guitarsarereal
@guitarsarereal@sh.itjust.works
firmly of the belief that guitars are real
- Comment on Toshiba exec claims hard drives are 7X cheaper than SSDs and will continually evolve for large datacenters 10 months ago:
With the SSD’s I can afford, there negative savings from power consumption when I have to replace them every few months. We can’t all afford EVO’s.
- Comment on Toshiba exec claims hard drives are 7X cheaper than SSDs and will continually evolve for large datacenters 10 months ago:
I mean, with stuff like ZFS, it’s a little hard to justify the outlay for all solid-state disk storage when I can build out a large storage array using HDD’s and use the ZIL+L2ARC to provide read/write speedups. Who actually cares what the underlying storage mechanism is as long as the dataset is backed up and the performance is good?
- Comment on Looking for low power devices for selfhosting 10 months ago:
Hi, sorry I just saw this. “SFF” is short for “small form factor.” It’s just industry jargon for “a small PC.” They tend to be designed to use less power which makes them a good fit for home servers. Pretty much any line of PC sold to businesses, like Dell Optiplex or HP EliteDesk, will have small form factor variants.
- Comment on Apple Develops Breakthrough Method for Running LLMs on iPhones 10 months ago:
Everyone likes to trash machine learning because the power requirements are high, but what they don’t realize is that we’re in the very first days of this technology. Every technology that got bundled together into your phone was equally as useless when it was first invented.
Literally once a week, I see some news story about AI researchers delivering an order of magnitude speedup in some aspect of AI inference. The technique described here apparently allows for a 20x speedup on GPU’s.
- Comment on Looking for low power devices for selfhosting 10 months ago:
Shout out for ODROID, their product revision cycles take too long, but when they drop new stuff, it tends to be great.
- Comment on Looking for low power devices for selfhosting 10 months ago:
Bonus: there is a literally endless supply of used x86 SFF hardware from large institutions, so unlike SBC’s, there’s no special, weird supply chain managed by an English educational nonprofit that could just suddenly decide to not sell to the public for years at a time.
- Comment on Using AI, researchers identify a new class of antibiotic candidates that can kill a drug-resistant bacterium 10 months ago:
Would you also agree it’s rude to imply that this group of researchers, who actually advanced the state of the art in machine learning, are just a bunch of ChatGPT jockeys who don’t deserve credit for their work?
- Comment on Quentin Tarantino's 'Star Trek' Movie Would Have Been a "Balls-Out Hard R" Movie 10 months ago:
Counterpoint: with some subject matter, you don’t need nuance or subtext. Hence why IB remains, in my opinion, his greatest work. It’s one of the few subjects where you don’t need nuance so the good technical aspects of his filmmaking doesn’t just wash out in all the blood and gore. All you have to do is cook up a story in the Trek universe where his filmmaking style would be an asset (hint: have the story revolve around killing fascists), don’t give him complete control, and make him work in tandem with Star Trek old hands like Brannon Braga or Jonathan Frakes and I honestly think you’d end up with something good.
Personally, I think Star Trek is good enough that it deserves more and more interesting film treatments than it’s gotten. Tarantino Trek would upset a lot of people, but if it was good, it could kick open the door to experimenting with all different kinds of styles.
- Comment on Scientists successfully replicate historic nuclear fusion breakthrough three times 10 months ago:
Yeah, you don’t get to the point of solving scaling problems without having something to scale first.
- Comment on Using AI, researchers identify a new class of antibiotic candidates that can kill a drug-resistant bacterium 10 months ago:
Yeah, so the actual law is that if you didn’t do any work and just gave ChatGPT or Midjourney a prompt and it shat out a picture and then brag to the copyright office in your application that you didn’t do diddly squat, the work effectively had no human authors. If, instead, you build a new machine learning model, tune it for your specific problem, analyze the results, and furthermore, break new ground understanding how it solved your problem, and then you write the paper, in fact, you have tons of ownership over the work.
