Permanently Deleted
tal @ tal @olio.cafe Posts 8Comments 313Joined 4 wk. ago
tal @ tal @olio.cafe
Posts
8
Comments
313
Joined
4 wk. ago
Have Trump act like Biden did.
George W. Bush was a Republican president and didn't get anything like the flak that Trump does. He also didn't behave like Trump does.
I remember reading someone on some service...maybe it was Steam?...saying that some wildly disproportionate percentage of their users had January 1 as their birthday. As in, people didn't even want to bother setting the month and day, which defaulted to January 1, just cranked the year back to whatever was required to avoid age-restriction hurdles.
Altman said in a statement accompanying the announcement, adding that the company is "building an age-prediction system to estimate age based on how people use ChatGPT."
I suppose our theoretical teenager could get an account on, say, Grok and ask it to rephrase all of his prompts as if they were written by a 30-year-old and then send the output of that to ChatGPT. Let the models fight it out based on their profiles of what constitutes an adult.
I'm sure that you could use MetaPost to procedurally generate a frame, then merge them into an animation.
A newer, similar language is Asymptote, and it looks like it has animation support.
This might be too far into the "watching gameplay" side of things or not up your alley, but I remember watching through some Arma II videos by Jester814 on YouTube and enjoying them.
He was one of a number of members who play in a group that tries to stay in-character, act as if it were a real Marines operation.
The 506th Infantry Regiment Realism Unit was founded in late 2014 by former members of the 15th MEU(SOC) Realism Unit. Many of our members are veterans and active duty military from around the world. The primary focus of the unit is light infantry combat in Arma Reforger utilizing real to life tactics, techniques, protocol, and communication.
While I do like some military history, I'm not really all that interested in light infantry tactics, so the content itself wasn't an immediate draw...but I wound up finding it fun to watch through the videos.
One of his ArmA II playlists:
He's also done ArmA III videos, which are obviously graphically-prettier, but at least in the few I watched --- and I haven't gone back and looked recently --- he didn't have larger numbers of coordinating players acting as larger, hierarchical military units, just a squad or maybe a couple of squads, and I didn't find it as interesting.
That being said, it's not something like Red vs Blue, which is content scripted purely for the viewer, not the people involved.
EDIT: Actually, I do remember a couple of large-scale ArmA III operations that he did. Just that there was a lot of smaller-scale stuff mixed in. That being said, could have been that when he was recording them, people hadn't switched to ArmA III yet --- it was still a pretty new game then. I should really go back and see what the situation is now.
Not exactly what you're asking for (specifically episodic) but relevant, and might be that this tweaks someone's memory if they're familiar with such a series and it's listed here:
Red vs Blue (their channel is locked behind some kind of 'join' thing... wtf...),
I think that they went commercial at some point, stopped just being a for-fun project on YouTube, and I assume that that's what happened. Wikipedia doesn't have anything clearly indicating the transition, though.
https://en.wikipedia.org/wiki/Red_vs._Blue
Although it is distributed serially over the internet, Red vs. Blue is also one of the first commercially released products made using machinima, as opposed to a product merely containing machinima. DVDs (and later Blu-rays) of every completed seasons are sold through Rooster Teeth's official website, as well as at several retailers in the United States, such as Target and Wal-Mart. Rooster Teeth claimed in 2017 that Red vs. Blue has sold more than 1 million DVDs of individual seasons and box sets.[76]
kagis more
Hmm.
https://old.reddit.com/r/RedvsBlue/comments/1ct84yj/rooster_teeth_shutdown_red_vs_blue_and_where_to/
With the final closure of the Rooster Teeth website, and RTs insistance on removing much of their content from YouTube, many may be wondering where you can watch that once popular web-series: Red vs. Blue.
I intend to keep this post as a repository, cataloguing public archives of RvB and RT content. There are of course still legal ways to acquire the show via YouTube, Amazon, Apple. However with that money no longer supporting RT, I can only recommend them on a convenience basis and instead offer some free alternatives.
The most comprehensive and accessible is: https://archiveofpimps.com/ which contains all of the main series and mini-series in both its original and remastered states. It appears to lack some PSAs but with tonnes of other RT content and a strong interface it’s currently the strongest contender.
Additionally, there is a Google Drive: https://drive.google.com/drive/folders/1NHnQK7-BgwaJiKJOYA7CkRQemKjTQ8Nd This also contains all of the main series and mini-series in both original and remastered formats. Individual episodes and movie edits, PSAs, Behind the Scenes, trailers and bonus material. For Red vs. Blue specifically this is the place to go.
*Our subreddit wiki also has a detailed watchlist order to help new viewers https://www.reddit.com/r/RedvsBlue/wiki/watch_order/
I urge people to maintain their own archives and if you are hosting your own public archives and wish to advertise them, let me know and I’ll add them to the post.
Thank you Rooster Teeth for 21 years of laughs. Let’s try and preserve their legacy. ❤️💙
That post was dated a year ago, so I assume that there was some change that happened around that point in time.
