support PD now. You can also buy adapters to go from USB C PD to whatever connector your old laptops use (although watch out for current limits so you don’t fry the adapter). I use this charger for everything. https://nekteck.com/products/nekteck-65w-usb-c-charger-gan-ii-pd-3-0pps-fast-charger

On my PD laptop, it complains that it’s a “slow charger” because it wants 100 W or something absurd, but 65 is fine. On my old Lenovo, I have an adapter and the adapter itself doesn’t support high current, so I can’t reasonably use the laptop while charging, but I can charge it while it’s off or asleep. I usually use the OEM chargers for those, but for travel it’s nice to bring less. For everything else it’s perfect.

For those that don’t know, you can download the latest APK directly. It takes a while for new versions to make it to repositories and this lets you get it as soon as it’s ready.

https://newpipe.net/#download

Edit: someone pointed out that NewPipe hosts their own F-droid repository, so you can add that to F-droid to get faster updates more conveniently.

https://newpipe.net/FAQ/tutorials/install-add-fdroid-repo/

Seeing your comment I wondered how someone publishing in Nature could have possibly left out the use of statistics for prediction. That would be a wild oversight that only someone with little knowledge of the topic would make, and surely not something that the editors of Nature would miss. Upon clicking the link I see that they mentioned it in the very first sentence and apparently ignore it if someone happens to call the prediction model a machine learning model. Using statistical models for prediction has been used since the start of the field, and renaming things that have been used for decades as “machine learning” doesn’t suddenly make them not statistics.

Artificial neural networks are statistical models, with numerous statistical approaches associated with their use, development and interpretation.

“If parameters aren’t neatly interpretable then it’s bad statistics.”

Haha, keep going guys. You obviously know a lot about statistics.

“such as neatly interpretable parameters”

Hahaha, hahahahahaha.

Hahahahaha.