Hey everyone,

It’s my first post here so my apologies if I mess something up.

Recently I’ve been refreshing my YouTube homepage constantly because I feel like it’s either:
A.) Suggesting me things I’ve already seen before
B.) Suggesting me things I have no interest in

So I’m going to go straight to the source and find some good YouTuber Channels I may have not heard of.

I primarily enjoy gaming critiques, history topics & natural disaster docs (kind of random I know), videos detailing scammers (SBF, Elizabeth Holmes, etc) but I don’t limit myself to these, I’m pretty much open to anything as long as it’s entertaining and/or informative… preferably both.

YouTubers I currently watch:

• NeverKnowsBest
• LiamTriforce
• Knowing Better
• Internet Historian
• Ordinary Things
• RennsReviews
• Scott The Woz
• SAWS
• ADoseOfBuckley
• DAngelo Wallace

Any recommendations are seriously appreciated. I don’t limit myself to a specific genre, but longer form content is definitely preferable.

Have you been visiting just too many different social media platforms lately, checking them out to see what the deal is? Well, same here. It feels like I've been a guest every night in different houses for the past month and I must say: I am exhausted.

But it's not over, far from it.

And I'm here to give you a heads up: we've witnessed platforms dying in the past, I'm guessing most of us have been a part of some sort of digital exodus before but I have a feeling that this one is going to be more painful.

Mainly because we've created so much data over the years and the majority of it got collected by centralised platforms. There are very few ways to take it with us and move elsewhere, it's all locked in.

Backing up your data now would also be a good idea, before some CEO comes with up the plan that it should be a paid feature.

I just want to say that this is all to be expected because the social web is in a transition period, and that golden bookmark doesn't exist yet. However, I think there are some contenders for it. What I want to ask is: where will you go next?

I've got some ideas, feel free to add your preference if I'm missing anything.

• Threads: Meta's Twitter clone that will be out some time this summer. It will be a federated (ActivityPub-enabled) platform.
• Bluesky: Jack Dorsey-backed Twitter clone. This one is also federated but it uses AT Protocol.
• Mastodon: The Twitter clone. It's got a fairly large userbase now, with lots of instances to choose from.
• Blog: Maybe it's not a bad idea to set up shop on a platform like Micro.blog (which is ActivityPub-enabled and has got community features built-in) and lead a quiet digital life.
• Threadiverse: Reddit-alikes.

I'm going to set up my first home server with an Intel NUC, but I can't decide what OS to use. Ubuntu seems popular but I like Pop!_OS and am not sure if that would be a good option. Then there's TrueNas and Unraid, but as a newbie, what's the best choice?

I'm also just curious what everyone else is using :)

Edit: Thank you for your great responses!

I've been a computer scientist in the field of AI for almost 15 years. Much of my time has been devoted to classical AI; things like planning, reasoning, clustering, induction, logic, etc. This has included (but had rarely been my focus) machine learning tasks (lots of Case-Based Reasoning). For whatever reason though, the deep learning trend never really interested me until recently. It really just felt like they were claiming huge AI advancements when all they really found was an impressive way to store learned data (I know this is an understatement).

Over time my opinion on that has changed slightly, and I have been blown away with the boom that is happening with transformers (GPTs specifically) and large language models. Open source projects are creating models comparable to OpenAIs behemoths with far less training and parameters which is making me take another look into GPTs.

What I find surprising though is that they seem to have only experimented with language. As far as I understand the inputs/outputs, the language is tokenized into bytes before prediction anyway. Why does it seem like (or rather the community act like) the technology can only be used for LLMs?

For example, what about a planning domain? You can specify actions in a domain in such a manner that tokenization would be trivial, and have far fewer tokens then raw text. Similarly you could generate a near infinite amount of training data if you wanted via other planning algorithms or simulations. Is there some obvious flaw I'm not seeing? Other examples might include behavior and/or state prediction.

I'm not saying that out of the box a standard GPT architecture is a guaranteed success for plan learning/planning... But it seems like it should be viable and no one is trying?

For the past week, I have been researching headphones/earbuds, buying them, and then cancelling the order immediately because I realized I was making a compromise on what I am wanting. Not a crazy amount of cancelled purchases, just maybe...three.

Anyways, I am in the market for the holy grail of headphones or earbuds that fits my needs but I am thinking I might have to buy multiple for the different scenarios that I am looking for. Which are:

• Preferably earbuds so I can take one out if I need to have an ear free for something.
• Great mic quality - I want to be able to use these for work call and when I go outside to walk. Really need something that handles wind and background noise pretty well that doesn't leave the other person on the phone annoyed talking to me. I don't want to walk around with a boom mic in front of face though.
• No preference on wired or wireless. If wired, I would like direct USB-C so I don't have to think about dongles or anything. I also read that mic quality degrades with Bluetooth? Not sure how accurate that is but the consensus I have read was that wired earbuds have better mics than Bluetooth. Wireless would be amazing if I knew that everything else worked well. I would take a short battery life to have top notch everything else.
• Decent sound quality for music. I already have a pair of aftershokz that I use for working out so I'm not stranger to compromised sound quality. But when I go out on walks I want to listen to language podcasts so I can hear clearly.
• Active noise cancelling - longshot but this would just be nice to have since I have gotten a taste for these with my Sony headphones.

