|
Sage Grimm posted:https://twitter.com/EccentricFlower/status/1616185035387715593 This has also killed off my beloved TweetBot, and I flatly refuse to use the Twitter app. If anyone ends up finding an alternative that allows you read tweets solely from the people you want to see tweets from, in the chronological order in which they were posted, I'd love to know what it is. The closest thing I've found is an RSS feed setup, but the options I've seen recommended have been pretty shady.
|
#
¿
Jan 21, 2023 03:49
|
|
|
|
| # ¿ May 4, 2024 03:07 |
|
|
withak posted:My favorite part of every google search now is the first dozen hits being “best [search terms] of 2023” articles written by AIs trained on Amazon reviews. The one nice thing about these is that they'll give you enough brand names that you can feed back into a search, and it'll usually pull up a Wikipedia article showing you what your choices are, or to introduce into a reddit / [insert appropriate community here] search in pairs or trios to find comparisons.
|
|
#
¿
Feb 12, 2023 20:49
|
|
|
Rocko Bonaparte posted:That's pretty funny because I saw that in this article posted earlier: I do that for the majority of my searches now, or if not Reddit, then some appropriate community. The initial search might start as a general googling, but it almost inevitably gets to using site:reddit.com or an equivalent. It just works better, period.
|
|
#
¿
Feb 13, 2023 02:12
|
|
|
VikingofRock posted:Apparently Google can't even afford to give its in-office workers a desk anymore. Instead, they are telling people (including people who currently come in every day) to come in two days / week and share their desk with someone else for the other two days. The headline is misleading. This applies specifically to their cloud unit, which is bleeding money and under some serious pressure to perform. This reads like a management of a struggling unit taking desperate, ill-advised measures to cut costs.
|
|
#
¿
Feb 27, 2023 20:57
|
|
|
Large-language model-type "AI" is not only applicable to tabletop roleplaying games, it's going to be enormously influential. In some pretty meaningful ways it has existed since 2007, and a version of it is at the cutting edge of TTRPG design. Back in '07, the Mythic Game Master Emulator was released by Tom Pigeon of Word Mill Publishing. It's what's known as a "system-agnostic" game supplement: a standalone book containing rules that will work with any game system, not just Dungeons & Dragons. Its purpose is what it says on the tin: it allows you to play tabletop RPGs, most of which are meant to have one or more player-characters and one or more game masters, solo, by essentially acting as the game master. I won't get too into the weeds on how it does this, but the short version is that any time you'd normally interact with the game master rather than the rules of the game, you consult the emulated game master in the form of - tell me if this sounds familiar - weighted randomized tables. The more detailed a response you need, the more you can drill down through its responses by asking whether X or Y is true, and assigning weights to each side. The emulator is also salted with outcomes which break you out of the track you expect to follow and send you down another trail (and another set of tables), such that you're never in complete creative control. It's a relatively crude system, but it works well enough to generate some legitimately compelling results, and it has remained in print for over 15 years. You may be seeing where this is going by now. (Note: this is all based on the 2007 version of Mythic GME. A modernized version came out late last year, but I haven't used it so I won't comment on it.) Now, all of this is ancient tech, since 2007 is an eternity in the timeline of TTRPG development. I suspect goons are more aware of this than most audiences, but it bears mentioning nonetheless: there are many, many tabletop RPGs other than D&D and Pathfinder, and the art form is currently in a golden age of design. Mythic GME hasn't been terribly influential on game design, since it's an overlay on top of other rules sets, but that has changed with the publication of Ironsworn and its sequel system Starforged, written by Shawn Tomkin and published in 2018 and 2022 respectively. Ironsworn's premise is that it's a system designed for solo or GM-less co-op play, which it accomplishes brilliantly by integrating something along the lines of the Mythic GME into its core mechanics. MGME's greatest strength and greatest weakness has always been its system-agnosticism: it works with anything, but it doesn't work beautifully or elegantly with anything. Ironsworn squared the circle by building its GM emulator into the system and elaborating on the techniques pioneered by Mythic. It retains the basic function of "ask questions of weighted probability tables until you get sufficient clarity to engage with the rules on your character sheet," then goes a step further with oracles. Oracles are random tables, usually with entries from 1 to 100, that are used as themed creative prompts that you engage with either when the rules of the game tell you to do so, or when your appraisal of fiction tells you they're necessary or interesting. This is all a very dry, surface-level description of it, but anyone who has played one of these games will tell you that the results are genuinely kind of spooky. It often feels as though you're playing with another person, but one who you can only communicate with through a Ouija board. If you've played around with ChatGPT enough to see it perform some of the spookier tricks in its repertoire, this will all feel familiar. It even