Vibecoded app or not, I really hope this refreshingly unhinged writing style is all you.

Looks like it’s behind an experimental flag, only applies to new rooms, and isn’t available on the public app.element.io instance right now (the Labs screenshot they showed doesn’t show the option on element.io+matrix.org when I checked just now):

https://github.com/element-hq/element-web/pull/31513

For it to work, looks like it needs another in-progress feature to allow new members to decrypt previous messages (Matrix can’t do that yet???):

https://github.com/matrix-org/matrix-spec-proposals/pull/4268

(That one isn’t even in the spec yet, it’ll take a lonnnnnng time before we see any real movement on it, possibly a year or two. A year or two before you can see encrypted messages from before you joined a room. Insanity.)

A contributor calls this encrypted metadata feature a “partial prototype:”

https://github.com/element-hq/element-meta/issues/1214#issuecomment-3714132569

And that issue was opened in 2017. Pretty crazy they’ve been discussing this for nearly 10 years and it’s just now “a partial prototype,” but that’s Matrix for you.

So it was all of your messages?

Can you explain what bridge you used and how you got the whole history moved over?

Every messages in your server’s entire history, or just the channel layout?

I know the Mautrix bridge can create the channels for you, but their backfill feature was not designed for full archival, it can’t really do that.

Thank you!! Really looking forward to this, Movim is already pretty nice.

And I believe the devs are working on adding discord-like channels with multiple groups under a single community.

If you have a link to a discussion or issue where they’re tracking this, I’d love to follow it.

They’ll also need to figure out how to let rooms have more than 1000 people, which is currently the limit on Movim (and it might be an XMPP limitation). Right now, Matrix seems like the only option for a Discord migration, but I hope that changes soon.

I might move my Discord community to Matrix, but I’d want a full 100% clone of the server’s history. Seems like that’s almost possible?

Discord scrapers like DiscordChatExporter can dump a whole channel to JSON (although I don’t see a lot of scraper options that let you pause/resume/retry an export… looking for suggestions if people have any).

And the Mautrix Discord bridge already makes it look like real users sent bridged messages. I think with a little customization, the Mautrix bridge could be modified to import an enormous JSON file to fully mirror an entire Discord server, instead of just grabbing messages in real time. It’s already got the functions there to “send a message as xyz user.”

Matrix or not, “take your Discord dump and clone it somewhere federated” would be huge for migrations like this. I really hope someone works on that soon 🙏

Be careful, that’s an AI generated review. Definitely try to find other human reviews before you buy it.

If you go through that user’s other reviews, while they are very funny, they’re also very AI generated. Tons of contrivances and the classic “and honestly?”

They were almost all posted on April 13th, and they all drop the “brand name” of the product in just really unnatural ways. This is 1,000% a bot.

Do go visit the profile though. The sex toy reviews are pretty funny.

If you want some quick examples of how this is obviously AI:

Ah I see. I’ve gotten into some back and forth with the maintainer of Caddy about these build inconsistencies and lack of versioning, and he responded by locking issues on me… twice.

After that communication, I resigned to just keep things as they are because upstream is not willing to make the experience better for use cases like these.

He is impossible to work with, but Caddy is just really good :/

What issue did you have? If I can handle it for the user automatically, I can add a best effort attempt to avoid it.

Hi!

The project is still active! I just haven’t needed to develop any updates because it’s currently stable, and the upstream Docker compose template, which is a reference for my project, hasn’t changed in 9 months:

https://github.com/LemmyNet/lemmy-ansible/blob/main/templates/docker-compose.yml

Whenever Lemmy releases a new Docker container, it will just pull the latest one automatically. This will continue to be compatible indefinitely until the compose deployment (linked above) has any breaking changes added to it that I need to incorporate.

Thank you for the shout out :)

Ah I see, that makes sense. Thanks for explaining, I learned something from your comment and the other one.

I am not up to date on all these license debates, but don’t you think equating Alpine Linux to “locked down DRM” is just a bit of a logical reach?

Alpine and its components are fully open source, you can make whatever changes you want to them. I am not seeing the argument here.

The people downvoting you have never experienced perfectly regular trees (and don’t understand CrossCode inside jokes).

Just being open source is not enough to solve this problem. You would need a record of your “home instance” stored somewhere that all Lemmy instances would always have access to it. Ignoring the security implications of that, you still can’t use browser storage or cookies for that storage. In browsers, those storage mechanisms are restricted to the websites that stored them in the first place. Browser extensions have their own storage, so they don’t have this limitation.

I can’t think of an architectural solution to this problem without a browser extension.

The closest alternative that could be upstreamed is what Mastodon does, which is prompt the user to enter their home instance every time they try to interact with something while not logged in. An upstream solution can never be fully automatic like a browser extension, but this may be “close enough.”

They explain it in the video. They already use algorithms to detect if things are buildings or not.

But if their algorithm can’t make a determination or is uncertain below a certain threshold, they send it to Maptcha to get a bulk human opinion.

That’s not the person I originally asked.

The person I asked actually did reply to me on this thread… but didn’t answer how they know all this.

How do you know all of this?

I just want a reasonably priced generational bump over the Index. Most PCVR headsets that have pancake lenses are either obscenely priced, are ridiculously heavy, or have reportedly terrible QA. From what I’ve seen lately, usually all three are true.

You can get close in features and price with something like a Pico or a Quest, but they lack direct DisplayPort connection, so it’s compressed wireless PCVR, compressed “wired” PCVR (which basically uses a networking protocol anyway), or no PCVR at all.

Myself, and I’m sure a ton of other people, are hoping for the Deckard to be “huge” for the PCVR market, just like the Index was when it released. Maybe we’re all coping, and we probably even are, but I think a lot of people are generally unhappy with the state of the PCVR hardware market right now.

So all this is to say… I really hope this thing is much better than a glorified flat screen projector.

If you don’t want to open this and solder a chip, then no, you can’t do what you want.

The closest you can get is to enable AutoRCM, which will cause the Switch to always boot into recovery mode and accept a payload. This skips the need to use a jig in the Joy-Con rail, but you still need to inject a payload. And because recovery mode is just a black screen, you don’t have any visual feedback to know if the Switch is actually in recovery mode, or if the battery is just dead.

Your best option is to just boot into whatever OS you use most, then make it a habit to keep it charged enough to not shut down.

However, I do believe they’re entering a very oversaturated market, with the likes of 8BITDO and GuliKit creating high-quality, affordable controllers in the same niche. ANBERNIC would have to aggressively outprice them to get any kind of attention, but only time will tell.

Not really. 8BitDo really blew it with the Ultimate. They confusingly have two different “versions” of it, and neither have the full range of device compatibility that previous 8BitDo controllers had. The most egregious exclusion from the Ultimate was Xinput over Bluetooth. I still have no idea why they decided to drop that.

Its design takes inspiration from a modern-day Xbox controller and is fully compatible with PC, Steam, Nintendo Switch, Android, and iOS using Bluetooth 5.3 and 2.4g connection. It can also be connected using a USB-C wire too.

If this new controller has Xinput over Bluetooth, all of the compatibility from above, and a strong battery life, it might be a day 1 buy. It will have hall effect sticks, so this sounds like everything I wanted the 8BD Ultimate to be. I hope there aren’t any showstoppers once reviews start coming out.