All these articles proposing "intuition" as secret magic dust is quite funny. This is culturally amplified by movies and media ("May the force be with you", "follow your feelings").
In reality, intuition is simply a pattern recognition mechanism that sometimes work. Entire science is basically testimony of how our intuition lead us astray and why we need to be disciplined about looking at data, evidence and crafting experiments. Our intuition has always said Earth is flat, Sun rotates around us, time is constant... Virtually every single thing in science is how our intuition (aka the primitive pattern recognition) was so magnificently wrong.
I think you have to distinguish here between your intuition, and the currently reigning common sense, which you just take for granted and which is not based on your own intuition.
> Entire science is basically testimony of how our intuition lead us astray and why we need to be disciplined about looking at data, evidence and crafting experiments.
This is not really the case. If you look into how a lot of great science is done it often starts with intuition followed by testing / experimenting in order to validate.
I suspect the scientists intuition is less reliable because the scientists are acting at the limits of our understanding. Even then it is still a critical component.
For those of us writing glue code to connect services for the 100th time or a manager dealing with our 20th PIP is our intuition likely to be as faulty?
Another way to put what I think they were trying to say -
If you have a lack of knowledge, the ‘rational’ set of things to try can be so large that it’s overwhelming/impossible to actually try them. You have to pick something.
Intuition can help there (and is commonly found in almost all major discoveries), even if it isn’t necessarily right. Since it’s still more right than not listening to it.
But then you need to pay attention and do some rational analysis to verify, and then iterate.
I missed this - body language experts the world over will tell you that body language is culturally and individually unique.
Some people share similar body language for displaying certain emotions, but they are not universal.
Further you may be able to tell what some people are feeling by their body language, but there will be many many more examples where you miss the emotion.
Worse, the ones that you can get an idea of, it's very VERY rare for you to know the reason that people are in that emotive state.
edit:
This even applies to people close to you, who you think that you know.
PowerShell should be taught in university as example of how poorly designed system devoid of taste and aesthetic looks like. It's an ugly monstrosity that makes me puke every time I try to use it. The only reason some people might like it is because they have been tortured by Windows defaults for many years and finally they got used to this utter mediocracy.
It'd be good if you could actually justify why you think PowerShell is inferior to UNIX shells, instead of going on an impassioned but unsubstantiated rant.
I think powershell is actually really well designed. Some of the syntax leaves a bit to be desired, but other than that I think it's really good. What do you think is poorly designed?
That's quite poor design for a shell! It's like saying an OS is well designed except the mouse and keyboard input leaves a bit to be desired - because they were designed for scripting first and users second.
YouTube is absolutely the business that is resting on laurels, just like Google Maps and Gmail. Sometime I wonder if these products have any real active development teams at all besides ads. YouTube massively screwed with users by forcing poorly executed botched migration to YouTube Music. Even outsiders can see that this was entirely internal Google politics which powerful people like Wojcicki should have been able to avoid but she didn't. It just makes me wonder if these billionaire leaders of Google products really care anymore about anything. There is visibly an utter lack of hunger at the top and these people clearly should have been spending more time with family leaving these products with more hungry minds. YouTube recommendations are crap and it's still amazing that in 2024 just clicking one video will fill up most of recommendations with same thing. It never got around to incentivize creators to produce concise content and to this day creators keep producing massive 30 min diatribe that could have been done in 3 mins. TikTok took full advantage of this but YouTube CEO just kept napping at the wheel. Ultimately, the original product mostly just kept going but the measure of success is not about retaining audience but what it could have been if there was an ambitious visionary leader at the helm.
> It never got around to incentivize creators to produce concise content and to this day creators keep producing massive 30 min diatribe that could have been done in 3 mins.
Why on Earth would you want shorter videos? The best thing about YouTube is that it's one of the only places you can find quality medium-to-long-form content.
Maybe not what the commenter was saying, but there is a difference between great multi-hour essays and pointless rants stretching out their length to meet a minimum ad requirement. I like watching a lot of multi hour videos, but you can tell the difference between one with substance and one repeating the same thing over and over so they can "clock out."
