He was shopping in Tbilisi for this video, does he have some connection to the USSR? That would be a really, really interesting perspective for a tech YouTuber.
On his page https://www.thomasburns.net/ he mentions "Our family runs a small computer repair business in Tbilisi, Georgia" (under the rotary phone tile).
I think projects like this, as well as Disney's recent incredible robots (e.g. https://youtu.be/-cfIm06tcfA?si=D3SKsNekOaDQFeVU) mean it's just a matter of time until we have pretty real, embodied robots in people's homes. There will probably be some viral videos about that, and I imagine some company that can sell at an attractive price is going to make a lot of money one day.
I love how the Roy Batty monologue from the original Blade Runner movie ("I have ssen things you people wouldn't even believe...") keeps popping up. The robot acutally looks like it was made J. F. Sebastian, so making it say the monologue at the end was a very nice touch IMO.
Not sure if this is accurate, but I was once told that the voltage itself is not too bad, it hurts but will not kill you as the currents are too low. But your reaction to the shock has the potential to injure you, cut your hand at sharp edges or make you touch something else with dangerous voltages and currents.
This is true. The CRT itself holds a small charge but at a huge voltage. So the current through you would be quite high but extremely short, and will not have time to do much damage.
Now, if you open up an old TV, what you REALLY need to watch out for are the big capacitors in the power supply. Those may be only the voltage of your power grid, but hold enough charge to power the whole TV for several milliseconds. On a 28”, that’s enough to kill you dead. Luckily the leads on those caps are not exposed unless you disassemble the TV (removing the PSU PCB).
I really like his point about how our realized future of the "robot companion" is so lifeless and utilitarian compared to what we used to imagine. I guess it makes sense given how the adoption of voice assistants predated widespread LLMs by about 10 years, and voice assistants were never even close to feeling "alive" in any sense.
I'd love to see something like this manufactured at scale once the LLM ecosystem matures a little bit. As someone who has never quite felt comfortable using voice assistants, it might actually be the ingredient I need to make the whole thing feel enjoyable.
We made some "One Minute Movie" robot reality TV spots at the Stupid Fun Club, about Empathy [1] and Servitude [2], written by Will Wright, who also participated in Battlebots with his daughter Cassidy and their vegetative robot Super Chiabot [3] [4].
The one minute movies never ran due to contractual problems between NBC and SAG. We used hidden cameras, and the humans were real unsuspecting people, but I'll admit it's true they didn't accurately portray the everyday lives of actual robots in the real world: the robot's injuries and Professor Johnson's phone number were fake, and the robot waiter was fired.
The point of NBC's one-minute reality shows was to tell short entertaining two-part cliffhanger stories, which NBC would broadcast as interstitial programming at the beginning and end of commercial breaks, to compel people to keep watching TV and submit to more advertisements.
But the point of Will's stories and performance was to explore how people interact and empathize with robots, discover what their beliefs and expectations are, and probe to test how easy humans are to fool or convince to play along with real-time tele-robotic wizard-of-oz man-behind-the-curtain mumbo-jumbo.
>In 2003, NBC attempted to add a new feature to prime-time television with "One-Minute Movies." Each original movie unfolded in its first 30 seconds and ended with a cliffhanger, and then a conclusion in the last 30 seconds. The One-Minute Movies were to be used as interstitial programming between commercials and possibly where a show ends a minute earlier than its scheduled running time.
>The idea for the series was brought to NBC by John Wells (“E.R.”) and director Paris Barclay. The ten unique films were written, produced, cast, directed and filmed by assorted talent. Among the talent appearing in or lending their voices (one was animated) to the projects were Michael Richards, Tom Arnold, Carmen Electra, Bill Bellamy, Eddie Cibrian, Garcelle Beauvais-Nilon, Paula Marshall, Jackee Harry, Danny Masterson and Amber Valletta.
>With mechanical robots designed and built by Will Wright (creator of “The Sims”), the two segments directed by James Moll are comedic short films featuring fully functional robots in hidden-camera situations.
>The first, “Restaurant” features a robotic waiter. Coffee shop diners were asked if they would be willing to be seated at a table that will be served by a mechanical server. Then, surrounded by tables filled with hired “extras,” the unknowing customer is approached by a 6-ft tall, fully functional, computerized, talking robot.
>The second film, “Empathy,” also features hidden-cameras. In this case, a broken down, severely damages robot was planted next to a dumpster on a side street in Berkeley, California. As people would pass the robot, it would suddenly move and talk to the passersby. “Help me,” the poor helpless robot would say. Hidden cameras captured various reactions, everything from apathy to empathy.
