Ottomateeverything @ Ottomateeverything @lemmy.world

Posts

0

Comments

322

Joined

2 yr. ago

I'm actually shocked to find how many people agree with the OPs sentiment, but maybe there's something about the demographics of who's using a FOSS Reddit alternative or something. I'm not saying everyone is wrong or has something wrong with them or whatever, but I entirely agree with people finding this valuable, so maybe I can answer the OPs question here.

I've been working remotely long since before the pandemic. I've worked remotely for multiple companies and in different environments. I am extremely introverted and arguably anti social. I tend to want to hang out with many of my friends online over in person. But that doesn't mean I think there's no advantage at all. To be honest, when I first started remote work, I thought the in person thing was total bullshit. After a few meetings my opinions drastically changed.

I've pushed (with other employees, of course) to get remote employees flown in at least a few times a year at multiple companies. There are vastly different social dynamics in person than over video. Honestly, I don't understand how people feel otherwise, especially if they've experienced it. I've worked with many remote employees over the years and asked about this, and most people have agreed with me. Many of these people are also introverted.

I think one of the big things here is people harping on the "face" thing. Humans communicate in large part through body language - it's not just faces. There's also a lot of communication in microexpressions that aren't always captured by compressed, badly lit video. So much of communication just isn't captured in video.

Secondly, in my experience, online meetings are extremely transactional. You meet at the scheduled time, you talk about the thing, then you close the meeting and move on. In person, people slowly mosy over to meetings. And after the meeting ends, they tend to hang around a bit and chat. When you're working in an office, you tend to grab lunch with people. Or bump into them by the kitchen. There's a TON more socializing happening in person where you actually bump into other people and talk them as people and not just cogs in the machine to get your work done.

I find in person interactions drastically change my relationships with people. Some people come off entirely different online and it's not until meeting them in person that I really feel like I know them. And then I understand their issues and blockers or miscommunications better and feel more understanding of their experiences.

Maybe things are different if you work jobs with less interdepencies or are more solo. I've always worked jobs that take a lot of cooperation between multiple different people in different roles. And those relationships are just way more functional with people I've met and have a real relationship with. And that comes from things that just don't happen online.

Im honestly really curious how anyone could feel differently. The other comments just seem mad at being required to and stating the same stuff happens online, but it just doesn't. I do wonder if maybe it has to do with being younger and entering the workplace more online or something. But I've worked with hundreds of remote employees and never heard a single one say the in person stuff to be useless. And I've heard many say exactly the opposite.

Yeah this is why I don't get it. If you expect me to respond instantly, you should be placing a phone call and not sending a text.

Suspicions? It literally says that in the article:

But administration officials made clear it was the first step in what could be a wide range of policy responses meant to stop low-cost Chinese electric vehicles... from flooding the U.S. market and potentially driving domestic automakers out of business.

But administration officials made clear it was the first step in what could be a wide range of policy responses meant to stop low-cost Chinese electric vehicles... from flooding the U.S. market and potentially driving domestic automakers out of business.

As much as I do not enjoy being spied on, it seems they're just banning them on the presumption that they might spy.

On the other hand, it sure would be nice to be able to fucking afford an electric car. I give exactly zero fucks about the US automakers. If they're going to keep up insanely high prices, that's their problem.

So essentially we're restricting American's access to electric cars, under the guise of danger of a fear we have, in order to protect businesses from having to adapt. Glad, as usual, that the needs of the 1% are out weighing the needs of the entire fucking country. Again.

Why not just sort me out of the Mailing-list?

Because that would cost their time instead of yours

To make sure millenials can't read your password, 𝔀𝓻𝓲𝓽𝓮 𝓹𝓪𝓻𝓽 𝓸𝓯 𝓲𝓽 𝓲𝓷 𝓬𝓾𝓻𝓼𝓲𝓿𝓮.

How would this mess with millennials? I think you mean gen z.

Aren't a lot of US's enemies closer if you go around the West instead of to the east? I think your maps backwards.

The idea that plane safety is tied to everyone together agreeing to and remembering to push a button on their devices is absolutely insane. You think that the regulating bodies that require multiple backups for every possible system also just trust that every passenger pushes a button and every flight attendant actually checks every passengers devices?

You seem to think I'm just talking about linearly expanding the vocabulary of the model, I'm talking about giving it an entirely new paradigm through which to work.

No, I don't. I know exactly what you're trying to say. But you're basically talking about trying to make a car fly. That's not how it was built and it's goals and foundations are entirely different. You're better off starting over and building a plane. Your proposal just doesn't fit within the paradigms of what was built and makes no sense.

I'm talking building in entirely new ways for the AI to understand.

Exactly. But the AI doesn't "understand" anything. In order to achieve this, you need to build something that "understands" things. LLMs don't understand anything.

