Skull giver

Revanced has that, as well as a load of other features. Can be used without root access these days.

They’re teachers, they already have a full time job, they don’t need a side job of syadminning their own laptops.

Upvoted for actually posting an unpopular opinion. I disagree with your reasoning, but at least your opinion has some arguments backing it.

Fuck Github for refusing to set up some kind of IPv6 proxy is all I’ll say on the topic.

Votes federate, but only for communities followed. I won’t see your votes in a community that I don’t follow, but I can see when you upvoted or downvoted what post in the community.

A scraper could simply follow every community on a Lemmy server and, barring Lemmy performance issues, will receive all comments and votes.

Just a quick and dirty SQL query of which votes of yours are in my server’s database:

select comment_like.score as score,comment_like.published as when, person.actor_id as who, comment.ap_id as what from comment_like join person on person.id = comment_like.person_id join comment on comment.id = comment_like.comment_id where person.actor_id = 'https://lemmy.ml/u/GolfNovemberUniform' order by comment_like.published desc; 

The same info is also available for posts, of course, I just didn’t want to bother making the query any longer.

Server admins/mods on Lemmy also have a button to see who upvoted and downvoted each post. This is just the inverse of that.

deleted by creator

Seems to me like they just want to greenwash (open wash?) their company by making it work with others. Saves a hell of a lot of trouble with legislators when it comes to stuff like the Digital Markets Act.

I’m sure they’re selling ads to their users, but I think that’s about it. There’s no money to be made analysing random internet accounts if you can’t show them ads when you’re big enough for EU regulators to care.

Why would they need threads for that? A whole bunch of companies are already doing that without running actual social media services.

They can analyse your likes and you wouldn’t even know it. All they need to do is follow the same servers you do here on Lemmy. On Mastodon they can set up a basic puppet domain, follow every user they can find, and then your Mastodon server will deliver your posts, likes, and re-tweet for them, no scraping or interaction necessary.

If you’re trying not to get analysed, the Fediverse is not for you. It’s simply not designed for privacy.

A factor in favour of jet fuel is that as the plane burns fuel if becomes lighter, thus consuming less fuel. Batteries stay the same weight. The difference between a full plane and an empty plane can be 18 metric tonnes. Super cheap operators tend to carry only a small extra margin of fuel over the amount technically necessary to make a trip, because it makes a real difference.

That means the energy density you need in this comparison isn’t really linear. If you’re doing Taylor Swift flights to the couch and back, you can save a lot of weight by having a minimal amount of fuel in the tank, but with an electric plane you’ll always have to have the full battery in case you need to go somewhere further away.

The difference between servers and countries is that servers aren’t countries and countries aren’t servers.

Servers aren’t a democracy. Well, most of them anyway.

The difference between a violent, oppressive authoritarian regime and a fee Fediverse server is that you’re free to join other servers. Multiple at the same time, even! You can just leave, no passports, no refugee status, no paperwork.

You can even set up your personal little server where you decide on the rules. A server for you and your friends can cost as little as ten dollars per month. Try that in any real country and you’d be considered an insurrectionist or a traitor, do it online and it’s just everyday business.

The unfortunate reality of most “everybody is welcome” servers is that hey generally attract a lot of people who have been banned elsewhere. Some for stupid reasons (like calling any criticism of the CCP “orientalism”), some for very valid reasons. You need some form of moderation, or your server is going to be a cesspool. Some server admins preemptively decide to block servers that don’t have moderation that’s up to their standards, others wait for abuse to spread to their server.

If you still use MBR, and Windows has an update to its bootloader, yeah.

I don’t even know if Windows 11 still supports MBR, though. Maybe it’ll happen if your firmware is broken and always boots from the fallback bootloader instead of the normal boot entry? But in that case Windows is right and the firmware needs an update.

My GPU is a 10 series, actually. Same generation as my laptop’s GPU, though that’s technically a Quadro I think.

I’ve been using Wayland on Ubuntu 24.04 with Nvidia’s 555 driver and I have yet to see something seriously break because of it. 545 was unstable as hell, but 555 has been running perfectly. I’ve even got a working external display on my laptop with the 555 drivers, something I’d previously given up on with Wayland.

Plasma is much further along (having stuff like HDR and VRR working for instance) but things have gotten a lot better quickly with Nvidia+Wayland.

Even if they’re marketing BS, some decent mainline support for Qualcomm chips is always welcome. It’ll help projects like postmarketOS and mobile Linux distros massively to have more usable Qualcomm code.

Damn, they really overfit their music models. With image generation and text prediction it’s very hard to prove a direct connection, but with four or five of those songs it’s unmistakable that the original songs were used to generate the music output

I wonder what the effect will be of fixing the models’ overfitting. I’m guessing it’ll generate worse music, or they would’ve done so already.

Quite sad that it took the music industry to notice before any lawsuits with a chance of succeeding got off the ground.

