fafferlicious

There’s a lot of people here unable to recognize how important attention span is for deep, meaningful learning.

Way too many people saying “ATTENTION SPANS NOT INTELLIGENCE DUHMASSES” and failing short because they don’t have the attention span to think of implications of decreased attention span.

Long term memory is built on repetition and attention. It is entirely plausible that short form media makes you stupider - by impacting your ability for deep learning

It’s not even strictly what capitalism is about. It’s some stupid bullshit interpretation that came out of the University of Chicago economics department.

Seriously, go look at Adam Smiths wealth of nations. The only mention of the “invisible hand” is so different from what is taught in economics now

… by directing that industry in such a manner as its produce may be of the greatest value, he intends only his own gain, and he is in this, as in many other cases, led by an invisible hand to promote an end which was no part of his intention.

It’s not that we need no regulation. It’s just trying to say that if we set things up correctly, we don’t need to worry about people pursuing SOLELY their own personal gain. Because the market seeks out the “greatest value”, which is not just about money. It’s also the value to society as a whole.

Instead, we got the fucking bullshit from Chicago saying that the only / best way of measuring value is by profit.

As someone who has never done any serious coding work or collaboration, hadn’t touched Linux in 18+ years and am really only fluent in windows and Mac, and with limited time to get up to speed, I fucking love opinionated guides.

Tell me exactly what to do to get it up and running. Let me learn along the way, but don’t expect me to be able to read and understand the pros and cons of lets encrypt vs other solutions.

I simply do not have the requisite base knowledge to make informed decisions on this.

If you wanna leave it up to me how to do a golden gate assembly, quick change reaction, or a gibson assembly, I can handle that.

Understanding the nuance of docker networking, reverse DNS,maintaining SSL, and just generally how to make it so I can use a hostname and not an IP address to access my services locally is something I want to learn. Eventually.

It’s not that your critiques of guides are invalid, but they may just not be structured for general learning. And their target audience is temporarily ignorant fools like myself.

A good journalist does not just write for themselves, they must also write for the audience as the audience. They are the readers proxy and ask questions on behalf of the reader.

Imagine a generic shmuck that has no idea ctrl +c and ctrl +v are hot keys.

Now. Do you think they have any concept of what type of duplication tools exist? Do you think they’d just want to use this “super cool everything” software that AI is billed as? After all, it’s supposed to be smart right? They say it’s the next best thing and it’s almost like magic.

Look, I’m not saying it’s a great article. But given all the bullshit hype regular people hear about AI. Is it really unreasonable to think “copilot, help me find duplicated files in my one drive” would be something good old Billy would try?

Maybe it’s not a great article because there are better ways to de-duplicate photos. But that isn’t the fucking point of the article. The article is “look at how AI still fucking fails at basic shit we expect it to be good at.”

And for that, I thank the author. We need way more of that.

Legitimate question.

How is the act of an AI company downloading a copyrighted work and adding it into their “dataset” to generate a summary different from an individual downloading a copyrighted work and adding it to their “dataset” and writing a summary?

Both instances require consuming the material in some way. Both instances generate something new, transformative of the original work.

Why do AI companies get to torrent the entirety of human knowledge, but if any single person does it… Well we know what happened with Napster. Limewire. Kazaa. Megaupload.

Because to me, that hints at a flaw in your logic. AI companies are violating copyright. They had no permission to consume the copyrighted works.

There’s definitely hope that it synergises with what’s going on due to the steam deck.

Personally, I can’t wait to buy two controllers and the machine. The flawless experience of the deck is amazing. And because it’s Linux, I’ll just install YouTube, jellyfin, any app as a non steam game and I’ll have the perfect smart tv appliance.

Stream games, play games, run any program I want through steam big picture - I can’t wait to bury my Shield.

I’ll never have to connect a tv to WiFi again. I’ll never see a fucking ad for anything on my TVs home screen again. With KDE connect my phone is a remote. I’m so fucking pumped.

Power is a function of the size of the difference you’re testing for. Idk how to really calculate it accurately, but this site would suggest to discriminate between 7.2% and 7.4% you’d need over 300,000 participants

I’m kind of skeptical of that, but I don’t know enough about clinical trials to know that.

I’m curious if you have recommendations on how to structure or keep the notes. I find that I struggle reading technical documentation or how to structure notes so they’re easy to refer to. Have any tips or guides you can share?

Do not mistake the opportunity cost for inviability. Billionaires are not lining up to build one with their money may be a fact (I don’t know), but even if it is, it is entirely possible that the rate of return is just too low for them to consider investing in. Why make 3% EV YoY when you could get 5% EV YoU? Almost double the rate of return!

Not so easily proven.

It’s really not some mythical user, man. I just laid out the two main methods we were comparing of updating firmware. One process requires literally 0 user intervention after device setup. The other require the user to intervene. In what world is the process where the end user has to do literally nothing to get updates inferior to any process that requires intervention? It’s not that the device has Internet that’s the issue. It’s that there’s nothing respecting privacy! We’re on the same side but you wanna…what…make it illegal to do OTA updates because internet connectivity for devices is intrinsically invasive somehow? I just don’t get it. Your position doesn’t make sense.

Make an app

And now the app pulls gelocation data of the user and takes the data on the device and uploads it through the app because no privacy laws.