Anyway, this is why no one likes pedants. If you want to actually engage in conversation, sure.

It's easy to label me as a pendant, but I'm explaining how this stuff works. You clearly have no idea, admitted yourself that you don't understand, and then keep going. You just keep spewing the same shit, but the shit you're spewing makes no sense. But you refuse to budge or engage in conversation here.

You're just talking out of your ass. You're admittedly uneducated but want to be treated like you're educated and make any sense. You don't. This is why people hate people pretending to be experts and talking about things they don't understand. It's a waste of time.

If you want to keep living in some imaginary world where this can be done, be my guest, but it's fake. That's not how this shit works. Enjoy your imaginary quest though.

I didn't say any researcher or anything had named it intelligence. Nor am I trying to be semantically correct.

Read the guys comments. He's trying to push the idea that we can "change" it's "understanding" about the things it's discussing. He is one of the people who has fallen for the tech bros etc convincing people it is intelligent. I'm not fighting semantics, I'm trying to explain to him that it's not intelligent. Because he himself clearly doesn't understand that.

I don't see any reason these kinds of relationships can't be integrated into generative AI, they just HAVEN'T yet

No, it's just fucking pointless. You're talking about adding sand to a beach. These things are way more complicated and trying to shovel these things in just makes a mess. See literally the OP.

each time you increase how the relationships interact, you're also drastically increasing the size and complexity of the algorithm and model.

No youre not. Not even fucking close. You clearly don't understand this at all.

The ALGORITHM will always be the same. Except for new generations of these bots. Claiming adding things like racial bias is going to alter the algorithm is just nonsensical.

The MODEL is the huge fucking corpus of internet data. Anything you tack onto it is a drop in an ocean. It's not steering anything.

Whats changing is they're editing inputs because that's all you can really do to shift where these things go. Other changes would turn this into a very different beast, and can't be done at the fine grained level like "race".

Claiming this has any significant impact on the size or complexity of any of this is just total hog wash and you must not understand how these work or how big they are.

You're just rephrasing the same approach, over, and over, and over. It's like you're not even reading what I'm saying.

The answer is no. This is not a feasible approach. LLMs are just parrots and they don't understand anything. They were essentially a "shortcut" that gets something that acts intelligent without actually having to build something intelligent. You're not going to convince it to be intelligent. You're not going to solve all it's short comings by shoe horning something in. It's just more work than building actual intelligence.

It's like if a costal town got overrun by flooding from a hurricane. And some guy shows up and is like "hey, I've got a bucket, I'll just pull all the water to the sea". And I'm like "that's infeasible, we need a different solution, your bucket even has fucking holes in it". And you're over here saying "well, what if we got some duct tape? And then we can patch the holes. And then we can call our friends, and we can all bucket the water".

It's just not happening.

Eh I really need to learn more about AI to understand the limits

Yeah. This. You just keep repeating the same approach over and over without understanding or listening to the basic failings of these chat bots. It's just not happening. You're just perpetuating nonsense.

These things are basically slightly more complicated versions of the auto complete in your phone keyboard. Except that they're fed hug amounts of the internet. They get really good at parroting sentences, but they have no sense of "intelligence" or what they're actually doing. You're better off trying to convince your auto correct to sound like Shakespeare than you are to remove the failings like racial bias from things like Gemini and ChatGPT. You can chip at small corners here and there but this is just not the path forward.

I don't know, maybe that would work, for this one particular problem. My point is it's more than that. Even if you go through the trouble of fixing this one particular issue with LLMs, there are literally thousands of other problems to solve before it's all "fixed". At some point, when you've built and maintained thousands of workarounds, they start conflicting with each other and making a giant spider web of issues to juggle.

And so you're right back at the problem that you were trying to solve by building the LLM in the first place. This approach is just futile and nonsensical.

Someone posted the version I saw below. In that one, the person has only typed "Fe". I haven't seen one that had "Febu" typed, but yeah, that obviously would throw it.

They didn't? At least in the version I've seen, they typed "Fe" and excel auto filled the "buary". That's the whole point of the meme.

Apparently without any correction there is significant racist bias.

This doesn't make it any less ridiculous. This is a central pillar of this kind of AI tech, and they're trying to shove a band aid over the most obvious example of it. Clearly, that doesn't work. It's also only even attempting to fix one of the "problems" - they're never going to be able to "band aid" every single place where the AI exhibits this problem, so it's going to leave thousands of others un-fixed. Even if their band aid works, it only continues to mask the shortcomings of this tech and makes it less obvious to people that it's horrendously inacurrate with the other things it does.

Basically the AI reflects the long term racial bias in the training data. According to this BBC article it was an attempt to correct this bias but went a bit overboard.