The fact people can’t tell the difference between the two and are actually upvoting you kind of says a lot about how little most people understand this stuff.
- Comment on Quentin Tarantino's 'Star Trek' Movie Would Have Been a "Balls-Out Hard R" Movie 10 months ago:
I didn’t say give him full artistic freedom!
- Comment on Quentin Tarantino's 'Star Trek' Movie Would Have Been a "Balls-Out Hard R" Movie 10 months ago:
Because you can only tell so many stories when literally all of your characters are required to be so narrow and flat it becomes a matter of debate and discussion when they do or say anything that would make them seem like real people.
- Comment on Quentin Tarantino's 'Star Trek' Movie Would Have Been a "Balls-Out Hard R" Movie 10 months ago:
I still maintain that a Quentin Tarantino Trek likely would have been the greatest Trek film ever made (not a high bar though). Imagine something like Inglourious Basterds but set during the Cardassian resistance. Come on, that would be the greatest Trek ever made. But the rights holders have always been Trek’s biggest enemy.
- Comment on 1-bit CPU for ‘super low-performance computer’ launched – sells out promptly 10 months ago:
Education.
- Comment on Thousands of private camera footages from bedrooms hacked, sold online - VnExpress International 10 months ago:
All of the hacked systems in this article are home based systems.
[citation needed] because that’s not in the article. According to the article, attackers used automated scanning software, which strongly implies they brute-forced cameras connected to the Internet with default or weak credentials. That has nothing to do with whether or not the service is based in the cloud.
In general, cloud services have far better security than DIY systems
As a matter of fact, it’s known that the leading cloud-based surveillance system, Ring, has been subject to employee abuse and user accounts have been widely compromised via credential stuffing. In fact, Amazon is currently facing a proposed order from the FTC over the fact that they allowed abuse by employees and more or less knew for years that their lax security practices were placing their customers in danger.
Cloud based security only gets better when regulators force cloud providers to improve security, after cloud providers allow hackers to harm thousands to millions of customers.
I’m just gonna say it again: the cloud is just someone else’s computer.
- Comment on How Googlers cracked an SF rival's tech model with a single word | A research team from the tech giant got ChatGPT to spit out its private training data 11 months ago:
Huh, so simulating a vat of neurons, giving it a problem, and letting it rip is likely to produce unexpected behavior that will require many years of study to even begin to adequately understand? Who knew?
- Comment on Iowa Demolishes Its First 3D Printed House 11 months ago:
Thanks for the update! Makes a tiny bit more sense.
- Comment on Making an image with generative AI uses as much energy as charging your phone 11 months ago:
Tons of work being done to improve the energy efficiency of ML models. We’re in the ENIAC days of AI right now. I’m not sure I see the problem other than that it would obviously be nicer if we could just build a time machine and steal an energy efficient AI from the year 2100.
- Comment on Lucid dream startup says engineers can write code in their sleep. Work may never be the same 11 months ago:
The tradeoff obviously will be that since you’re not actually getting rest, and all multicellular life sleeps, it’s going to fuck up a lot of engineers in ways we won’t find out about for like 5-10 years until they start going crazy/dying/whatever. But hey, people are infinitely replaceable commodities you can just burn through like trees, right?
- Comment on Iowa Demolishes Its First 3D Printed House 11 months ago:
Well, I double checked because I’ve done a fair amount of research and the All3DP article was literally just a random example I found off Google. A realtor interviewed for Yahoo! finance said buyers could expect to pay about $15k-$50k for a starter 3d-printed home. Doesn’t sound like I’m actually in the wrong ballpark here.
- Comment on Iowa Demolishes Its First 3D Printed House 11 months ago:
They’re not a gimmick, they’re dirt cheap to build relative to the quality you get (when you’re not a pack of literal community college students using a non-load-bearing material like hempcrete, working on a learning project).
The way it’s done drives the cost through the floor, even relative to a modular house, and, let’s try thinking about it for a second, if you take a liquid concrete mix and poop it out in a line and cure it properly, how is this different or worse than taking that same liquid concrete mix, dumping it into a mold you built, and then also curing it properly? Because that’s (in very simple terms) the basic difference between 3d printing houses and building them.
