How long are we talking in terms of a short movie? 🤔
@stigatle@yarn.stigatle.no Holy moly that’s awesome dude! 😱😍
@eapl.me@eapl.me Id love some help on yarnd 🙏
You are absolutely fluent in English if you can understand these - YouTube– This video (if it’s true this whone phun thing is British in origin?) makes me realize that I don’t actually find puns funny at all 🤦♂️ In fact I find them quite outrightly stupid 🤣
@abucci@anthony.buc.ci already tried very hard to get that on there but it spews errors 😢
Anyone have any ideas how you might identify processes (pids) on Linux machine that are responsible for most of the Disk I/O on that machine and subsequently causing high I/O wait times for other processes? 🤔
Important bit: The machine has no access to the internet, there are hardly any standard tools on it, etc. So I have to get something to it “air gapped”. I have terminal access to it, so I can do interesting things like, base64 encode a static binary to my clipboard and paste it to a file, then base64 decode it and execute. That’s about the only mechanisms I have.
@eapl.me@eapl.me Hmm interesting 🤔
codellama or codellama (See Models), or whether I'm expecting far too much out of these "glorified" token prediction machines, but all this seems to be good for is banging out repetitive keystrokes.
Was sort of hoping for a more objective response and experiences with using any LLM local or Oyherwise as a “coding assistant” 😁
codellama or codellama (See Models), or whether I'm expecting far too much out of these "glorified" token prediction machines, but all this seems to be good for is banging out repetitive keystrokes.
Is it actually any better using the much more (supposedly) powerful ChatGPT from OpenAI and wll that jazz that runs some crazy $250k/day to run?! 🤔 Anyone?
Been playing around a bit with Continue.dev and Ollama.ai in VSCode (which all runs locally). I have to say, Continue.dev is not a bad tool in terms of “utility” and the overall UX is kind of nice. However; I dunno whether I’m just using inferior models like codellama or codellama (See Models), or whether I’m expecting far too much out of these “glorified” token prediction machines, but all this seems to be good for is banging out repetitive keystrokes.
The darn thing is just so well umm, fucking stupid and just umm clueless?! 🤦♂️ I’m not really sure what to think of any of this anymore… It’s been so heavily hyped up over the past couple of years, but why? LIke you can’t really get these models to do much for you, even its “summarize this …” is kind of garbage really 😅
@stigatle@yarn.stigatle.no Gonna miss your lovely ocean scenery, but we’ll do something about that soon™ 😅 I believe I do still intend to build an external fully supported Twtxt<->ActivityPub bridge, so ya never know, you might just be back and ya’d never know 😅
@stigatle@yarn.stigatle.no Just saw this 😱 Sad to see you go mate 😢
@movq@www.uninformativ.de I think I misunderstood some aspects of Wireguard as mentioned here, not 100% sure, but so far things are much happier now with assigning /32(s) as Tunnel IP(s) for Peers and being a bit more thoughtful about the AllowedIPs 🤞 I’m only playing around with 3 devices right now, my core router (RouterOS), an Ubuntu 22.04 VM over at Vultr and my iPhone.
I think this is what I was missing in my understanding:
In other words, when sending packets, the list of allowed IPs behaves as a sort of routing table, and when > receiving packets, the list of allowed IPs behaves as a sort of access control list.
This is what we call a Cryptokey Routing Table: the simple association of public keys and allowed IPs.
@movq@www.uninformativ.de What’s your setup like? How many peers? How are they configured? (if you can share)
Hmmm really not getting this at al 🤦♂️ So far things appear to be a bit more stable, but the only changes I made was to assign addresses to peers of the form 172.30.0.X/32 instead of 172.30.0.X/24 and setting AllowedIPs to 0.0.0.0/0 for mobile peers (phones, etc) and X.X.X.X/24, Y.Y.Y.Y/24 for more static peers (remote VMs) where X and Y are the LAN and Wireguard subnets.
Hmm when I said “Wireguard is kind of cool” in this twt now I’m not so sure 😢 I can’t get “stable tunnels” to freak’n stay up, survive reboots, survive random disconnections, etc. This is nuts 🤦♂️
Huh hmm Boring Proxy actually uses SSH under the hood (written in Go) for the tunnelling 🤔 Clever, I would have done the same if I hadn’t learned about Wireguard 😅
@mckinley@twtxt.net Now that I have real experience with Wireguard, I’m seriously thinking about building my own “Cloudflare” replacement infra 😅 – And commodifying that somehow. Boring Proxy kind of does this too, but I may have a slightly different takes on things 🤔
I’ve set PersistentKeepAlive = 25 on both side. Let’s see if that improves things a bit…
@movq@www.uninformativ.de Wow that is wicked cool! 😅
@movq@www.uninformativ.de Only problem I seem to have is the connection keeps dropping out and never re-connecting until I forcefully disconnect/reconnect one side. Hmm 🤔
Wireguard is kind of cool 👌
@lyse@lyse.isobeef.org Hahahahaha 🤣🤣🤣
@movq@www.uninformativ.de Oh nothing has changed for me 🤣 I stay away from humans as much as possible, never shake hands, refuse to use public transport, etc, etc
Finally, in this day and age I do enjoy that I’m not reachable everywhere
This is the primary reason why I NEVER install “Work” apps on my iPhone. I’ve gotten into the habit for many years now, never to mix work and personal stuff. If I’m not on my Work Macbook, I’m not available – end of story.
@lyse@lyse.isobeef.org Off-by-one eh? 😅
@shreyan@twtxt.net you know I’ve already done this right? 😁
@shreyan@twtxt.net Nice 👌
@rrraksamam@twtxt.net Raspberry Pi Laptop and dumb phone?
This should be illegal.
@movq@www.uninformativ.de LOL 😆 That game was freak’n awesome! 🤣
I play chess, is that considered retro gaming?! 😆
@lyse@lyse.isobeef.org Haha that sounds like so much fun for the kids 😆
@movq@www.uninformativ.de Just flying down half a day early so not to stress too much 😆
@movq@www.uninformativ.de Cool! 👌 I’m away interstate at the moment for a work thing 😁
@movq@www.uninformativ.de Sure is! 😆
@linux_gizmos@feeds.twtxt.net Hmmm 🤔 This is a cool looking embedded board! 👌
AmaIng that on these plane WiFi networks, clearly powered by satellites, even get this much bandwidth 🙃
Latency is pretty piss poor, but you kind of expect that with a moving target 🤣
@juhi Hi 🤗 Welcome to Twtxt 👌
@juhi Hello! 👋
@thecanine@twtxt.net Very cool! 👌
@lyse@lyse.isobeef.org The way you wrote this made you almost sound “Aussie” 🤣
@yakumo_izuru@tsuki.chaotic.ninja Micely put 👌
It just doesn’t scale IMO 😅 Much better to have smaller interconnected decentralised social interactions 🤣
@yakumo_izuru@tsuki.chaotic.ninja This in and of itself ☝️ is why I believe so strongly that the whole “Fediverse” concept is utterly broken. You cannot scale human interactions like this. It doesn’t work on a centralised system, and it doesn’t work on a distributed system either. 🤣
@yakumo_izuru@tsuki.chaotic.ninja The Fediverse has bureaucracy?! 😱 Tell me more! 🤔 – Me just writing some code (for another project)