works in a similar way: ChatGPT is just ("just") picking out the next word in the sequence via a hellaciously complex set of weights salted with randomness to make it feel more human, while Ironsworn and its descendants are doing something very similar, but at a much slower pace, and with prompts that are tailored to generate outcomes appropriate to their genre. And it works. Mythic GME functioned well enough to cement a small but persistent place in the hobby for 15 years and counting; Ironsworn works so well that it has become A Big Deal and is inspiring tons of new work, including the current topic of discussion in TradGames general chat, Across A Thousand Dead Worlds. There is absolutely no doubt in my mind that you could build an RPG whose mechanics linked into ChatGPT right now, and have it be a genuinely compelling GMless experience. It wouldn't be easy, and the first people to do it will fail, because TTRPGs have only a handful of designers who are capable of more than cargo cult design at best. But sooner than you think, we will see games that use this technology to good effect, and as soon as the public can get a hold of LLMs as advanced as ChatGPT, in a way that lets us tailor them to a specific purpose, GMless RPGs are going to absolutely explode. As an aside, if anyone's curious about how Ironsworn plays out for real, I ran an entire Ironsworn adventure a couple years back in The Game Room, which will illustrate how this works much better than I can explain it here. Ironsworn itself is also free, so you can check it out for yourself and see what I mean about the spookiness of oracles.
|
|
#
¿
Mar 8, 2023 05:06
|
|
|
NomNomNom posted:Just hinged your whole Ironsworn thread, that was dope and I want to read more (but I fear I don't have the creativity to play it myself) You definitely do! That's the remarkable thing about the "technology" that Ironsworn et al are based on: it's like a cybernetic augment for creativity. And, to bring it back to the thread topic, that's why the intersection of TTRPG design and LLMs has so much potential. You don't need to replace the other people at the table with replicants, you need to have a structure that fires your creativity and, crucially, challenges your assumptions and gives you surprises within the bounds of genre and setting. Modern solo RPGs already have a remarkable capacity for this, but it's very much in the era of stone tablets and abaci compared to where it can go as game design and technology advance. It may seem like I'm overhyping this, but anyone with the least curiosity about it ought to spend an afternoon with an oracle-based system (reminder that Ironsworn is free!), and an evening mulling over what skillful assistance from an LLM could do in that context.
|
|
#
¿
Mar 8, 2023 21:46
|
|
|
Ghost Leviathan posted:In general I think the idea of being able to fire all the designers and press a button that makes large chunks of the game for you is a pipe dream, but I wouldn't be surprised if down the track algorithmic generation can be a useful tool. But that's the thing, just a tool like any other, like the procedural generation we've had for decades by now, and it's still going to need a lot of design work at every stage to make it actually fun and immersive. This is dead on, IMO. Anyone expecting these tools to replace game designers or become a full-fledged GM for them in the near future is going to be sorely disappointed. As a tool for structured creativity though, they have enormous potential. I’m excited to see what happens when future models get into the hands of the few really gifted game designers bobbing around in the indie tabletop scene.
|
|
#
¿
Mar 9, 2023 23:56
|
|
|
Magic Underwear posted:I know jack poo poo about banking regulation but this seems like it will create huge second and third order effects. Or maybe it won't because investors and bondholders and executives are being wiped out? As a fellow not-bank-knower I’m also quite curious to see some professional takes on this, some real wonk stuff, but I have no idea where to start. Any finance goons with recommendations on whose takes are going to be worth reading? Name Change posted:Someone else can absolutely never read my Dragonlance cookbook that was itself handed down to me, is all I'm saying. The recipe in there for Otik’s Spice-Fried Potatoes is legit pretty good, can recommend. Also: it is my obligation to inform the thread that the one time we see the name pronounced phonetically, it’s pronounced “Drisst,” so this drone company is either basing its name on something else, or it’s Doing It Wrong 
|
|
#
¿
Mar 14, 2023 00:41
|
|
|
feedmyleg posted:I love how punishing bad behavior is impossible in these tech ventures. In any sane world, the solution for creeps groping another person's avatar would be to ban that person from being able to interact with others. But then number go down. So these spaces will always be filled with awful bad actors seeing no consequences for their actions and making the entire venture insufferable to normal people, keeping number from going up. "Petition to have someone banned" isn't a scalable solution for the kinds of virtual worlds they want to build. If your goal is to have millions of simultaneous users (let alone tens or hundreds of millions), you need to design systems that prevent as much misbehavior as possible. Any time you need to involve a human judgment call, even for something as straightforward as "yeah they were humping someone, mash the ban button," it introduces enormous - and abusable - latency into the system. Having humans with the banhammer is necessary, but you can't rely on that being The Way Things Get Solved.