>After airing only one of the films – featuring “The Pussycat Dolls” -- NBC decided not to continue airing the remaining twelve films.
After a short and unsuccessful career as a waiter, Slats eventually decided that his one true goal in life was to reproduce himself.
So we took him out in public around Berkeley, and shot some video of Slats attempting to seduce men and women off the street.
It was kind of like Speed Dating meets Demon Seed: he tried to talk people into giving him their eggs and sperm, because he needed their genetic material to reproduce himself.
>Cupid $959 Are you tired of dating weasels, hedgers, pussyfooters and wafflers? Are you ready to dispense with all the beating around the bush you're forced to put up with just to find a yes at the end of a marriage proposal? Let Cupid, the Equivercator help you cut to the heart of things and land yourself the mate of your dreams. No need to wear yourself ragged trying to win the affections of the one you desire. Let Cupid help you secure the proposal, and then you can spend your time on more important matters: You've got a wedding to plan!
The results of our research: It turns out to be a lot easier for a robot to talk a man out of his sperm, than to talk a woman out of her eggs.
This is Slats, the unemployed robot waiter, roaming the streets of Berkeley:
Side-note: I watch quite a lot of Youtube videos in this space, and while I quite liked the presenter and the project, I've rarely watched a video with more 'fluff' padding out the content. There were multiple sections of music and irrelevant images that stretched out to 30-40 second each which did nothing but waste my time. At least they're easy to skip.
I get that not everyone wants a dense information-packed how-to video - and indeed, a lot of DIY creators (Colin Furze, for example) seem to have gradually trended away from this style over time. It may therefore be that trying to please the Youtube algorithm ('all hail the algorithm') can drive video producers in counter-intuitive directions, but I really hope that this is an outlier trying to boost their video length, and not the sign of yet more things to come.
It’s also a rather traditional format outside of YouTube. I didn’t fault him for it, it’s nice to see that kind of presentation sometimes instead of some presenter talking non stop for ten minutes.
The "fluff" is really stylish and I don't mind it. What is a pity is that there isn't more of a demo of the robot, operating. It's nice "technopunk", it doesn't have to be practical. He himself points out that the plain old Echo device does the job and the bot just adds personality. So using an actual CRT just as an oscilloscope display is just fine!
good god he built this, it's his channel so he can video it however he wants. i dont understand the point of this comment. why are you questioning the video length and supposedly intention behind it? just skip it or don't watch.
Pretty sure Mark Robar has a rant about this that creators are pretty much forced (mathematically) to make longer videos if they want to be successful.
Wanting to be successful in a field that is driven by capricious technical whimsy and uncorrelated luck is generally a mistake. Do it for yourself, come what may.
Of course he has the right to do whatever he wants (as long as it is legal), but he shared it and even monetized it, so we also have the right to comment, including pointing out the parts we didn't like.
I think it is fair, although personal criticism (I liked this, disliked that). The questioning is also constructive. Just look at the replies. They are interesting, as you can see why people liked or didn't like the "fluff". It also suggests a simple, but not obvious answer to the questioning. Creators do this because people like it.
I know, right? God forbid we should share our opinions of things posted to an internet forum, on an internet forum. Whatever is the world coming to?
To answer your question less sarcastically, I watch quite a lot of 'technical' or project Youtube videos - e.g. Stuff Made Here, Superfast Matt, Colin Furze, Aging Wheels, Bad Obsession Motorsports, Matthias Wendel, Peter Sripol, Tom Stanton... and the video style in this case was such an outlier (even considering the trend I've seen in recent years across multiple channels) that it was mildly noteworthy. I can't remember the last time I skipped such large segments of a video from any of the producers I listed above (apart from sponsor segments, natch).
If it's the quirk of the individual producer, then of course it's his prerogative, no question. But if his goal is to grow his channel, such feedback may actually have value, given it might turn some viewers off.
I can't speak for OP, but I did comment how it also made my feel a bit weird, I get suspicious given the amount of shills/plants large companies will create as 'spontaneous' social media success people. Its used to create a more authentic connection to a person who then shills the brand they've been planted by.
Prevalent or not, the ide gives off creepy vibes to me personally .
> Sounds like you people just lack a solid attention span
Sorry but that's really not it, I just don't watch a video on a topic I'm interested in for 30 seconds of city skyline views with some light music..