That's all due to changes by YouTube to reward length and frequency, which of course makes sense for maximizing their ad revenue. But the result is creators are incentivized to pump out 20-minute fluff videos, not well edited/written videos.
People on here complain about SEO sites being filled with meaningless garbage. That's what YouTube is starting to be. The difference is their search bar still works whereas Google's will only give you the garbage. Though I still get "such and such breaks down their career" even though I've never clicked on that.
I agree that there are a lot of inflated videos to hit some ad target. But the solution is not to encourage people to create short videos, or at the very least, not the way TikTok did, making it almost impossible to popularize anything longer than 3 minutes.
And despite all the dredge, there is a lot of good content on YouTube, at least in certain niches. Video essays on media and politics, lots of video-game analysis and other fan communities, history content, lots of e-sports to name just a handful that I personally enjoy.
Search is literally one of the things YouTube is poorer at than ever and it blows my mind. I get a handful of results that might be relevant and then it’s just pages and pages of completely unrelated content that has nothing to do remotely with my search.
I find it a small price to pay if a few videos are too long (you can usually tell within three minutes anyway), to have a platform that generally encourages 30 minute videos and even 3 hour videos that do have content.
There's almost no meaningful 3 minute content possible, so a platform like TikTok that only works for short videos is basically condemned to be meaning-less, to be pure entertainment.
Clearly the add-supported side, that likes to pad and pad and show more adds, is working against the premium/fee-supported side, that wants to maximise value and engagement.
Premium subscribers should be able to give feedback on a video's density IMHO...
Length is shown in the thumbnail. Too long, no click, less views. I also wouldn't be surprised if the recommendation algo uses premium status as an input
Why on earth would you watch a 1.5 hour movie when you can watch a 2 min TikTok that explains the entire story?
In a world full of distractions I for one love the more slow-paced videos than “shorts” churned out by content mills designed to feed the modern day digital ADHD…
Few years ago “long burn” story telling was hot and we are still feeling the effects. Take any show on Netflix and it will be 8 45min episodes from which first 3 are absolutely garbage filler.
Youtube learned the wrong lesson and started to optimize the algorithm for retention and length. It is annoying to click for a review of some product that looks like a lengthy one with probably tests and what not only to see painfully slow unboxing and a wikipedia read of the history of the product and company and then sponsor read and then they turn on the device for a minute and give arbitrary score.
Exact same info could have been communicated in 30seconds, but then they wouldn’t get sponsor money and mid video ad roll
I beg to disagree. I don’t watch movies to “get information”. I watch movies (and long form YouTube videos) to be entertained. Why travel places? You can look up photos and videos online and get the same “information”.
YouTube videos were originally limited to 5 or 10 minutes I think. And probably 480p or so. You have to remember when it started, video on mobile didn't exist and there was absolutely no bandwidth for it. So people watching YouTube were watching it on their PC, probably with a 1024x768 CRT screen, and that's assuming they had something faster than dial-up internet.
Oh, I do remember, I was around in the early days. I think (but maybe that came later?) longer form videos did exist, but only paying accounts could post them.
Genuine question: How the heck crapeware like CloudStrike got into all critical systems from 911 to hospitals to airlines? My understanding was that all these critical systems are just super lazy to upgrade or install anything at all. I would love to know all the sales tactics CS used to get into millions of systems for money!
Reading other comments here, sorry I don't have the link, one crowd strike salesperson threatened to cancel them as a Client, yes you read that right, if the client wasn't easier to work with. So they're bullies or at least that one salesperson in crowd strike is a bully.
Another article talked about crowd strike being required for compliance, people here talking about checkbox compliance. So there's a systemic requirement from perhaps insurers for there to be some kind of comprehensive near real-time updated antivirus solution.
