@prologic@twtxt.net

I’ve been playing around with AI at home over the past few months and building my own neural networks from scratch (in Go) with genetic algorithms

Oh, is that all 🤣

That sounds like some intensive ‘playing around’ haha

@prologic@twtxt.net yep for sure. The part about concentrating too much power and reliance on the wealthy elite also resonated with me. Seems a good way to potentially end up in one of those dystopian futures you usually see in fictions where massive corporations have too much power and control over people.

A good blog post that makes some good points: Can I ethically use LLMs?

@movq@www.uninformativ.de reminds me how many Windows games using Proton (or WINE with similar patches) on Linux run better than some of the old native Linux binaries.

Woops, sorry if my Pod was offline for a few days, I hadn’t checked and needed to renew the domain xP

I just have to say the buttons page gives me the warm fuzzies with all the old school animated gifs. The internet seemed so fun back then…

I don’t think I’d personally be worried about migrating, just re-fetch. Sounds cleaner anyway?
Sorry I’m late to the party!

@prologic@twtxt.net Fully agreed. I’m far more likely to buy such mediums when DRM-free. I never go near Amazon eBooks etc because of their lock-in, and I have a Kobo eReader which needs to have the books side loaded unless directly from the Kobo store. I prefer DRM-free files every time.

Ehhh yeah, what could go wrong 🤔 😵‍💫

Hope you stay safe mate!

That’s pretty darn neat, the little indie movie (‘Flow’) made in Blender beat out some Disney/Pixar heavyweights. I actually watched it a month or so ago, nice little movie. It’s not the highest detailed animation or anything, but it has a style that it makes work.

Oh the ****ing irony 🤣 with all the IP infringement AI models usually do, and the companies are like “nothing to see here”!

Ahh yes, what I like to call “wild wild west” upgrading.😂
Felt like that when I upgraded/updated an Arch Linux machine that had been sitting for a couple years unused.

@movq@www.uninformativ.de That’s so awesome! I really oughta make use of the telescope I was gifted a few years ago…

@movq@www.uninformativ.de ha, very cool!

@movq@www.uninformativ.de my friend, I’m curious what is that interface? It’s like WindowMaker meets dwm, meets…? :D

Sounds like a plan

Of course they do

Hahaha wtf

What’s the alternative, plow straight through them? lol

@phoronix@feeds.twtxt.net oh gee the syntax of that thing 😆

@funbreaker@we.loveprivacy.club welcome!

@prologic@twtxt.net Cheers, mate, just saw this reply so thank you. And hope you are feeling better now!

I agree with what you say too. The whole thing is just an odd approach and can’t possibly be effective, all the while causing inconveniences or at worst, being plain weird and invasive like ID verification.

@prologic@twtxt.net yes.

@prologic@twtxt.net All good mate.

Oof, is it any wonder some of us don’t want to just give out our info online willy-nilly.

Also that credit card ‘encryption’ will likely land that company in very hot water, no doubt far away from PCI DSS requirements.

@prologic@twtxt.net what do we make of Labor’s proposed social media minimum age ban, I.e ID verification, and the likes of Yarn? I haven’t been able to find out exactly how far the legislation goes, but some have said it’s broad enough to include any site that even has a comment section 🤔 but that could be FUD.

I like this comment on Slashdot in the above link:

LLMs don’t have an understanding of anything. They can only regurgitate derivations of what they’ve been trained on and can’t apply that to something new in the same ways that humans or even other animals can. The models are just so large that the illusion is impressive.

So true.

This data selling shit needs to be more heavily criminalized.

@slashdot@feeds.twtxt.net Ugh, do we really need more brainrot, even if it is more open? 😅

@bender@twtxt.net 🤣

Lol, what a disgrace. And not surprised sadly.

@aelaraji@aelaraji.com Hmm that is worth trying. It is the same base Firefox I guess 🤔

@aelaraji@aelaraji.com Yep seems alright! Really fast too. I’m still using my main Firefox in general cos.. well it’s set up so much and it’s hardened, profile running in RAM, all that crazy stuff that got it working the way I want 😂

But keeping a good eye on Zen Browser’s progress.

Lol, this is actually a good thing by Apple. Doesn’t kill social apps at all, just prevents some harvesting of your entire address book by abusive apps like WhatsApp.

Seems to be working OK 🤔

@prologic@twtxt.net That’s definitely a little less depressing, when thinking of it that way 🤣 Be interesting when the hype dies down.

I’m not the biggest Apple fan around, but that is pretty awesome.

Getting a little sick of AI this, AI that. Yes I’ll be left behind while everyone else jumps on the latest thing, but I’m not sure I care.