|
|
#
¿
Jul 7, 2023 20:50
|
|
|
Boris Galerkin posted:This is going to become more and more common as time goes on and yet free speech absolutists keep arguing that people who make generative AI software shouldn’t be handicapping the software to prevent poo poo like this from happening. Nothingtoseehere posted:The point is more you can't. The versions that can do this are the "basic" ones without any fancy additions, creatable from open source academic code already published (and not intended for this) and are just trained on the large, large amount of existing pictures of naked women on the internet. You can certainly make it so you have to go to some shady russian website to download the model rather than be able to sell in the open, but that doesn't stop piracy (or porn) right now, it won't in the future. Nothingtoseehere is correct, there's no getting this genie back in the bottle. The software that can do this isn't just hosted apps, it's freely available on Github for local installs, and the system requirements are trivial. Even if every site/app on the internet capable of running image-generating models for users went down right now and never came back, there would still be hundreds of thousands if not millions of local copies of Stable Diffusion and its interfaces. And by god, those people will not be denied their porn-generating software. I learned Stable Diffusion to do tabletop RPG stuff for my home games, and it's been fantastic for that. But if I had a penny for every time I told a SD model to make some character art and its first draft gave me borderline porn, well, I'd have a lot of pennies. The users who are training these models (and it's primarily users now, not the companies who came up with the original model) are mostly training them to make pictures of sexy ladies, to the point where you can't browse Civitai, the big model archive, without being inundated by porn models and add-ons (although it's better than it once was). The models tend to get "mixed" and folded into one another to achieve different aesthetics, creating what amount to genetic lineages of models that aren't well-documented. As a result, it's common to grab a Stable Diffusion checkpoint that says it's for, I don't know, spooky autumn vibes, only to discover that any character you put into it will be a big tiddy witch in panties unless you specifically tell the model to do something different. As for handicapping it, it's impossible. Anyone who can run Stable Diffusion can also teach models new concepts, if they're patient enough to assemble the material and let their computer chew on it for a while. Stable Diffusion XL (SDXL) came out recently, and by all accounts it's even easier to train on than SD1.5. Despite SDXL being deliberately stripped of porn (to the best of its makers' abilities), everyone who's used this technology knew exactly how easy it would be for regular users to insert those capabilities back in, and sure enough, just scrolling through Civitai will get you model after model of porny stuff for SDXL. All of this is to say, this technology is out there now, it's trivially accessible, it's captured the "horny people" audience, and as such no force on earth will suppress it. Any future version of this software that include things like SDXL's invisible watermarking will be insufficient, because Stable Diffusion 1.5 is already perfectly capable of making photorealistic people, places, and things. Our cultures need to adapt to this capability's presence, fast.
|
|
#
¿
Sep 25, 2023 02:27
|
|
|
Mr. Nemo posted:And my homebrewed cocktail of 5 different adblockers stopped working, I added a few more and now they work, but only in incognito mode. I'm still holding on to Chrome, but I may have to switch. Apparently it’s more reliable to just stick with uBlock Origin and no other adblockers, as they seem to interfere with one another. Firefox with uBlock is pretty much a one-and-done solution for the time being, though.
|
|
#
¿
Nov 10, 2023 00:04
|
|
|
Leon Trotsky 2012 posted:Unless they changed it recently, Netflix has a DRM feature that hard codes all browsers to 720p - except for Microsoft Edge. Edge was the only DRM compliant browser as of a few years ago. Do you have a link to those add-ons handy? It sure would be nice to not have to resort to piracy to get decent resolution on a service I pay for.