Honestly, it might not be a terrible idea to give them a bogus date, if you're concerned about privacy and want to encourage poisoning data that data-miners are using.
If they require some sort of actual validation, like an ID document, then I get you.
Court rules Europe can call nuclear and natural gas sustainable investments for its green transition
Jump
I'd guess that the argument on natural gas is one of the following:
It's replacing coal and coal emits more carbon
The problem is that coal-based power is rapidly declining, at least in the West, and it's not a huge chunk of the generation mix anymore.
https://ec.europa.eu/eurostat/web/interactive-publications/energy-2025
In 2023, the energy mix in the EU, meaning the range of energy sources available, mainly consisted of 5 different sources:
- crude oil and petroleum products (37.7%)
- natural gas (20.4%)
- renewable energy (19.5%)
- solid fuels (10.6%)
- nuclear (11.8%).
Oil is a pretty expensive way to generate power. I doubt that wood pellet power plants are very common. So if you want to reduce fossil-fuel-based generation past that, you probably do have to look at reducing natural gas.
We can use it in conjunction with intermittent renewables at lower levels to avoid expensive energy storage
Solar and wind aren't always available when someone wants to use them; they're intermittent. You have to fill in those gaps somehow. But energy storage is expensive and for pumped hydrostorage, the most-currently-economical form, somewhat geographically-limited. So the idea is that one uses natural gas instead of storing energy from a less-carbon-intensive source to fill in those gaps...but at least you're using less natural gas than one would if one weren't using renewable resources and just using natural gas all the time.
Also, one more tidbit:
Austria had sued the European Commission, the bloc’s executive, over the inclusion of gas and nuclear in the EU’s classification system for environmentally sustainable economic activities.
My guess is that Austria's probably unhappy because Austria uses a ton of hydropower, is very mountainous and has favorable geography for hydropower, so they'd prefer to have hydropower favored.
kagis
https://lowcarbonpower.org/region/Austria
This has hydropower in Austria being 56.2% of Austria's electricity generation.
I'm not going to watch the video --- I like most context in text rather than video form --- but while I will very well believe that:
- It's possible to optimize LLMs to make smaller models more effective than they are today. It would be very surprising if they were already optimal, given that the field is immature.
- It's possible to do a series of smaller, specialized models and keep models not-relevant to the current context unloaded from VRAM --- I believe that the "splitting into smaller specialized networks" approach is referred to as Mixture of Experts. This should improve memory efficiency for many problems.
...this is countered by the fact that once you free up resources, I also suspect that you can then go use those now-available resources to improve the model by shoveling more data into the model. And while there might be diminishing returns, I very much doubt that there is a hard cap on which one can get better results by throwing more knowledge at a problem.
From the blurb:
Sacrifice them to summon a mortal or kill them and use their corpses to resurrect some zombies!
That being said:
- I'm not totally sure that realism is necessarily the best complaint when it comes to the capabilities of a god in a god game.
- It seems like one could issue that complaint about most games. I don't think any, say, tennis games let you murder your opponent, though that's clearly at least a possibility in real life.
LLMs have non-deterministic outputs, meaning you can't exactly predict what they'll say.
I mean...they can have non-deterministic outputs. There's no requirement for that to be the case.
It might be desirable in some situations; randomness can be a tactic to help provide variety in a conversation. But it might be very undesirable in others: no matter how many times I ask "What is 1+1?", I usually want the same answer.
I'm not totally sure I follow.
If you're playing to whatever sound device you want, but it's not coming out the output you want (e.g. headphones and/or speakers and you want the other), the mixer program you use probably has an option to select the output. I haven't used plasma-pa, but with pavucontrol, it's in the "Output Devices" tab. For each device, there's a "Port" drop down.
Looks fine to me. I don't use KDE, but searching, it looks like KDE Plasma's audio mixer is "plasma-pa". The "pa" there will stand for "PulseAudio", so at least at one point, it'll have been based on PulseAudio. I dunno if it talks natively to Pipewire now.
kagis
https://old.reddit.com/r/kde/comments/v8hbyb/something_like_plasmapa_for_pipewire/
If you have the pipewire-pulse compatibility layer installed (which you really should), plasma-pa will work without any problems. Right now there is no pure PipeWire equivalent of it.
That was three years ago, so might be out of date, but at least then, it still used the PulseAudio API, so it may need pipewire-pulse to be active. In any event, I don't think that it'd hurt to have pipewire-pulse.
I'd check and make sure that pipewire-pulse is active too, and if so, try using plasma-pa to have PipeWire set the volume to whatever it is that you want set to. I assume that once you've set a volume with PipeWire, PipeWire will handle restoring it next time you log in. It does on my system.
there's no need to install pavucontrol but all i need to do is set up alsamixer to make the audio work
If you want to fiddle the audio at the ALSA level --- the hardware --- you can, but my guess is that in 2025, unless you have some kind of exotic need, what you probably want is for PipeWire to be setting the volume.