@movq@www.uninformativ.de If it still existed I bet the first thing he’d do is convert it to Golang 👌🤣

@prologic@twtxt.net ‘Clownflare’ 🤣🤣🤣 Love it.

But yes the idea of a cheap VPS as a tunnel and keeping home network all local is a good one I reckon.

https://arstechnica.com/gadgets/2024/07/logitech-has-an-idea-for-a-forever-mouse-thatrequires-a-subscription/

The fuck…?!

@prologic@twtxt.net Good to know. I must admit I’ve never actually used a Docker instance, probably as I just assumed the overhead might be a bit much for my usual very modest servers.

@bender@twtxt.net Is it so maxed out you couldn’t fit a pretty small program like Headscale on it? Headscale by itself and only personal home type use as far as amount of peers go, it really isn’t noticeable I don’t think resource-wise. The Docker version I guess could be a different story.

@bender@twtxt.net Mine is about the same, though I have 20GB left 😅 In terms of resources, Headscale is using next to nothing though.

@prologic@twtxt.net Yes I suppose that is true. There is an article on Tailscale’s site that explains it all quite a bit: https://tailscale.com/blog/how-nat-traversal-works

To me, with CGNAT, it’s a small miracle that a direct connection can be made between peers (as opposed to going through a relay constantly) but it does indeed work. I guess to host it at home you would need to have it WAN accessible, and if you’ve already gone to the trouble of port forwarding etc… well 😅
Not that I could personally do that, but for those with static IPs etc.

@bender@twtxt.net on my hosted VPS, as I’m on Starlink which is CGNAT, I need some sort of external intermediary.

I setup and switched to Headscale last night. It was relatively simple, I spent more time installing a web GUI to manage it to be honest, the actual server is simple enough. The native Tailscale Android app even works with it thankfully.

@prologic@twtxt.net Interesting! Had no idea about that, but trust you to know of a self-hosted implementation 😅👌

I must admit Tailscale is really cool and why I haven’t used it before now is beyond me 😅

@movq@www.uninformativ.de

(I don’t really trust Android, though, and I suspect that apps can still install background services that are always active. Pure speculation and paranoid on my part, but still.)

Which is fair, but I would say the GrapheneOS devs in particular are also quite paranoid about this stuff and go to great pains to make sure this stuff can be controlled by the user.

I admit I’ve always compromised on this way too much myself, always to this day having Facebook Messenger just to communicate in my families group chats. Sure I run it in a Work profile on my GrapheneOS phone that I can switch off at any time, I can completely cut it off from network access any time as well, I can have a lot of rudimentary control over it, I use it as sparingly as possible, but it doesn’t change the fact everytime I use it we’re funneling private convos through bloody Meta’s servers and trackers etc.

Yeah, though sometimes the most clever devs aren’t always the best to deal with on a personal level. I seem to remember the (former?) lead dev on GrapheneOS (IIRC) was an ass hat and threw tantrums at the smallest things and would get stalkery and weird if someone criticised him, but he’s undeniably a brilliant coder and problem solver. Some people need to be more self aware of how their efforts might be harmed with their behaviour though.

After that talk about the Ladybird browser the other day, I see this article just pop up:

https://devclass.com/2024/07/03/ladybird-web-browser-project-now-funded-by-github-co-founder-promises-no-code-from-other-browsers/

Seems it’s gaining some recognition and support, I hope it can gain traction as we sure as anything need some genuine alternatives.

@mckinley@twtxt.net I must admit I was tempted to use EndeavourOS for an install on a HTPC (N97 mini PC) when it arrives to quickly get up and running, but then again I haven’t done a fresh install of Arch in quite a while so it sounds like things have simplified even more since then. Hmm…

I love Arch but they could probably do to have some of these warnings come down pacman itself

There is JavaScript, but not everything is implemented (properly). They’re writing everything including the JavaScript engine from scratch.

A huge effort 😲

@prologic@twtxt.net Very nice 👌

Very cool! Interestingly using your web app, the result was a higher bitrate than when I downloaded the best audio only option in yt-dlp (258 kbit/s vs 140 kbit/s).

Don’t quite understand that but nice work 😅

The whole gift card thing is crazy. You get scammers calling up people pretending to be from the IRS (just one example) and getting people paying with gift cards 😳

@prologic@twtxt.net Perfect 👌

@movq@www.uninformativ.de Golly, you are right 😅

Oh no, don’t tempt me. I’ve been on KDE for a while to not tinker and make it possible for my Windows using partner to use my laptop now and then, I’m trying to avoid the dwm/l addiction 🤣

@prologic@twtxt.net phew 😁

@prologic@twtxt.net yep pretty much!