Exactly. This is a core failing of LLM tech. It's just going to repeat all the shit it was fed to it. You're never going to fix that. You can attempt to steer it in different directions, but the reason this tech was used was because it is otherwise impossible for us to trudge through all the info that was fed to it. This was the only way to get it to "understand" everything. But all of it's understandings are going to have these biases, and it's going to be just as impossible to run through and fix all of these. It's like you didn't have enough metal to build the titanic so you just built it out of Swiss cheese and are trying to duct tape one hole closed so it doesn't sink. It's just never going to work.

This being pushed as some artificial INTELLIGENCE is the problem here. This shit doesn't understand what it's doing, it's just regurgitating the things it's consumed. It's going to be exactly as flawed as whatever was put into it, and you can't change that. The internet media it was trained on is racist, biased, full of undeniably false information, and massively swayed by propaganda on all sides of the fence. You can't expect LLMs to do anything different when trained on that data. They're going to have all the same problems. Asking these things to give you any information is like asking the average internet user what the answer is. And the average internet user is not very intelligent.

These are just amped up chat bots with data being sourced from random bits of the internet. Calling them artificial INTELLIGENCE misleads people into thinking these bots are smart of have some sort of understanding of what they're doing. They don't. They're just fucking internet parrots, and they don't have the architecture to be "fixed" from having these problems. Trying to patch these problems out is a fools errand and only masks their underlying failings.

Problem 2 also shows they have no double checks on access to private video feeds. Mixing up what's being requested at any step and not reverifying anywhere after that point just reveals fucking terrible security practices.

No, this does actually sound like a solution. But it's a solution that should be scattered all throughout the process, and checked at multiple steps along the way. The fact that this wasn't here to begin with is a bigger problem than the "client library failure" as it shows Wyze's security practices are fucking garbage. And adding "one layer" is not enough. There should be several.

To give a bit better context, which I can only be guessing at by reading between the lines of their vague descriptions and my first hand experience with these types of systems...

Essentially your devices all have unique ids. And your account has an account/user ID. They're essentially "random numbers" that are unique within each set, but there appear to be devices that have the same ID as a some user's user ID.

When the app wants to query for video feeds it's going to ask the server "hey, get me the feed for devices A, B, and C. And my user ID is X". The server should receive this, check if that user has access to those devices. But that server is just the first external facing step. It then likely delegates the request through multiple internal services which go look up the feed for those device IDs and return them.

The problem that happened is somewhere in there, they had an "oopsie" and they passed along "get me device X, X, X for user ID X". And for whatever reason, all the remaining steps were like "yup, device X for user X, here you go". At MULTIPLE points along that chain, they should be rechecking this and saying "woah, user X only has access to devices A, B, and C, not X. Access denied."

The fact that they checked this ZERO times, and now adding "a layer" of verification is a huge issue imo. This should never have been running in production without multiple steps in the chain validating this. Otherwise, they're prone to both bugs and hacks.

But no, they clearly weren't verified to view the events. Their description implies that somewhere in the chain they scrambled what was being requested and there were no further verifications after that point. Which is a massive issue.

It doesn't even need to go that far. If some cache mixes up user ids and device ids, those user ids should go to request a video feed and the serving authority should be like "woah, YOU don't have access to that device/user". Even when you fucking mix these things up, there should be multiple places in the chain where this gets checked and denied. This is a systemic/architectural issue and not "one little oopsie in a library". That oopsie simply exposed the problem.

I don't care if I was affected or how widespread this is. This just shows Wyze can't be trusted with anything remotely "private". This is a massive security failing.

The idea that you think people in the Bush administration sent soldiers to Afghanistan to make money is insane, and shows me you have never worked in government or met anyone who has.

The fact that you think this is so insane shows that you have no idea how the actual finances of sovereign currency works. What'd it cost them? Numbers on the "debt" that's so astronomically high that it's a joke?

since the US was attacked and the whole world agreed on going into Afghanistan

Yeah, sounds like you "worked" too closely to this militarization. That's just blatantly false. Portions of the fucking US itself, the target of the attacks, still protested and was against going there.

War is a net negative (look up broken window theory) and everyone in government knows it.

Many huge corporations disagree, and profit off of this. Even in the early 2000s, while it was happening, Haliburton and Cheneys relationship were heavily criticized, because even if it's some "net" negative or positive, there are people that stand to make a lot of money off one side of that equation.

The point of war is to change the global order, not pad pocketbooks

There were large issues people took with many international conflicts being about money and companies lining pockets. Whether it's oil in the middle east, fruit in central America, or any of the others, there are many conflicts in the "global order" which have had huge impacts for the aggressor and their economy. If you want to try to justify each one, sure, but many points point to a trend.