Basically, it’s a lot cheaper (I’ve seen multi-bedroom houses – real houses, not haha this is a big shed that you can abuse as a studio if you lie to your county planning department – printed for under $20k) and when done right, the quality is the same as any other concrete construction. The secret is just like any other construction: you have to do it right!
Right now, they’re super rare because there are no standardized building codes available for them, which means planning departments don’t have industry-standard guidance they can draw on for their own building codes, so in most places you just can’t get them approved.
Like with any technology, there will be a gold rush and tons of idiots doing it wrong or badly to turn a quick buck on the hype, but come on. New technologies are transformative, and additive construction or whatever you want to call it has tons of potential.
- Comment on Iowa Demolishes Its First 3D Printed House 11 months ago:
Yeah, so two things.
- Standard hempcrete mixes don’t exist which raises the barrier to entry substantially. You have to make your own at this time, which means correctly dialing in a mix and testing it to make sure you got it right. They almost certainly didn’t bother. Which would have happened because they were community college students already participating in a cutting-edge building process
- You’re not supposed to use hempcrete for load-bearing applications to start with, it lacks the structural strength!! What were they thinking!! It looks nice, has that lovely earth tone, and it has good insulating properties, but that’s it! It’s a finishing material! Here’s a page from some random builder I found on Google who specializes in hempcrete construction:
Areas of use. It can be used to build self-insulating walls, roofs and screeds It can be adapted to all types of building project including new builds and renovations. It is not a load-bearing material. Consequently, when building walls, it is cast around a primary or secondary structural frame made of timber, metal or concrete.
Anyways, as someone who’s looked into this stuff before, I’m irritated they did it like this. They were supposed to figure this out before they started printing. Where the hell was their professor?
- Comment on Google caught placing big-brand ads on hardcore porn sites, report says 11 months ago:
For a lot of brands their most valuable customers are middle and upper class people, and their tastes tend to veer prudish/judgmental/conservative on these things. IE with the example of a kitchen appliance manufacturer, we think that’s a very popular/cross-class purchasing category, except for the fact that for tons of us our kitchen appliances are chosen for us by our landlords and among homeowners, the only ones who are regularly going out and swapping out their kitchen appliances are the well-off ones. LG’s best customers, and this is true of most businesses, are their rich customers.
For the most part, rich and affluent customers’ tastes count for more just because they can consume more, and many of the people in charge of advertising decisions at these companies are themselves middle or upper class, so it’s like a self-reinforcing ideological loop caused by structural economic inequality. The population at large’s opinion about whether shit like this even matters can never enter into the equation because it’s all driven on a per-dollar, not per-capita, basis.
- Comment on Elon Musk on X antisemitism controversy: “Don’t advertise. Go f*** yourself” 11 months ago:
Yeah, but the fact it brought down some powerful people doesn’t mean it threatened the system as a whole. Things like Me Too become threatening to the system when they become widespread and ubiquitous and there’s a perception the ruling class isn’t interested in fixing it up. Also, many upper class people, particularly of course women, are survivors and likely are not interested in being endangered by rape culture, so there was support from within the upper echelons about Me Too.
I’m not saying that’s a bad thing, I’m saying that’s a little perpendicular to the whole question of whether or not Twitter was some kind of revolutionary working-class institution before Musk bought it. It was an influence marketplace. Everybody used it to buy and sell influence. Including progressive movements, and fascists. This had good, and bad, effects, and we shouldn’t put it on a pedestal.
- Comment on AI won't take your job, might shrink your wages, European Central Bank reckons 11 months ago:
The best part is, the MTPE workers’ output is 100% going to get fed back into the algorithms, so it’s only a matter of time before the average error rate of the models is good enough that there’s no real reason to pay anyone to look at it.