Could you perhaps just have a check box to do the opposite, like “Don’t remember me”? I’ve seen that a couple of places I think. Sort of an opt in short lived login, if you’re at a public library or something etc.

@prologic@twtxt.net hey mate, all working well here so far. The login issue isn’t really an issue as far as actually logging in goes, rather if I get my password wrong it gives the response error code in console, the response of which contains the HTML for the wrong password page if you inspect it, but on the frontend itself nothing actually happens which is the confusion. Just stays on the login page as if it was never submitted. Am I alone in having this issue as well?

@prologic@twtxt.net Yes very very strange! I truly don’t know where to start on that one 🤣 Must be one of those really weird edge cases. Thanks for your help on this, I can at least post normally now.👌

I’ll check logging in etc tomorrow, time for bed lol 😴

Test

EDIT: Okay, convo works properly now at least

@prologic@twtxt.net It’s weird though cos I could reproduce it on any of my browsers on either my laptop or phone 🤔

@prologic@twtxt.net Firefox 126.0.1 is my primary

@prologic@twtxt.net Fix works!

Test

@prologic@twtxt.net Woops my blockquote got mangled but you get the gist haha

@prologic@twtxt.net I was wondering if my reverse proxy could cause something but it’s pretty standard…

server {

    listen 80;
    server_name we.loveprivacy.club;

    location / {
            return 301 https://$host$request_uri;
            <a href="https://txt.sour.is/search?q=%23proxy_pass">#proxy_pass</a> http://127.0.0.1:8000;
    }

}
server {

    listen 443 ssl http2;
    server_name we.loveprivacy.club;

    ssl_certificate /etc/letsencrypt/live/we.loveprivacy.club/fullchain.pem;
    ssl_certificate_key /etc/letsencrypt/live/we.loveprivacy.club/privkey.pem;

    client_max_body_size 8M;

    location / {
            proxy_pass http://127.0.0.1:8000;
    }

}

Test

@prologic@twtxt.net Firefox but it happens on Brave and Chromium on my laptop, or any browser (Brave, Chromium, Vanadium) I try on my Android phone.

@prologic@twtxt.net Sorry, my messages don’t get included in the current convo unless I tag you. Guess something gets lossed in translation with this weird posting issue. ANYWAY, it is rather perplexing. Clearly only an issue on my Pod, but what could the source of it be 🤔

Yeah sorry just realised, but just checked again and the referrer is the same (/post) on either the POST or the GET 🤔

Testing

Indeed /post is referrer…

GET /post HTTP/2
Host: we.loveprivacy.club
[…]
HX-Request: true
HX-Target: content
HX-Current-URL: https://we.loveprivacy.club/conv/vcpt7gq
Referer: https://we.loveprivacy.club/post

@prologic@twtxt.net Okay lemme see

@prologic@twtxt.net Yeah realised I hit the character limit lol

Hmm…

Jun 19 23:31:38 yarn_init.sh[61567]: [yarnd] 2024/06/19 23:31:38 (127.0.0.1:40254) “POST /post HTTP/
1.0” 200 0 3.402208ms
[…]

Jun 19 23:31:39 yarn_init.sh[61567]: [yarnd] 2024/06/19 23:31:39 (127.0.0.1:40262) “GET /post HTTP/1.0” 404 729 123.474001ms

@prologic@twtxt.net The login issue I cant yet narrow down as to when it happens as sometimes I login fine. But it gives off a 401 forbidden error. Anyway I’ve been focusing on the posting error as I figured it must be related. Registering and logging in as a new user works every time, which is weird.

@bender@we.loveprivacy.club hmm you may be right…

Unfortunately not on that front. Still the same 404 posting errors and oddly occasional login errors.

That’s why I was wondering if using Go 1.22.4 could be an issue. I don’t know how exactly. Only way to test is to rebuild it with an older version I guess, which is why I did the make clean in the first place. Old habits die hard lol.

Haha, yeah sorry about that, I wasn’t even trying to nuke the database either but it worked out that way 😩

I guess it got rid of all the dead spam bot accounts at least 🤣

@prologic@twtxt.net Correct… epic fail 🤣 Been a long day and I just wasn’t thinking, nor backing up properly. Oops indeed. My usual errors still exist though.

@prologic@twtxt.net Righteo, so rookie error - I obviously had some untracked, rather important files for starting my pod and I ran a make clean. Why I originally had them in the git directory is anyone’s guess. Anyway it blew away those files including the database so that’s that. So your good self and @bender@twtxt.net etc - apologies but your profiles got nuked as well (as did my own but easily recreated).

Another thing I noticed which was the reason I ran make clean in the first place. I noticed my pod was being built with Go 1.22.4. Could this be a problem @prologic? preflight.sh actually errors out about it…