Giphy has a documented API that you could use. There have been bulk downloaders, but I didn’t see any that had recent activity. However you still might be able to use one to model your own script after, like https://github.com/jcpsimmons/giphy-stacks

There were downloaders for Gfycat - gallery-dl supported it at one point - but it’s down now. However you might be able to find collections that other people downloaded and are now hosting. You could also use the Internet Archive - they have tools and APIs documented

There’s a Tenor mass downloader that uses the Tenor API and an API key that you provide.

Imgur has GIFs is supported by gallery-dl, so that’s an option.

Also, read over https://github.com/simon987/awesome-datahoarding - there may be something useful for you there.

In terms of hosting, it would depend on my user base and if I want users to be able to upload GIFs, too. If it was just my close friends, then Immich would probably be fine, but if we had people I didn’t know directly using it, I’d want a more refined solution.

There’s Gifable, which is pretty focused, but looks like it has a pretty small following. I haven’t used it myself to see how suitable it is. If you self-host it (or something else that uses S3), note that you can use MinIO or LocalStack for the S3 container rather than using AWS directly. I’m using MinIO as part of my stack now, though for a completely different app.

MediaCMS is another option. Less focused on GIFs but more actively developed, and intended to be used for this sort of purpose.

Wouldn’t be a huge change at this point. Israel has been using AI to determine targets for drone-delivered airstrikes for over a year now.

https://en.m.wikipedia.org/wiki/AI-assisted_targeting_in_the_Gaza_Strip gives a high level overview of Gospel and Lavender, and there are news articles in the references if you want to learn more.

This is at least being positioned better than the ways Lavender and Gospel were used, but I have no doubt that it will be used to commit atrocities as well.

For now, OpenAI’s models may help operators make sense of large amounts of incoming data to support faster human decision-making in high-pressure situations.

Yep, that was how they justified Gospel and Lavender, too - “a human presses the button” (even though they’re not doing anywhere near enough due diligence).

But it’s worth pointing out that the type of AI OpenAI is best known for comes from large language models (LLMs)—sometimes called large multimodal models—that are trained on massive datasets of text, images, and audio pulled from many different sources.

Yes, OpenAI is well known for this, but they’ve also created other types of AI models (e.g., Whisper). I suspect an LLM might be part of a solution they would build but that it would not be the full solution.

Unless something has changed, it did. The page linked reads:

And, obviously, this POC is open source, the code is publish here on our forge.

The link takes you to their repos. The server repo has instructions on self-hosting directly on your server or with Docker. The app repo has code for both the iOS and Android apps. That’s good, because the iOS app at least doesn’t have a built-in way to select a different backend server.

Whisper is by OpenAI and as far as I know they have not shared the training code, much less the data sets, so the best you can do is fine-tune the models they’ve provided.

If use of Whisper is a problem, but the project is otherwise interesting to you, you could ask them to consider using a different STT solution (or allowing the user to choose between different options). I’m not aware of any fully open STT applications that are considered to be as capable as Whisper, but if you do, that would be great info to share with them.

Just don’t use Ubuntu. They do too much invisible fuckery with the system that hinders use on a server.

Would that warning also apply to Mint, since it’s based on Ubuntu, as well as other Ubuntu-based distros?

Your comment makes no sense.

The article you posted is from 2023 and PERA was basically dropped. However, this article talks about PREVAIL, which would prevent patents from being challenged except by the people who were sued by the patent-holder, and it’s still relevant.

• Assembled: 1200 USD
• Kit: 950 USD

The rules text says it creates an area of darkness, and with your interpretation, it doesn’t, which means your interpretation is wrong. Yes, the ability could be written more clearly, but the logic for a reasonable way for it to function follows pretty cleanly. Your interpretation is not RAW or RAI.

There’s a reply on RPG StackExchange that covers a similar line of logic to what I wrote above.

Remember that Fifth Edition D&D is intentionally not written with the same exacting precision as games like M:tG. The game doesn’t have an explicit definition of magical darkness, but it’s pretty clear that the intent is for magical to trump mundane (when it comes to sources of light and darkness). Even the Specific Beats General section says that most of the exceptions to general rules are due to magic.

If you have normal darkness everywhere, there isn’t a reason to use it, but you don’t always have darkness everywhere. In fact, you generally don’t.

Not all monsters with darkvision have access to light sources. Even if they do, they may need an action to use it or may be out of range. A torch or the light cantrip only has a 40’ range. If you collaborate on positioning with the caster, you can basically set yourself up to have advantage every turn thanks to the darkness, since as a ranged attacker you don’t have to stay within 40’ of your enemies.

Also, Gloom Stalkers can’t see through Darkness like Warlocks can, so this effect is useful to them in a way that the Darkness spell isn’t.

That all said, Tricksy wouldn’t do anything if it didn’t block nonmagical illumination, so it’s reasonable to run it as though it does. Sure, it still wouldn’t block even a cantrip, but it would block torches, lanterns, the sun, etc…

And running it as though it doesn’t block nonmagical darkness results in nonsensical behavior. You’re in a torchlit chamber and use the ability - now there’s a cube of darkness, blocking the light of all four nonmagical torches. If you move one of those torches away and back, why would it suddenly pierce the magical darkness? If it wouldn’t, why would a new nonmagical light source?