My attention span has nothing to do with it, my patience for wasting my time does however.
Edit : Just to reiterate, I liked the video, cool idea.. Just the balance seemed off to me. It's not worth an argument to be honest. Maybe I shouldn't have said anything.
No that's fair. I didn't mean to shit on your opinion. I just really enjoyed the slower pace and tbh I'm a little frustrated by how quickly internet people talk these days.
Sure, but it's a 20 minute video, ˜40-50% of which is arguably unnecessary: it could have been a kick-ass 10 minute tightly-edited informative video. It's not about a short attention span - it's about the value of the time spent.
That’s your perspective- I hate tictok videos because they are just so ungodly short and compressed to the point of being in enjoyable - I don’t care how much information they convey. I enjoy longer form videos that indulge, it’s a better production and more enjoyable
Like your argument boils down to why watch planet earth when you could look at a few photos and read the wiki
> I've rarely watched a video with more 'fluff' padding out the content. There were multiple sections of music and irrelevant images that stretched out to 30-40 second each which did nothing but waste my time. At least they're easy to skip.
I felt the same way but didn't want to comment as it felt a bit rude, I even want to see his profile because the amount of production 'fluff' seemed excessive for a relatively low 84k subscriber count.
A low amount of videos (14 right now) and ~80% of the video views on this one, it feels weird.
All in all, interesting video and at least he pointed out the elephant in the room about merging to some GPT model.
Well, to comment on the 80% thing, that's usually how these algorithms go. Someone makes video after video gaining some to little traction. And then they make their first video that scrapes above the margin, gets posted somewhere, gains traction and becomes vastly more popular than their other videos. What we're witnessing is this person's largest video yet, i mean it got posted here, and we're all discussing it.
As for the disproportionally produced nature of it, sometimes smaller creators who have money to spend will pay producers, editors, artists to help them on their video. These people have a portfolio that they advertise and they sell their services with some personality. Either this individual has experience video editing and has created all the fluff themselves for some reason, or the person has enough money to spend they dicided to hire a professional editor, who has made all these sections and spliced them into the video. Possibly the creator gave the editor a minimum length to produce and not enough real footage. Possible this exact thing happened but the creator and editor are the same person.
Youtube has changed a lot in the past 5 years. There a huge industry of video editors and artists who support the professionalization of even first time youtube creators, as the bar is raised ever higher and higher to be heard above the noise
I was going to suggest that this might have been to get the video over the 8 minute limit which triggers mid rolls adverts. But the video is almost 20 minutes long..
He did this for fun and made a video about it. Based on his posting history it doesn't seem that he is playing anything just wanna make sure that his work gets visibility (thumbnail)
Please don't pick the most provocative thing in an article or post to complain about in the thread. Find something interesting to respond to instead. [...] Please don't complain about tangential annoyances—e.g. article or website formats, name collisions, or back-button breakage. They're too common to be interesting.
The trouble with such 'side notes' is they end up eating the thread and are effectively completely unrelated to the thing being presented.
Just run SponsorBlock or similar then. If you still want to support creators, you can tune the plugin to only skip filler, or as I do, just prompt with a Skip button and I can choose to jump as needed.
We don't talk about this enough but those kinds of techniques are just manipulation techniques, the same way movies and tv are designed to affect our senses and emotion (rising music, etc). They exist to short-circuit our logical minds and appeal to the emotional parts of our minds to create attachment and familiarity and sympathy with the content and creator. Which is fine if that's what you're asking for.
I dislike that you're being criticized. In a technical forum we should be demanding more efficient uses of our time and less manipulation and more facts-based presentation. I feel a little sorry for people who grew up in the youtube influencer age. These techniques are totally normalized for them and I don't think they realize how deeply manipulative your average popular youtuber is.
A lot of us just want a fact-sheet kind of thing that is neurodiverse and "busy technical person" friendly. The marketing-style media is really meant for executives and potential customers. I always shy away from those kind of presentations, find them hard to focus on, get impatient, and just google for a fact sheet, credible review, or technical reddit posting or other text-based info. And if there is video, it should be justified. Like a short clip illustrating something that is hard to explain or appreciate otherwise.
The other big bonus is how easily I can copy and paste these text items, make a bulleted list, etc to share with other busy people. "Hey Boss, check out this robot project, I think we can try that here," is going to be an easier sell with a page of text and some photos than "Hey Boss, here's a 40 minute video of a robot project."
