+1 for MXroute. I have unlimited domains with 25GB of storage for $30 every 3 years. So less than a dollar per month. Looks like they are still offering it. It's more than enough for email especially considering the Gmail account I used for 15 years was under 5GB.

I switched to them at the beginning of the year so about 9 months ago and have not had any issues.

The answer will always be no to any general question like this. It's not like the U.S. is a monolith. It's a culturally, racially, and geographically diverse collection of 350 million people.

Grab 10 random people and the chances of all of them agreeing on almost any subject is near zero.

Even on Linux where their drivers are supposed to be better, my 7900XTX has been crashing randomly for at least a month and it was only fixed in the latest 6.10.9 kernel release yesterday.

Similar story here. Asperger's runs on both of my parents' sides of the family. In addition Dad has ADHD and mom has BPD. I ended up with the Autism+ADHD combo with sprinkling of CPTSD on top. I don't even know where the neurological problems end and the psychological problems begin.

Have not talked to either in well over a decade.

Copyright isn't meant to help independent creators. At least not small ones. You have to pursue legal action against people to enforce it. Small creators do not have the money for that.

(unless any human creativity is involved then that has copyright )

This is the part people usually forget when they spout "it can't be copyrighted." If a human edits the output in some capacity then that is still copyrighted. It's not really the gotcha a lot of people seem to think it is.

Yeah at least Google will let you in after you solve 5 puzzles. It's shit but it's possible. With CloudFlare you are at the mercy of whatever hidden criteria they're using.

If you change your user agent from Firefox to Chrome for instance, CloudFlare will never let you through.

I remember Overwatch was one of the first DirectX 11 games to run really well when DXVK was new too.

Oh hey, I love your work on Plasma's HDR and color management. Glad to see you on Lemmy.

Phoronix is the ONLY website I disable uBlock Origin for.

I thought it was weird such an old piece of software had so much Rust in it. I noticed all the Rust-related things while Firefox Librewolf compiles but never looked into it further.

I have been playing through Elden Ring again with a friend using the seamless co-op mod and my friend on Windows gets (what we assume is) shader compilation stutter in every new area while my game has been smooth as butter.

There is a pretty big difference in terms of usability between Arch and everything else because of the rolling release model and the AUR. Lots of things you would have to manually install from a git repo or track down a PPA for can be installed like a normal package.

Only because people like you refuse to change. That's a self-fulfilling prophecy.

I don't understand why you want to be wrong so badly. What do you even gain from that? In this case using the wrong word isn't even just wrong it's objectively harmful but for some reason you want to keep doing it anyway? Why?

Why do you need to call him something if it's inaccurate? Just call him a child rapist and be done with it. Words have meanings and there is no reason to use them incorrectly.

As for who it's hurting: everyone. Pedophila literally has nothing to do with rape. It is a disorder. It's estimated it effects as much as 3% of the population. The vast majority of those people have never done anything wrong. They were simply born that way. The more people conflate the medical term with crimes the less likely pedophiles are going to seek help or treatment. Nobody wants to risk being outed as part of the most hated group of people on the planet.

If you want to actually help children instead of being angry the best course of action is to destigmatize the disorder so people can get counseling without fear of ruining their lives. Forcing people to bottle up their feelings/urges/etc. does not lead to good outcomes.

The majority of child molesters are not pedophiles. It is more about power dynamics / them being an easy target.

This breaks any site that uses CloudFlare's Turnstile for me. It will loop forever and never let me through if my user agent is set to Chrome.

Even if they fully render them into the video with absolutely no way for an extension to tell where it is something like Sponsorblock where people manually enter time codes could still get around it.

Usually being "lazy" is also the result of some kind of untreated medical issue or disorder like ADHD, autism, sleep apnea, etc. or any combination of those. Almost no healthy person would make the choice to sit around doing nothing for years at a time. Treating it like it's some kind of lack of will or moral failing only makes it worse.

Plasma actually has a UI for smart TVs if you weren't aware, although I have never used it myself so I'm not sure how good it is. https://plasma-bigscreen.org

Interesting. I've been rooted and running custom ROMs for a decade at this point and have never had an issue with banking apps.

I've had to explain this so many times. The majority of molesters are not pedophiles and the majority of pedophiles are not molesters. Pedophilia is a disorder, one that people suffer from at that. There is a (almost certainly) sizable group of pedophiles who have never actually done anything wrong but are caught in the crossfire of anger and therefore can't seek help/treatment.

I agree with the sentiment, when things get too popular every sub becomes more generic and filled with recycled or low effort content. But there's a happy medium. It would be nice if there were enough people that some more niche communities had activity.

Really? For me rspamd blocks at least 15 spam emails a day, usually from China or Russia. An additional 2-3 go to the junk folder, and some still slip through the cracks especially if it's coming from a gmail address.

But it could be as simple as it being because my email is publicly available (github, my website, etc.) so scrapers are picking it up.

That's a really clever login system.

It should all be opt in

Then you introduce self-selection bias and the data is worthless.