Why does using a USB stick to flash the firmware equate with chewing your own food, but using an app to update the firmware not? They didn’t compile the app from code!

there is no compelling use case for constant internet connection for these types of appliances

I literally just gave you one. The ability to push out OTA updates for bugs greatly reduces the complexity of maintaining technical support and development to support legacy features. I don’t know, maybe instantly patching a critical Bluetooth vulnerability to protect users privacy via OTA updates is compelling enough ?Do they have to do it that way? No. The device could just get it via BT from the phone app. Basically the same thing as the device itself doing it.

But the privacy issue still remains even if we remove WiFi from the fucking vacuum and then just let the phone app have WiFi access instead.

Did I make my point better? Removing internet access to the device, doesn’t remove the privacy concern. It just moves it somewhere else. And yes I know there absolutely is a way to do firmware updates in some privacy respecting way, but if you maintain that a 0 step process is less preferable to the majority of users than any process that requires multiple steps… Then I don’t know. Interact with users more? There’s plenty of boomers peck typing their way around the internet.

I admit you have points, friend. I share your concern about privacy. But come on… We got where we are in part because the general users don’t give a flying fuck or even think about privacy.

The sheer fact that internet access exists for them indicates there’s a use case. And it’s obvious. Over the air updates are objectively a better user experience. Maybe even the best. For the average consumer that doesn’t even know to use ctrl + c, what do you think they’ll choose: dragging out the USB stick/cord or just having the device take care of itself?

The problem isn’t that the devices have internet access. It’s that we don’t have appropriate legislation governing protection of privacy that would keep this kind of snooping in check.

People don’t sympathize with their trash when it ends up in landfill.

We all have our hard lines, and I don’t fault yours.

However for the scrollers passing by, they have a lifetime membership of about ~$45 USD. I found it worth it. It was super easy to get used to, and I honestly can’t stand any other launcher now.

I like it a lot and would hate to see it die out, so it was easy to justify for me.

The other consideration it has is on net luggage weight. Yeah, they’re still checking bags for “free” but they’re smaller and lighter - saving fuel by increasing efficiency. And that’s also a good thing from a pollution point!

Because it’s not. The base architecture of how it works is by probabilistic word suggestion. That isn’t thought.

We have a concept of self. We understand our place. We can interpret and respond to entirely new situations. LLMs routinely fail that. They regularly fall into local minima that keep it on the wrong path, and I’ve personally seen them just… Get lost in the weeds and swing back and forth based on what you tell it.

Give it a protein sequence and tell it to calculate the pI. Then tell it it’s wrong. “Oh my bad yes you’re right it’s {whatever you said it was}.”

Tell it you lied and that the number you said was wrong, and it turns up saying “Yes, you’re correct, the pI is {original value}” - that is objectively false.

That is not the behavior of something that thinks. That’s the behavior of a simple probability model updating priors and weighting things differently by the most recent information you gave it.

LLMs are soulless, brainless, thoughtless word generators. And they have some uses.

Why does everyone take statements about a population and then try to use a specific example to disprove it?

I’m talking about generalities. Look at who the majority of boomers vote for. Spoiler, it’s not democrats. I don’t know your grandpa. I don’t know what anyone’s grandpa did. But collectively, generally speaking, the stats say our grandpa’s share blame for voting R for decades and at the very least, for voting for corpo neolibs every primary.

Because hey, guess what, it’s not the new generations that have the highest participation in primary votes.

The boomers have held the levers of power longer than any other generation, at least in America. It is their fault because they run the companies and they run the government.

don’t upload it to the internet!

or use a smart phone

or corporate searches that track you

or go to any website with ads - they track you

hell don’t even search the internet! your ISP tracks dns requests

or use a modern tv that tracks what is on your screen

or you can do custom phone from - just unlock the bootloader, root it, and install! then just setup pihole/adguard/self-host everything

it’s simple, for privacy just go live in a yurt in the woods to not be tracked 24/7

You’ve definitely given me something to think about - evaluate if even 110 million would have prevented or given us another decade before we hit +1.5c.

However, your Bangladesh stat is absolutely meaningless and misleading. It seems impressive at first glance, but it’s not. The proper context is global CO2 production. In 2014, 35,000 million (or 35 billion) tons of CO2 were produced. And that’s just fossil fuels. And that’s more than a decade ago. I don’t have the numbers, but I suspect it’s even more.

110 million / 35,000 million = 0.3% reduction

“The Truth” is not self-revelatory nor is it self-reinforcing, particularly for a lay person.

It is not self-revelatory, but there are objective truths. If a lay person lacks the expertise to understand, they should defer to experts - not politicians or pundits.

Falling for propaganda is a reason, but it is not an excuse. The electorate has a responsibility to be informed.

You have to ask, you haven’t bothered to look.

I’m incredibly proud of what has been happening in my home city of LA. That’s what we fucking need everywhere. Burn cities down until things change.

But fair point! I was being more rhetorical and less literal. But that’s my miscommunication error. My question wasn’t to say they don’t exist or haven’t happened. I asked it to highlight that it isn’t enough. That for the magnitude of what is happening and its importance, the response is impotent and not proportional.

The world is increasingly on fire (almost literally). I’m living in a downtown metropolitan area minutes from city hall and protests are not daily.

I don’t believe an entire generational cohort is irredeemably stupid.

Nor do I. I never said that. I said I blame them for their willful ignorance and their decision not to prioritize climate change politically.

My position is simple. More could have been done, and because of that, we share blame and responsibility - however small. This is why I also blame myself.

Anyway, I think we’ve kind of hit a natural end. I appreciated our conversation, and it’s given me some things to mull over.

Thank you❤️