I can't stress how badly the move from a text based internet to a video-based one has badly hurt people like me. I'm autistic with adhd and the constant demands to "hey just watch this video if you want to learn this technology," is just too much. I seek out text tutorials with screenshots or code snippets or whatever. I think we're becoming what happened to the boomers. That is to say the rise of TV, TV news, etc moved them way from books and newspapers that were far less centralized and had more potential for merit, variety, and lower barriers to entry. For someone like me, I could never be a slick youtube video producer, but I could write a good hack-a-day article. Why should the former have more merit? Why should we be applauding slick production techniques outside the ability of nearly all technical people?
I'm sorry if this rant is going odd places, but honestly, the mass videofication of all things has incredible downsides, but is incredibly profitable hence we're stuck with it, but we should be able to call it out if we are displeased with it.
I agree with you in general but you might have missed that this particular video is about an art project. The robot doesn't really have any practical purpose at all. The technical details are not the focus. The robot isn't that interesting from a purely technical side of things.
The video is really about the sense of nostalgia for a past where we had a more optimistic idea of a future. The emotional level is the point of the video.
So please call out the mass videofication of things but this is not an good example of it.
They’re also just techniques in art where the artist wants you to feel something specific rather than just dumping their creation on you without comment.
This is art, and entertainment. A guy made a cool project and he wanted to create an artistic and engaging video to show it off, not a fact sheet documenting how to reproduce his efforts.
If you don’t want to consume videos for entertainment, then don’t watch it.
Thanks; and I'm totally with you. Most informational topics (except perhaps obviously mechanical topics - e.g. changing an inner tube on a bike) are by far better dealt with in written/pictoral form. Funny you mention boomers - I remember my Dad, many years ago, complaining vociferously about how much TV documentaries were padded out with extraneous content versus those on radio. What goes around, etc. :)
---
Even more of a tangent, but regarding the march of videos on the internet, I have been surprised by how much I've enjoyed (and even derived value from) TikTok, which I would have expected be the worst of the worst. But thanks to its nature, it's both far more efficient at enabling discovery, and almost always far more concise, than Youtube. And oddly, TikTok has done a lot to drive off-internet pastimes, thanks to discovery of new books to read, recipes to shop for and cook, films to watch, exercise routines to try, etc. It's so weird so write this, but I find (my little slice of?) TikTok unbelievably wholesome.
A lot of boomers didn't like the TV-ization of the world. For a long time Neil Postman's "Amusing ourselves to death," was a top rated academic commentary on how TV is ruining society by moving us more towards sensationalized news, easy manipulation, centralization, etc. Its sad to think that war is long over and TV has won. And now that same war has been lost on the internet with things like youtube.
I'd argue tik-tok is a new artform and not TV or youtube like at all, and works pretty well for what it does. It doesn't ever, at least for me, take the place of written materials or photo tutorials or technical articles, but youtube is doing its best to replace those items. Tik-tok is more a unique media that wants to create a new space in your life and badly constrained by its short video focus, but youtube trying to eat the internet by eating up traditional spaces like written media.
(For me, a wonderful example of a channel that almost perfectly balances information and entertainment. One of only ˜2-3 channels that make me want to drop what I'm doing, whenever I see a new video has been posted.)
Sorry for the tangent dang but since this video mentions Alexa I gotta say that as an Australian I find the ubuiquity of Amazon in the US perplexing.
To me, Amazon is a total mess. The account management is terrible. It has all these crazy region restrictions when you try to buy books, and I just assume that whenever I buy something from the store I'm getting scammed. Everything on Amazon looks like a scam.
Then I hear people in the US buy their groceries on Amazon and use Alexa. It's crazy, what am I missing? The entire company seems like it's strung together with duct tape. I have zero trust in Amazon as a brand to do anything right.
Also in Australia, agreed about Amazon being a deeply scammy-feeling website. The only reason I'm subscribed is because I love Wheel of Time that much and I'll be damned if I'm going to wait a minute to watch it when it premieres (very convenient time for me in Australia, 10am). Then I figured well, if I've got a Prime account, I might as well check and make sure something I'm going to buy isn't on Amazon with free shipping for a lower price, and it turns out that happens very regularly. I see the value now, although it's probably not worth what I pay for it as the savings are small. But it's well worth it for me because of Amazon Prime Video, particularly Wheel of Time.