Aggregate data can be used to personally identify

You can't identify someone based on how they interact with a service. If you spend 5 minutes on one page and 2 minutes on another that could be anyone. Even if you for some reason personally knew someone's browsing habits it would be nearly impossible to pick them out in a sea of millions of data points.

I see you linked privacyguides.org in the thread as "alternatives", one of the services it recommends is Proton (Mail, Drive, etc.). Look at their privacy policy:

2.1 Visiting proton.me or protonvpn.com website: We employ a local installation of self-developed analytics tools. Analytics are anonymized whenever possible and stored locally (and not on the cloud). IP addresses are not retained and stored for such analytics.

When you use our native applications, we (or the mobile app platform providers) may collect certain information. We may use mobile analytics software (e.g. fabric.io) app statistics and crash reporting, Play Store app statistics, App Store app statistics, or self-hosted Sentry crash reporting to send crash information to our developers in order to rapidly fix bugs.

Or how about addy.io that privacyguides recommends for email forwarding? From their privacy policy:

We use a self-hosted instance of Umami, an open-source, privacy-focused and lightweight option for website analytics. All the site measurement is carried out absolutely anonymously.

ALL online services collect this kind of data. Even the privacy-focused ones. There is nothing nefarious about it.

Like the comment I replied to already explained, this information is necessary to make informed development decisions. If you don't know who is using what feature you might be wasting resources on something barely anyone uses while neglecting something everyone needs.

You also need some of that data for security purposes. You can't implement rate limiting or prevent abuse if you can't log and track how your services are being interacted with.

And this is aggregate data. I can promise you not a single person cares about what any individual user is doing (assuming it's not illegal)

Yeah as someone who has worked in web development for over 20 years everything in here is completely standard. Almost every major website in existence collects this kind of analytical data.

The hardware survey doesn't ask every single user, it just gets a sample. So it probably just happened to hit a few more Windows 7 people this month.

This happens to me constantly. Just the other day I asked some friends for something and then they sent the literal exact opposite of that thing. Pretend I asked for blue with red stripes they gave me green with yellow polka dots. And it wasn't just one person it was three separate people who all decided that made sense for some reason.

I was extremely specific too, even more than usual because I know people constantly misinterpret me. I made extra sure to not use any language with vague meanings and it still happened anyway. It's like we live in alternate realities where words have completely different meanings.

It makes me not want to talk to people at all.

Again, even an exact copy is not stealing. It's copyright infringement. Theft is a different crime.

But paraphrasing is not copyright infringement either. It's no different than Wikipedia having a synopsis for every single episode of a TV series. Telling someone about what a work contains for informational purposes is perfectly fine.

Sorry, I misinterpreted what you meant. You said "any AI models" so I thought you were talking about the model itself should somehow know where the data came from. Obviously the companies training the models can catalog their data sources.

But besides that, if you work on AI you should know better than anyone that removing training data is counter to the goal of fixing overfitting. You need more data to make the model more generalized. All you'd be doing is making it more likely to reproduce existing material because it has less to work off of. That's worse for everyone.

What you're asking for is literally impossible.

A neural network is basically nothing more than a set of weights. If one word makes a weight go up by 0.0001 and then another word makes it go down by 0.0001, and you do that billions of times for billions of weights, how do you determine what in the data created those weights? Every single thing that's in the training data had some kind of effect on everything else.

It's like combining billions of buckets of water together in a pool and then taking out 1 cup from that and trying to figure out which buckets contributed to that cup. It doesn't make any sense.

If the model isn't overfitted it's also not even copying. By their nature LLMs are transformative which is the whole point of fair use.

For me on Arch, Flatpaks are kinda useless. I can maybe see the appeal for other distros but Arch already has up-to-date versions of everything and anything that's missing from the main repos is in the AUR.

I also don't like how it's a separate package manager, they take up more space, and to run things from the CLI it's flatpak run com.website.Something instead of just something. It's super cumbersome compared to using normal packages.

Same here. Switched to Arch in 2015 so I am also coming up on the 9 year mark. I have had very few issues, and the ones I have had were usually my fault for doing something stupid. I used Windows, OS X, and Ubuntu previously and compared to those Arch is a dream. Hence why I've stuck with it for so long now.

https://librewolf.net/

Obviously you've never used Arch btw. We live for the sudo pacman -Syu.

I pretty much never reboot the Pi. It currently has over 18 months of uptime on it. My NAS on the other hand I probably restart for one reason or another maybe once every 6 months. So yeah I'd say I reboot it minimum 3x more often.

Plus a reboot takes much longer on my NAS than on the Pi. The server board is slow to start, the SAS cards are slow to start, and unRAID is slow to start. Then I need to manually enter the password for disk encryption. Then wait for the array to start up. Then wait a bit more for the docker containers to start. Add all of that up and even the absolute fastest reboot is like 10 minutes while the Pi probably takes 30 seconds.

And what if I want to swap hard drives? Now it's down for an hour. I guess I could wait until 3am to do all my upgrades so everyone is asleep, but I'd rather not. I suppose if it were just for myself it would matter a lot less. But again, it's only $15 to not have to think about it at all.