Typewar

They already had a controversy some time (years?) ago, saying stuff like "Bitcoin is all we need", "everything other than Bitcoin is a shitcoin", "we don't want to associate with the criminal record Monero carries"... something like that.

What you can do is to use non-kyc exchanges, send XMR in, exchange to Bitcoin, and withdraw directly to Proton. Proton won't be all hurr durr about dirty Bitcoin.

There are 2 instances that use anti bot features that I could find. That is https://search.canine.tools/ and https://search.fredix.xyz/

Personally, the fastest and most reliable in my opinion are https://priv.au/ and https://searx.tiekoetter.com/ I used to use Tiekoetter's instance for nearly a year before I hosted my own. Otherwise I would recommend checking here: https://searx.space/

With that said, I still didn't answer your question, because I am hosting my own instance, but I would not recommend joining unless you really want to. I'm going to move it to central Europe, it's currently hosted in Romania.

Damn, is it actually Eivith? Thanks for all the support help in the Matrix chat

Damn, I too went to this same route of Google -> duckduckgo -> Searx. While it is true searx is unreliable at times, if you enable more engines and find an instance that has anti botting features, it should be pretty solid

My friend is a crazy heavy anime watcher. One time he decided to watch an anime without subtitles, and he understood soo much... Pretty much got it all. Didn't study Japanese, just casually having nearly 300 days (7200 hours) of watch time according to MyAnimeList.

It's cool and all, but there are MUCH better and faster ways to learn Japanese than spending 2 years non-stop watching anime to reach his kind of level

I made a website to practice reading my wristwatch: https://aadniz.github.io/niwa-practicer/ (works best on PC, and I'm well aware of many issues)

Since depth is important to recognizing the odd and even, quickly mapping them to the number, I made it "fake" 3D, tracing each layer in krita.

There was no deep motivation for this other than refreshing myself a bit of React from University. With my neverending list of project plans, I felt like this one was a good choice for that. Here is the source code: https://github.com/Aadniz/niwa-practicer

I use Reddit to investigate niche topics. Sometimes I want people's say instead of these blogs showing top 10 house renting service.

The pure scale of Reddit users makes it more attractive in this regards right now

Having 2 loads gives the illusion that it's fast, aka. not waiting staring at something not doing anything for too long.

From a business perspective, isn't it best to just yeet most stuff to the front end to deal with?

First there was moral reasoning and feelings, then there were religion, and now there is laws. I think it works quite well for countries with a healthy government.

So many problems with this. I assume they're thinking of using an LLM since it would need to read language. It would need to adapt to our ever-moving Internet culture, knowing what intent is meant.

How well does it know irony? Slang? Taboo topics? Fresh new gen-z TikTok language?

"He should step on lego.. in a video game..", no way it will work at this early stage of AI.

I think AI could be useful to help actual human moderators to THEN determine if the activity is bad or not. But that's only doing some of the work.

I think manual reports from the users goes a long way on its own.

I remember there was one fact I was really beating my head on; A dishwasher should always have some food or other gunk on the dishes before starting the machine, otherwise the detergent will attack the coloring on the dishes instead.

How has no company solved this problem? It makes no sense. Many people do wash their kitchenware so it doesn't stink up the entire dishwasher if it has been sitting for a while... idk.

I would be happy to hear if anyone can help confirm or dismiss this.

I would like to see your raw words and feelings instead though