Thanks for clarifying! I’ve heard nothing but praise for Kagi from its users so that’s what I was assuming, but Searxng has also been great so I wouldn’t have been too surprised if you’d compared them and found its results to be on par or better.

By the way, if you’re self hosting Searxng, you can use add your own index. Searxng supports YaCy, which is an actively developed, open source search index and crawler that can be operated standalone or as part of a decentralized (P2P) network. Here are the Searxng docs for that engine. I can’t speak to its quality as I still haven’t set it up, though.

there is a better open source meta search engines

I already use Searxng and have never used Kagi, but I’m curious why you say that Searxng is “better.” Are you saying that because the quality of the searches is better, because it’s open source and Kagi isn’t, or for some other reason?

Maybe they just wanted to improve it for Gloom Stalker Rangers?

“But tante, then we will never have Open Source AI”. Exactly. That’s how reality works. If you can’t fulfil the criteria of a category you are not in that category. The fix is not to change the criteria. That’s playing pigeon chess.

This is a bad take. If your criteria aren’t grounded in reality, they aren’t useful, so of course you should change the criteria.

It’s also a missed opportunity to point to an AI model that did things right and that would qualify as “open source AI” even if that definition were not watered down. For example, OLMo (which I just learned about) says that they provide full insight into the training data as well as “full model weights, training code, training logs, training metrics in the form of Weights & Biases logs, and inference code.” Their most complex models are 7B models, which is enough to be relevant.

Saying “Meta and Alphabet will never release Open Source AI that meets the proposed definition” is fine. Saying “we’ll never have Open Source AI, period, that meets the proposed definition” means your proposed definition needs rewritten.

Yes, but have you seen some of the decisions the Supreme Court has come up with?

Do you only experience the 5-10 second buffering issue on mobile? If not, then you might be able to fix the issue by tuning your NextCloud instance - upping the memory limit, disabling debug mode and dropping log level back to warn if you ever changed it, enabling memory caching, etc…

Check out https://docs.nextcloud.com/server/latest/admin_manual/installation/server_tuning.html and https://docs.nextcloud.com/server/latest/admin_manual/installation/php_configuration.html#ini-values for docs on the above.

Your Passkeys have to be stored in something, but you don’t have to store them all in the same thing.

If you store them with Microsoft’s Windows Hello, Apple Keychain, or Google Password Manager, all of which are closed source, then you have to trust MS/Apple/Google. However, Keychain is end to end encrypted (according to Apple) and Windows Hello is currently not synced to the cloud, so if you trust those claims, you don’t need to trust that they won’t misuse your data. I don’t know if Google’s offering is end to end encrypted, but I wouldn’t trust it either way.

You can also store Passkeys in a password manager. Bitwarden is open source (though they did recently introduce a proprietary, source available SDK), as is KeepassXC. 1Password isn’t open source but can store Passkeys as well.

And finally, you can store Passkeys in a compatible security key, like the YubiKey 5 series keys, which can each store 100 Passkeys. This makes them basically immune to being stolen. Note that if your primary interest in Passkeys is in the phishing resistance (basically nearly perfect immunity to MitM attacks) then you can get that same benefit by using WebAuthn as a second factor. However, my experience has been that Passkey support is broader.

Revoking keys involves logging into the particular service and revoking them, just like changing your password. There isn’t a centralized way to do it as far as I’m aware. Each Passkey is only used for a single service, after all. However, in the same way that some password managers will offer to automatically change your passwords, they might develop a similar for passkeys.

Do any of the iOS or Android apps support passkeys? I looked into this a couple days ago and didn’t find any that did. (KeePassXC does.)

You have your link formatted backwards. It should be Vaultwarden, with the link in the parentheses.

Up until a year ago, the README explicitly said they didn’t claim to be an open source project: https://github.com/jgraph/drawio/commit/8906f90ac0cc50a0c6da77c28cf9b2b2339277b1#diff-b335630551682c19a781afebcf4d07bf978fb1f8ac04c6bf87428ed5106870f5L10

For starters, it was never “open source”…

From your link:

Instead, as Winamp CEO Alexandre Saboundjian said, “Winamp will remain the owner of the software and will decide on the innovations made in the official version.” The sort-of open-source version is going by the name FreeLLama.

While Winamp hasn’t said yet what license it will use for this forthcoming version, it cannot be open source with that level of corporate control.

If I upload the source code for my project on Github/Forgejo/Gitlab/Gitea and license it under and open source license, allowing you to fork it and do whatever you want (so long as you follow the terms of my copyleft license), and I diligently ensure that code is uploaded to my repository before being deployed, but I ignore all issues, feature requests, PRs, etc., is my project open source?

Yes.

Likewise, if Winamp had been licensed under an open source license, it would have been open source, regardless of how much control they kept over the official distribution.

Winamp wasn’t open source because its license, the WCL, wasn’t open source.