I would be interested to hear your product recommendations or solutions to my wants. I have a feeling that the perfect product doesn't exist but at the bare minimum I would take something that sounds decent and has a good mic that handles outdoor sounds well. ANC and wireless are optional but would be very amazing to have.

I would be using these with a Samsung S22.

The vast majority of free and open source software available is well known for being clunky, having very unintuitive UI/UX and being very inaccessible to non-nerds.

We can see this in Linux distros, tools, programs and even fediverse sites.

I understand that a lot of it is because "it's free", but I also feel like a lot of people who make and use FOSS don't actually value user-friendliness at all. I feel like some of it is in order to gatekeep the less tech savvy out, and some of it is "it's good enough for me".

What are the best theories for why this is the case?

EDIT: A lot of replies I've been getting are focusing on the developers. I'm asking more why the users seem okay with it, rather than why the developers make it that way.

https://postimg.cc/phNYcrTJ

Just deleted my Reddit account. I was a Digg addict, and thereafter way to absorbed in Reddit for my own good.

Wanted to thank Christian for a brilliant app (if he ever was to see this: you poured your soul into that thing. Thank you for all you did). I’ve now deleted the app on all devices and am moving on!

Am looking forward to a fresh change.

I really like the feel of this place. Low key, easy to navigate and not crowded. And the civil conversations just blow my mind!

PS: sincerely appreciate the invite link!

Hi,

I saw that Proton now has a Family Plan that includes all of their services including mail, calendar, drive, etc.

I've been looking to detach my and my family's life a bit from the big corps like Google, Microsoft, Apple, and Proton sounds interesting.

At the same time I am wondering if they're really to be trusted any more than any other company. I found some critique of the E2E encryption being basically snake oil and not properly implemented etc. I am of course aware that there're always tradeoffs and while privacy is definitely nice, I don't need to go to an extreme degree of "no logs whatsoever, never even turning anything over to the police", I think I am just mostly looking to make me and my family less of a product of advertising, possibly AI/ML and so on, so some compromise is definitely possible.

I am curious, does anyone here have any experience with or opinions on the company and their products?

I am looking for a wi-fi tablet for singing from pdf sheet music on a shared Google Drive. The phone is too small of a screen.

In the past, tablets at this price are often junk, already end-of-life, and run so badly they are unusable (balky when simply scrolling). So, I am still using paper for now, afraid to buy yet another disappointing piece of junk despite not needing anything high end.

Any suggestions?

I, clearly along with many others, recently left Reddit and have personally decided not to return unless it sees drastic, lasting change. However, this has made it clear just how reliant I was on the site for multiple aspects of my life, ranging from local and world news, to hobby related announcements, to perspective, advice, and memes directly from marginalized communities. In Tildes I've already found some of what I've been missing, and I'm sure I'll find more as the community continues to grow, but Tildes doesn't have the same setup to allow for finding the same niche communities that I once had (nor should Tildes ever need to have that).

With all of that said, what gaps are you currently experiencing? Have you found any good sites or resources that have helped you or might help others in the thread?

I signed onto Reddit six years ago because that was where I could discuss stuff I liked intelligently. Due to social awkwardness and poor conversational skills, I feel like I contribute nothing to the conversation anywhere, on forums or IRL. I even signed up for Stack Exchange, Hacker News, and Ars Techinca for discussion, but they all look like you need a Ph.D. to contribute anything meaningful?

Am I alone on this? How can I be a productive part of the conversation?

What's it like over there right now? IIRC the blackout was supposed to finish today. I've decided to quit reddit unless something changes, so I want to do my best not to visit the site, but I am very curious about what the website and culture has been like for the past few days. And now that the initial blackout is 'over', how many subreddits have started to emerge again? Are people coming back now and acting like they've won? Has reddit responded to the blackout at all?

As the title says, could someone explain the Fediverse or this whole concept of federated instances?

I looked at Mastodon a few months back and recently at Lemmy and read a few articles and I still don't quite get it. I joined a Lemmy instance but didn't see any content from other instances. (God this makes me feel old)

To me, a non techie (who is reasonably computer literate), the problem with all these federation based platforms is that they don't seem immediately intuitive to laypeople.

It also seems to have the problem of individual instances being quite parochial in that there seem to be a lot of niche instances but few that are necessarily generic for an international audience, or conversely instances that are too vague ('this is an instance for humans') whereas with something like Reddit you just sign up and post in subs just like vbulletin based forums and subforums back in the day. Tildes is essentially similar in principle.

Please feel free to tell me, if I'm an old man yelling at the Cloud.