It's actually funny, ever since Wheel of Time released and we got Prime to watch it, it's not just that we started buying from Amazon instead of individual stores, but we've actually bought a bunch more stuff over the internet than we would have before anyway. Stuff like air blowers for cleaning out electronics, verifiably safe toys for the cats, spring-loaded arms to hold phones up, a Kindle to read ebooks. The Kindle is the other thing I'll give credit to Amazon for being just very good and very good value like their streaming service, the Kindle is a very good device. Annoying DRM on my books, sure, but I've got them backed up for when it gets cracked.
Amazon has been going downhill for a long time, when I first started using them it was quick, it was easy, and I knew what I bought was going to arrive within a reasonable time frame. I bought Alexa devices, I bought their TV box thing, all was good.
Fast forward a decade or so and it's exactly as you say, finding anything reputable on their website is an exercise in frustration and unpronounceable randomly generated "brands" which have paid to be pushed up the search results. Alexa works maybe 50% of the time when I try to use it now (mostly I've switched to HomePods and Siri), the devices with screens are a constant flurry of pointless distractions ("There's nothing on your shopping list"... cool, maybe just don't tell me about that then). FireTV is a barely functional shit show which often takes multiple seconds to validate I've entered the parental control PIN correctly, and can take anything up to half a minute to fully populate the home screen with adverts.
The thing that really took the biscuit for me was when I cancelled my Prime subscription recently. Now whenever I try to buy something from Amazon I get interstitials trying to dupe me into reupping, often cryptically worded such that it looks like I'll be paying £10 for delivery if I don't. This is probably a good thing as it just encourages me to go elsewhere.
I've used Amazon for ~10 years. It has certainly gotten worse, but it isn't bad.
I live in Seattle. A lot of random things I can get shipped the next day (sometimes even the same day!) for free (I pay for Prime, but there is no additional cost), like if I need a USB drive, compost bags, or dish soap.
I'd prefer to buy things from Amazon vs other sites simply because I'm used to Amazon. I have an account; they have my address and payment information. I just need to click a few buttons and have what I need the next day.
If something goes wrong, customer support is still pretty good in most cases, although sometimes Amazon has dropped the ball. Amazon is great with returns -- most of the time, I don't even need to box up and label my return. I just need to show up at an Amazon Fresh store or some other dropoff point (there are many in Seattle) and hand them my item/scan a QR code, and that's it.
People have complained about fake products on Amazon. I've bought hundreds of things from Amazon, from CPUs to consumer electronics to batteries. I'm sure I received some knockoff at some point, but I've never noticed.
I would gladly switch away from Amazon if something better popped up, but I don't see how any alternative could be better without eventually becoming Amazon again.
I'm in Europe and I only buy stuff that is sold by or delivered by Amazon. Never had a problem with any scam in years and hundreds of products. The prices are good. Delivery to pickup locations is very convenient. 30 day return policy for any reason and it's very easy and free to actually send it back.
I see a lot of hate for Amazon on HN, maybe it depends on the location. I never had a problem with them and neither did any of my friends.
It is rare, but some fake products due get through even if you always get things fulfilled by Amazon. I've never had the problem myself in years of buying things including PC parts via Amazon, though a friend has. (we are both in the UK)
Due to the way things are co-mingled in the warehouses there isn't really much distinction between “sold by Amazon” and “fulfilled by Amazon”: if something has the same product code the system assumes it is the same thing, so the picker is sent to the closest one on the assumption that they are the same thing. In my friends case this resulted in him getting a counterfeit SD card. He noticed oddities on the packaging before actually opening it, so we don't know how far it was from spec. The return process was smooth and the right product was received the next day, but it was a faf he could have done without and if he'd not noticed he might have had a shock later if the card corrupted itself after writing a small amount (many fake storage devices pretend to be far larger than they are and stop writing (or worse: data is silently corrupted) once you approach their real size). Luckily it wasn't needed urgently, so the extra day wasn't a massive inconvenience.
Same here - my experience is that Amazon works very well in Europe. We picked up the Amazon habit when we moved house during the pandemic, and must have bought hundreds of times since.
Second this. I'm putting together a new computer and have bought some parts from Amazon and some components from other sellers. The Amazon purchases have been easy (not perfect of course), while the other purchases have been annoying or confusing (cancelled orders, unclear stock availability, confusing tracking, etc.).
Amazon is just so much more reliable than other retailers. Is it perfect? Of course not, but it is much less of a hassle than most others.
for what its worth im also in europe and amazon isnt even in my country, if I want it I have to buy from neighboring country and it takes many weeks to months to arrive. Seeing as the interface and product selection seems tailored to getting you to buy the most flimsy knockoff imaginable for every product there is no reason to buy from them
Amazon isn’t great, but they are still better than their competitors for me. They are fast, have reasonable prices, and good customer service. AFAIK, I’ve never been shipped a counterfeit item and they’ve never given me a hard time when I want to return stuff.