|
|
#
¿
Jan 5, 2024 21:41
|
|
|
Leon Trotsky 2012 posted:I don't use the Chrome one, but I believe this is it based on Google: Excellent, I'm on Firefox as well so I'll search around. Thanks!
|
|
#
¿
Jan 6, 2024 03:21
|
|
|
mobby_6kl posted:How are we going to accomplish that? Adobe and IIRC Getty for example have generative AI trained entirely on explicitly licensed material. Let's say their models are good enough to make all stock photography obsolete as well as a good number of Photoshop monkeys. Do we just ban this or what? Not to mention there’s already any number of Stable Diffusion checkpoints and LoRAs already floating around out there that people can use, and people hoard them - even more so now that Civitai has begun its inevitable monetization pivot. You could pass sweeping generative AI bans today, and those files would still be out there, easily distributed on filesharing sites, for anyone to grab.
|
|
#
¿
Jan 20, 2024 19:39
|
|
|
Antigravitas posted:I went with a Kobo instead specifically because Kindles jumped the shark years ago. The Kobo Libra II has nice buttons and a touch interface, and weighs basically nothing, I love it dearly.
|
|
#
¿
Jan 29, 2024 11:43
|
|
|
burnishedfume posted:I admit this isn't representative of all of her other work, but specifically her videos on trans people/healthcare are some of the worst I've seen outside of actual overt transphobes. This video/post by skepchick/Rebecca Watson is a pretty good overview of Hossenfelder's "Is being trans a social fad among teenagers?" video (still up on Sabine's channel, not gonna bother linking to it). While Sabine doesn't go full transphobe and acknowledges trans people as real, she also presents "Rapid Onset Gender Dysphoria", a conspiracy theory that memes are turning kids trans, as a real thing citing only a poll of mumsnet posters as evidence. She also then goes on to question health risks of puberty blockers citing only incomplete studies into potential health risks of puberty blockers (such as minor changes in bone density which may not even be irreversable), while completely dismissing studies that showed massive mental health improvements for trans kids on puberty blockers (for lacking control groups that the studies absolutely had). This seems like one of those weird outgrowths of people not being around kids and forgetting what they're actually like, combined with Social Concerns. For context, I've done volunteer work with kids for many years, seen whole cohorts of them grow up at this point, and some of them are my Actual Friends at this point - and yeah, a couple of them did actually go through a phase where they were calling themselves trans, and they later changed their mind. Others identified as gay or lesbian, then decided they were bi or omni, then straight, then went back to being full-on gay again over the span of 12 months. And y'know what, nobody should be surprised by this! One of the very best things about Gen Z and Alpha is that for a lot of them, being some flavor of queer is normal and supported. When kids feel like they have support and safety, they feel free to try on gender and sexual identities in experimental ways, in the same way that earlier generations tried on a bunch of social identities. Only one of the kids I know had their coming out as trans actually stick, though about 20-30% them have ended up as some variety of LGBTQIA+. I suspect it's easy for olds to look at this phenomenon, especially filtered through and exaggerated by things like TikTok, and think "memes are making the children trans," when the reality is that it's a combination of a couple generations of trans kids who feel safe being out when they wouldn't have in the past, and kids who genuinely are playing gender-and-sexuality dress-up as a form of self-exploration. This isn't super on topic for the Tech Nightmares thread though, so if someone wants to continue this discussion I can do it in PMs or another thread.
|
|
#
¿
Mar 4, 2024 23:39
|
|
|
Leaving vacations aside, there are innumerable cases where you need to get a specific human being to a specific place, in person, and that person and place are thousand miles or more apart. We built a world where flight is a necessity, that’s not changing any time soon.
|
|
#
¿
Mar 22, 2024 21:23
|
|
|
I would look at Kagi through this lens: they’re definitely recording everything you search through them, because of course they are. When the inevitable enshittification happens, are you comfortable with them having however many years of your searches and that profile squirreled away? If yes and you can afford the subscription, Kagi is probably a good choice.
|
|
#
¿
Apr 3, 2024 19:53
|
|
|
|
| # ¿ May 4, 2024 03:07 |
|
|
dr_rat posted:Obviously the Cybertrucks AI is so intelligent and human like, it decided to binge watch its favourite show for four hours and fifty-eight minutes, before finally, begrudgingly getting around to doing the two minute reboot. Every instance of a Cybertruck contains Murderbot? Checks out.
|
|
#
¿
Apr 22, 2024 10:28
|
|