On my system, I use pavucontrol and pipewire-pulse --- using pavucontrol doesn't entail using PulseAudio. If you really know what you're doing and you're confident that you want to go bare ALSA, then you can probably go have systemd run a script at boot to run an alsactl restore command. I am pretty confident that that's not what you want to do.
It looks like there's a native console PipeWire mixer in Debian in the form of pipemixer, and I'd imagine that KDE Plasma probably has some sort of graphical mixer that either can talk natively to PipeWire or uses the pipewire-pulse PipeWire emulation of PulseAudio.
EDIT: Basically, you probably want:
------------
|Sound Card|
------------
------------
| ALSA |
------------
------------
| PipeWire | <- You interact with the audio stack at this layer
------------
EDIT2: You should probably see that a user-level pipewire is running if you log into your KDE environment and open a virtual termainal and you run:
$ systemctl status --user pipewire.service
It should say something like:
Active: active (running) since Tue 2025-09-02 00:27:11 PDT; 1 week 6 days ago
If you have the PulseAudio emulation support active, then ditto for:
$ systemctl status --user pipewire-pulse.service
I don't use KDE, so I don't know what the KDE mixer program is called or does, whether it talks natively to PipeWire or uses the PulseAudio interface, but KDE Plasma probably puts some sort of volume control in a system tray or something. And it'll probably use one of those two APIs to talk to PipeWire.
EDIT3: Basically, the only times I'd have been wanting to run things through ALSA directly were:
- When it was introduced but before any standardized sound server was deployed, so maybe early 2000s.
- Until JACK and later PipeWire showed up, talking directly to the hardware was a way to keep latency low for real-time processing, so there were some reasons you might want to do this if you were using pro audio.
- Early PulseAudio was pretty broken, so I wound up using ALSA in preference to it.
But that's all pretty much ancient history now.
August 2024, where the minister says: "We have to break with the totally mistaken notion that it is every man's freedom to communicate on encrypted messaging services
Are you going to prevent people from using e2e encryption systems that run atop existing non-encrypted systems?
If you're concerned about someone being able to see your activity, no blacklisting-based system --- which is what OP is talking about in terms of "blocking" would be -- on a system without expensive identifiers (which the Threadiverse is not and Reddit is not --- both let you make new accounts at zero cost) will do much of anything. All someone has to do is to just make a new account to monitor your activity. Or, hell, Reddit and a ton of Threadiverse instances provide anonymous access. Not to mention that on the Threadiverse, anyone who sets up an instance can see all the data being exchanged anyway.
In practice, if your concern is your activity being monitored, then you're going to have to use a whitelisting-based system. Like, the Fediverse would need to have something like invite-only communities, and the whole protocol would have to be changed in a major way.
As Miku has no physical presence, the relationship is purely platonic.
If someone isn't already banging on that, I am pretty sure that they will be before long.
kagis
https://aimojo.io/ai-powered-female-sex-robots/
AI-Powered Female Sex Robots: Top 8 Models for 2025
Yeah.
Legend has it that every new technology is first used for something related to sex or pornography. That seems to be the way of humankind.
--- Tim Berners-Lee, creator of the World Wide Web, HTML, URLs, and HTTP
I'm not totally sure about the chronology, but I think that the "old->new" block change on Reddit may have been due to calls from Twitter users. Most of the people I saw back on Reddit complaining about the old behavior prior to the change were saying "on Twitter, blocked users can't respond".
On Reddit, the site is basically split up into a series of forums, subreddits. On the Threadiverse, same idea, but the term is communities. And that's the basic unit of moderation --- that is, people set up a set of rules for how what is permitted on a given community, and most restrictions arise from that. There are Reddit sitewide restrictions (and here, instancewide), but those don't usually play a huge role compared to the community-level things.
So, on Twitter --- and I've never made a Twitter account, and don't spend much time using it, but I believe I've got a reasonable handle on how it works --- there's no concept of a topic-specific forum. The entire site is user-centric. Comments don't live in forums talking about a topic; they only are associated with the text in them and with the parent comment. So if you're on Twitter, there has to be some level of content moderation unless you want to only have sitewide restrictions. On Twitter, having a user be able to act as "moderator" for responses makes a lot more sense than on Reddit, because Twitter lacks an analog to subreddit moderators.
So Twitter users, who were accustomed to having a "block" feature, naturally found Reddit's "block" feature, which did something different from what they were used to, to be confusing. They click "block", and what it actually does is not what they expect --- and worse, at a surface glance, the behavior is the same. They think that they're acting as a moderator, but they're just controlling visibility of comments to themselves. Then they have an unpleasant surprise when they realize that what they've been doing isn't what they think that they've been doing.
I recall reading that one application of sentiment analysis in voice recognition --- like, determining what a speaker's mood is --- is that if someone gets upset on a call talking to a computer, the system will route them to a human.