The Alexa stuff is a little weirder for me. It’s gone in waves of being amazing and being terrible. I just noticed that for the last six months they’ve stopped with the “by the way” bs after I ask for the time, so that’s nice. A month from now they might turn the dials and make it annoying again.
Their e-Ink Kindle hardware is very nice and they have more ebook titles than anybody, but the DRM is such a downer. I’m cheering for the DeDRM people to crack it, but so far KFX seems to be holding…
My biggest Amazon complaint is their review system. It’s so gamed it’s almost useless. I pretty much only look at 3 star reviews because 1 and 5 are rarely fair or accurate.
It’s hard to overstate the scale of their offerings and ubiquity in consumer mindshare in the states. Comparatively, we’ve had Amazon in Australia for what, seven years at this point?
Possible tangent enabler, but as an Australian i've felt the same + have the same perplexion.
I've bought I think three things from amazon just to try it out. My lesson is that the entire site has a toxic interface and my credit card got falsely charged sometime afterward, which I cannot prove but I'm pretty sure was related to my use on their site. It looks like a cheap scam bazaar.
I don't understand its apparent US ubiquity...but then I don't understand the desire for something like Alexa either.
Canadian here who lived in the US for a few years. One thing I discovered while living stateside is that there are a lot of features in the big tech companies services which only exist for American users. When I put a US SIM card into my google pixel and got a US number, a bunch of stuff I had never seen before started appearing. Similarly the number of products available to me on Amazon grew by what felt like at least an order of magnitude, and I could have nearly all of them on my doorstep in days, sometimes on the same day I ordered them. And that was in a small city in a flyover state.
Agreed, the ubiquity of this stuff is hard to understand from the outside but for me it was extremely clear from within.
> I don't understand the desire for something like Alexa either
What people value them for varies. I had smart bulbs anyway (so I can vary between soft low light for watching TV and bright white for crafting, etc) and a compatible remotely control able heating control, I wanted to be even lazier than needing to pick up my phone to change things, and a pair of them were dirt cheap in an offer. Setting of timing reminders in the kitchen is pretty handy too. I don't use them for much else them that though. Occasionally I'll have an audiobook play that way via Audible, but the speaker isn't great for music.
I looked into open source variants for simple home automation like this. I look back every now and then and, while things are improving at a pace, it still seems like a chore to setup and potentially flaky (not that Amazon devices don't need resetting almost monthly because they are a bit flaky too) and more expensive in terms of getting good microphones/speakers/etc together.
[All the above would apply the same would to Google's offering; I went with Echo devices because they were cheap at a time that my wallet control was weak, and you can call them “computer” rather than “brand-name”]
I don't know of anyone personally who buys all their groceries on Amazon. However, I will buy a few some canned/bottled stuff on Amazon if it's hard to find in the stores around me, or if it's cheaper (sushi ingredients like nori and rice vinegar). I also think I only know about 3 people personally who have Alexa.
I think there are fewer region restrictions here, maybe because they were a US-first company. Their account management and customer service is certainly lacking.
It's easy to buy groceries on Amazon if you live somewhere that Amazon delivers fresh groceries from a grocery store. I did it for a few months. I go online, click a few buttons, and a few hours later, someone shows up to my door with bags of groceries. It works well, especially if you can't leave the house for whatever reason.
My experience hasn't been that my Alexa or that Siri tries to control me. I use it to set timers and play music and that's about it.
On the generative AI front, OpenAI/Microsoft doesn't control me either. I use it as a coding assistant instead of searching StackOverflow, or when I want to brainstorm lists of other ways of writing the same thing when I don't like what I've written and am not sure the best way to improve it.
Could just be you're approaching the tools wrong? But I can't imagine what you're doing where they're controlling you.
You may be using a stricter definition of "control you" than GP was. Perhaps the better word would be "influencing" you. That can mean something as subtle as reminding you that pizza exists and it's yummy and mmm isn't pizza just so yummy? Basically what commercials have been trying to do since forever.