Then, what prevents whosoever, to copy that file through cloning the complete disk?

Nothing. At most, you can have a hardware encrypted drive that won’t permit access to the encrypted data without a password, but the file will remain available after unlocking that. Plus, dedicated people (law enforcement, data recovery specialists) may be able to get access to the flash chip itself unless you buy one that self destruct on any tampering attempts (and even those have flaws).

You cannot prevent copying of data if that data is readable at disk level. At most, you can make the data useless by padding a layer of encryption (as well-encrypted data may as well be random data without the key material). That’s why everyone is going for encryption: encrypted files may as well be inaccessible to anyone who doesn’t know the passphrase. There’s no sense in copying a file which you cannot possibly read any bytes from.

If the key is gone (i.e. the real key is a password protected file that gets overwritten so even the password doesn’t work anymore), the file becomes irretrievable. This is sometimes called “cryptographic erase” in the context of disks. There are variations of this, for instance, storing the key in the computer’s processor (fTPM) behind a password, and clearing that key out. There’s no way to get the key out of the fTPM so it cannot be backed up. Even if someone were to guess your password, the file will forever remain locked. Or at least until someone manages to break all cryptography, but even quantum computers don’t know how to do that part yet.

If you’re willing to go deep, you could reprogram the firmware on your SSD/HDD to refuse reading the file. A few years back, someone made a proof of concept firmware that detected disk imaging attempts (because all blocks on the disk were read in order) and had the firmware return garbage while secretly wiping the disk when this detection triggered. You could, in theory, write firmware that refuses to read that block of data. However, if whoever you’re hiding this file from know about that, they can take out the platter/memory chips and dump them directly, bypassing your firmware entirely.

“undoing the protection should include filling in a password” That sounds like an encrypted drive. There are USB keys that’ll require software to enter an encryption password before you can do anything (including deleting the contents).

If you’re on Windows, try Bitlocker or Veracrypt. You can create hard disk images that can be mounted temporarily with a password.

Same can also be done on other operating systems, though I don’t know what tools yours come with.

In a pinch, you can just create a password protected 7zip archive, though viewing and editing those files usually involves a temporary copy.

There’s no way to prevent a file that’s loaded in memory from making it back to the disk. The best you can do is also encrypt the system drive so only people who know the encryption password can boot the computer that’s accessing these files.

The edits are what makes it made with AI. The original work obviously isn’t.

If you’re in-painting areas of an image with generative AI (“context aware” fill), you’ve used AI to create an image.

People are coming up with rather arbitrary distinctions between what is and isn’t AI. Midjourney’s output is clearly AI, and a drawing obviously isn’t, but neither is very post-worthy. Things quickly get muddy when you start editing.

The people upset over this have been using AI for years and nobody cared. Now photographers are at risk of being replaced by an advanced version of the context aware fill they’ve been using themselves. This puts them in the difficult spot of wanting not to be replaced by AI (obviously) but also not wanting to have their AI use be detectable.

The debate isn’t new; photo editors had this problem years ago when computers started replacing manual editing, artists had this problem when computer aided drawing (drawing tablets and such) started becoming affordable, and this is just the next step of the process.

Personally, I would love it if this feature would also be extended to “manual” editing. Add a nice little “this image has been altered” marker on any edited photographs, and call out any filters used to beautify selfies while we’re at it.

I don’t think the problem is that AI edited images are being marked, the problem AI that AI generated pictures and manually edited pictures aren’t.

Looks like people are finally finding out they’ve been using AI all along.

Seems to me that employing the use of AI to alter an image should be labeled as “made with AI”. It’s not made by AI, AI was merely one of the tools used.

If you don’t like admitting you used AI, just strip the metadata, I guess. This feels like something you should be able to turn off in your editor’s settings, but I guess Adobe hasn’t implemented that.

This comment was made with AI, as my phone’s keyboard uses AI to automatically complete words, in a process strikingly similar to how ChatGPT works.

Now that single window mode is a thing across all platforms, it’s not as bad anymore.

I still don’t know how to draw a circle, though. I think I may need to turn a selection into a path? I just use Photopea for this stuff these days.

Gimp is what you use for image editing if you’re poor or Stallman-like. It’s good at being scripted for programmers that want to automate their image processing workflow.

Krita is what you use for drawing stuff if you don’t want to spend money on software or are Stallman-like.

Photoshop is what you use if you do photo editing or digital art for a living. Paid, proprietary options exist if you don’t do photo editing.

There’s one tool that comes close to competing with Photoshop and it’s Photopea, a subscription based/ad ridden website clone of Photoshop.

Recommending GIMP as a Photoshop alternative is like recommending LaTeX instead of Word. It’ll work for some people, and it’ll do some things much better even, but it’s ridiculous to assume any normal user of the proprietary product is going to be able to use the open source alternative without weeks or months of training.