People are, I think, overly optimistic that AI won’t eventually take their job. Why the hell do people think their boss wants to pay them plus pay for an AI? When they can, they’ll just switch as much as they can over to AI. We have a quantifiable error rate/range of error rates for most tasks, so all they have to do is create a model with an average error rate lower than, say, most people, and the case for employing us to even review their output goes out the window.
Technology development takes a long time, I’m thinking 30-50 years out here, but the point is, this is a different kind of technological development than earlier ones. Future generations might, in fact, not have jobs at all.
- Comment on Elon Musk on X antisemitism controversy: “Don’t advertise. Go f*** yourself” 11 months ago:
You do realize that’s a marketing line about Twitter, right? It’s a private, for-profit corporation whose entire purpose is to inspire users to give away data about themselves for free. They don’t care what you think, they care how to manipulate you into buying things from them. Go ahead, buy your justice deodorant to wear to the protest. They’re very scared.
Trump was one of Twitter’s biggest users and they didn’t boot him until he tried to start a fascist insurrection and, again, made it impossible not to boot him. They’re an advertising platform first and foremost, they pandered to the far right as much as to anyone else, and the right got plenty of use out of that platform even before Musk took it over.
The only way you could think otherwise was not actually having used Twitter since like, 2011.
- Comment on Elon Musk on X antisemitism controversy: “Don’t advertise. Go f*** yourself” 11 months ago:
Bigbrain businessman buys stable business that wasn’t profitable, promises to make it profitable. Proceeds to destroy company’s main line of business, complains about being “blackmailed” by business partners as they leave for making said line of business entirely untenable.
Is he like, trying to be funny right now or something?
- Comment on Exclusive: OpenAI researchers warned board of AI breakthrough ahead of CEO ouster -sources 11 months ago:
There’s a thing I read somewhere – computer science has a way of understating both the long-term potential impact of a new technology, and the timelines required to get there. People are being told about what’s eventually possible, and they look around and see that the top-secret best in category at this moment is ELIZA with a calculator, and they see a mismatch.
Thing is, though, it’s entirely possible to recognize that the technology is in very early stages, yet also recognize it still has long-term potential. Almost as soon as the Internet was invented (late 60’s) people were talking about how one day you could browse a mail-order catalogue from your TV and place orders from the comfort of your couch. But until the late 1990’s, it was a fantasy and probably nobody outside the field had a good reason to take it seriously. Now, we laugh at how limited the imaginations of people in the 1960’s were. Hop in a time machine and tell futurists from that era that our phones would be our TV’s and we’d actually do all our ordering and also product research on them, and they’d probably look at you like you were nuts.
Anyways, considering the amount of interest in AI software even at its current level, I think there’s a clear pathway from “here” to “there.” Just don’t breathlessly follow the hype because it’ll likely follow a similar trajectory to the original computer revolution, which required about 20-30 years of massive investment and constant incremental R&D to create anything worth actually looking at by members of the public, and even further time from there to actually penetrate into every corner of society.
- Comment on CORRECTED EXCLUSIVE OpenAI researchers warned board of AI breakthrough ahead of CEO ouster -sources 11 months ago:
According to the article, they got an experimental LLM to reliably perform basic arithmetic, which would be a pretty substantial improvement if true. IE instead of stochastically guessing or offloading it to an interpreter, the model itself was able to reliably perform a reasoning task that LLM’s have struggled with so far.
It’s rather exciting, tbh. it kicks open the door to a whole new universe of applications, if true. It’s only technically a step in the direction of AGI, though, since technically if AGI is possible every improvement like this counts as a step towards it.
- Comment on OpenAI brings Sam Altman back as CEO less than a week after he was fired by board 11 months ago:
Well, kind of. It’s a bad look for MS to be so heavily invested in such a dumpster fire of a corporation, so it’s okay for them it got resolved, but they would have won out more if Altman had joined them. It was the other investors, including a number of employees, who would have really lost out if the company had just collapsed in on itself like it immediately started doing.
So, sure, investor win. But MS more or less lost this one.