Even if you don't believe Alexa and Siri are recording everything you say, I think it's pretty clear that asking them to play music gives them some power to influence you. You're telling them your musical tastes, and you're (presumably) letting them pick some songs for you. (At the very least, by simply not having the song you asked for available.) One reason, perhaps even the biggest reason, that Amazon and Apple created those products is to sell you more shit, and both of these give them some power to influence your buying habits. Perhaps that power is tiny, and the tradeoff is worth it. They're also trying to influence who you buy from, or what services you subscribe to. Vendor lock-in and all that. Is that not them trying to influence you? Perhaps even to control you?
If you agree with me what they said about "control" and "lending my soul" is wrong, why pretend otherwise? If all they intended to say was that they don't like that if someone asks Alexa to play a playlist, that Amazon can choose which songs are on that playlist assuming its one Amazon put together, then they would have said that.
Instead they said a different thing. No, that's not an example of control. It's a feature I paid for!
I’m not seeing what me being downvoted has to do with anything. My response is exactly what Aldous Huxley was saying.
The viewpoint has the potential to trend toward the insane if it is the unconsidered default, for sure. You can’t go around thinking everything that happens to you is a grand conspiracy.
In your situation, though, you are informing systems (that are larger than human ability to productively think about) on how to perform your job.
You do this because those systems allow you a local competitive advantage — you get more done for less work. But your only real hope is that it takes them longer to replace you than it takes you to escape.
I have no idea why you were downvoted (wasn't me) and wasn't referencing it. I was referring to the circular logic of your reply to me where you imply that the very fact that I don't realize I'm being controlled is evidence I am controlled.
And the reason I use Alexa to set a timer is because my hands are occupied in the kitchen. You're overthinking it. Nothing to do with giving up control.
The reason you use Alexa to set a timer is because Amazon as a company wanted a presence in your house, and they internally understand and regularly use the term operant conditioning. Otherwise you’d be using a timer.
I actually think we are mostly on the right path, with the greatest exception being the current cost and overhead required to train the very best language models. But we are already seeing smaller, fine-tunable, user-deployable models.
I think (and hope) it'll track a similar trajectory to 3d printing:
1. Expensive systems only approachable to large corporations
2. Lower-cost but affordable as a business expense to cottage industry
3. Affordable to hobbyists, but requiring domain expertise to operate
4. COTS plug and play solutions for the wide audience
Right now we are at (1) with the huge scale LLMs, but scaled models are runnable by hobbyists (2-3) and we are at (4) when it comes to things like image generation and music splitting.
In the end he suggsted that further development of AI models will help with such projects. I think there will be a lot of opensource personal AI's available in several years when reasonably cheap accelerator chips become available enough.
We're on the correct timeline. It's just that people are extremely reluctant to embrace technologies which make part of the ego redundant.
For example, an AI slave (lit. robot) is useless to you because then you have the master-slave mentality so half of you is still a slave. Once we're actually OK inside we won't even think we need technology.
This is a cool project but I strongly disagree with its premise that we would need machines with "more personality".
I don't want that.
I want machines that are predictable; not "intelligent" ones that may do one thing one time and another thing another time, for no discernible reason. I expect machines to be deterministic.
Why not both? Someone might want to talk to a hologram of their celebrity crush in a skimpy outfit that always talks in a submissive way and sultry voice, while someone else might want a disembodied voice without discernible sex.
This is the classic HAL vs the Star Trek "computer" question.
I think initially things like Alexa, Siri and Google Home were so much verbose and personality based and now just barely speak and don't do any fun personality stuff, outside of kid-centric features like "guess the animal game", bedtime stories, etc. I think people are just too busy or in a hurry to deal with personality from technology. In fact, Cortana has entirely been cancelled and replaced with a faceless and nameless Bing Enterprise Chat.
Another case is how stoic and boring and lifeless ChatGPT is compared to some of these more personality based AIs out there. People just want the facts and not to be bothered. They actually don't want a virtual friend. Things like MS Bob, Cortana, and various avatars and personality-based agents keep failing compared to simple and fast interfaces. I think that'll probably stick with robotics too. I think the Star Trek "computer" will win out every time. People just want to feel empowered and have a minimal of their time and effort spent.
Siri, Alexa and Cortana all sucked. They simply weren't smart or robust enough for the whole personality thing to work. This has clearly changed. LLMs will literally perform better with emotionally fused text. https://arxiv.org/abs/2307.11760
Bing may be faceless and nameless but i assure you, it's definitely not personality-less.
Character ai is very popular and is based on personality infused LLM instances. There is clearly a big market here. That Open ai try their best to train the personality out of their LLMs by default is not indication there isn't.
I was reading some chatGPT programming workflow Twitter thread the other day and the guy was unusually polite with the thing. Like, does saying "please" or "thank you" actually improve its performance or is it just tech roleplay?
Yes, in the early days of ChatGPT I had noticed that polite conversation leads to less adversarial responses. In retrospect it does seem logical that a language model would choose more amenable responses and be more willing to help when spoken nicely to. I do not know if that behavior has been programmed out of it by now, I simply talk nicely to it with please and thank you out of habit at this point.
The person who started the thread said "I don't want that." Explaining that other people do want that is a perfectly valid response. If you would like to further elaborate on why you think sugar is the right analogy here that would be very welcome too of course.
If you mention "personality", in the context of robots, the first thing that comes to mind is Marvin the paranoid Android, from Douglas Adams. He famously built Marvin with 'Genuine People Personality', though it seems to do little good for him, or those around him for that matter.
I generally agree but part of the purpose of machines like AI is unpredictability. For example that AI girlfriend trend started exactly because LLMs can now produce more varied output than a Skyrim NPC* while making more sense than a random word generator.
*Of course not including the new LLM-based Skyrim NPC mods
I bet you lose out in the marketplace. The same selection pressure that drives neoteny will drive the evolution of hardware-with-a-personality. Compare Bratz dolls with Barbie dolls. Or even a 1960s Barbie with a 2020s Barbie, to a degree.
Yes yes, the market rules all and is always correct, amiright? Some of us don’t want the world ruled by the biggest company that won the game of capitalism.
Even if every robot was artisanally hand-made by Sandbenders in deepest Oregon, I predict the cute robots would still be more popular. It's not capitalism that's providing the selection pressure, it's our own evolution.
Given time, we'll find out. Maybe I'm completely wrong.
I wish the people of the past hadn't forgotten to tell us that the robots of the future would collect our private data and store it on the computers of some gargantuan corporation so that they could profile us better.
Yes but it seems they can't do anything about uBlock Origin. I have it and only found out about the blocking spree from YT videos complaining about it.
It’s worth noting that the ad blocker-blocking is a slow roll-out. People with the same OS and browser/ad blockers will have different experiences. I’ve also never been blocked from watching but I do commonly see “Youtube is not free” dialogs pop up because I’m using an ad blocker. In my case it’s AdNauseam[0] which uses uBlock Origin behind the scenes.
For me, yes. I do generally make sure that content creators that I like get paid, e.g. via Patreon. I still buy all my music, primarily on Bandcamp. I subscribe to streaming services and haven't sailed the seas to find movies for years.
However, YouTube I will never pay for. Google has lost my trust, and I fear that they are in the process of destroying the internet and turning it into a walled garden riddled with unblockable ads everywhere. I use uBlock to block their ads because (1) I don't want to watch them, and (2) I want to actively hurt Google financially by spending their compute resources but not giving them ad impressions. I make sure that the content creators I follow get paid otherwise, so I think this is perfectly morally consistent.
And for the longest time I didn't even block Google ads.
It was kind of a deal between them and me: they serve me ads in exchange for excellent free services.
Lately however the quality has sunk through the floor and the annoyance has skyrocketed.
It is almost like they don't care about the ads pr se, it seems they only want to turn the lever up until I start subscribing. And it won't surprise me a bit if after a few years it is pay + ads.
I use Google the way I use Microsoft, when it is convenient for me. They have broken so many implicit contracts with everyone of us that none of us owe them anything except what the law says.
------
The ads were always rubbish for me (I hear other people had scarily good ads) but "my" Google ads hit a few times in a decade. (Facebook was however scarily good: I hardly use Facebook or any of their properties at all, and still every time I log in they seem to have something I am tempted to buy).
He’s stuck in childhood. Computing should be ubiquitous in my opinion and completely non intrusive. Like the Universal Computer in Asmiovs “Nine Tomorrows”. I don’t ever want a 1980s goofy gremlin bot. I want the computer to completely disappear into the ether and only respond when spoken too. Less technology in sight not more.
Edit: and if I need a physical robot I want its form to follow its function, not an outdated notion of a future we’ve already passed.
Edit2: I think Alexa is the perfect model, but I don't have one because privacy is not taken seriously where I live (United States). Computing as a universal basic service needs to be wrestled away from the capitalism-winning megacorps, and that's about a century off.
He's got a really fun video with an old Soviet games console I particularly enjoyed.
https://www.youtube.com/watch?v=9mqXT-zj47s