Furthermore, the haste makes waste philosophy seems to not be honored, in my opining mind, by the minds who drive The impacted sectors of our economy. Hospitals, Banks, airlines. This kind of vulnerability should not have been accepted. It's a single point of failure. Even on crowdstrike's website they have this kind of like radar ring hotspot Target kind of graphic, where they show at the very center one single client app .. theirs, as if that one single client is the thing that's going to save us?
This is amazing sales tactics! So, you buddy up with insurance, they create a checkbox and recommend you for a revenue cut! Now you suddenly have millions of customers out of nowhere and your product gets installed on billions computers before you even know it. I have seen this tactic get used for many mediocre products. For example, 3rd party dishwasher soap recommended by dishwasher company. Amazingly powerful. I don’t think most of CrowdStrike employees even knew they were in more than billion computers with paid service. The CEO was just busy doing brutal marketing of this pointless product.
I used to make this argument until I came to know of law (at least in USA) that if you don't use your house and someone else starts using your house then they can claim the legal ownership of your entire house in as little as 10 years! It was very shocking and complete antithesis of property rights in US that are so dearly held. When you think about this, in long run, it might sometimes make sense that future generation can use what previous generation has built.
many countries have variants of this, for many years/since olden times. In my country, it is called 'hævd'.
Instinctively, it makes incredibly much sense to me, even more so than property law. The way I view it, it is precisely the foundation that ownership sort of rests on.
(ie, i respect 'we have been using this for a long time' more than 'our ancestors whipped everyone in the village').
No, those laws are actually a recognition of property rights. Although it has become a popular social media trope to criticize things like adverse possession and "squatters" rights as some evil thing, the reason they exist is to protect people from being kicked out of their homes by others exploiting lack of or technicalities in paperwork.
Let me give you some examples of why these laws are actually good:
* Adverse possession: Imagine someone has been living and maintaining a property for 30 years. They have a single family home, and a driveway. They have a fence around all of their property, and they pay taxes on all of the area. The footprint of the property is exactly as they bought it 30 years ago. Their property is beside a piece of undeveloped, unoccupied, unmaintained land. Today, a land developer purchases the undeveloped land, and they come up with a 100 yr old document that shows that the part of the lot where their driveway is should be part of their lot. The developer then demands that the fence be torn down, and the driveway be given to them, despite the fact that the homeowner bought the property that way, has paid taxes on and maintained that property for 30 years, and nobody has brought this up as an issue in the past 100 years.
* "Squatters" (really, tenants) rights: Imagine someone rented a home in 2020. They've been paying rent every month since then, on a month-to-month basis, without a written lease. In 2024, the home was sold to an investor who wants to renovate the home and flip it. The investor wants to flip the home as quickly as possible, so they tell the tenant to leave immediately with no notice. They call the police and tell them that someone is squatting in their home. The police tell the owner that they must file for eviction, because they are not capable of determining whether the person living in the home is a "squatter" or a tenant. Only a court can do this.
Adverse possession is really just a recognition that mistakes made a long time ago shouldn't undo current realities. And "squatters rights" are really just a recognition that tenants don't have to defend their own property rights in front of a cop on their front porch at 2am on a Tuesday, they get to defend their rights to their leased property in court.
These studios used to make ton of money because only a few are in town. Now, every teenager is his/her own studio and content pool is so vast that their revues must shrink. They have nowhere to run. I think their end state is to publish on YouTube/Netflix and continue living at fraction of revenue that they are used to. That's what they should prepare for and plan for. Their existence and importance was supported by scarcity imposed by cable network and they need to understand that. But instead they live in fantasy of becoming next Netflix and burn in billions of dollars in debt.
This is compelling and I think many elements of the spirit of this are true. The only thing I would add is that it will be interesting to see if professional, medium/high budget, polished, "produced" content will remain a stable and significant niche that is distinct from user-generated content. And does the high-production-value content directly compete with low-production value user-created content? And what percentage of a user's consumption it will represent in the future? And does the pool of available time for an individual consumer grow to accommodate both?
To put it another way, on some level, all content competes for our time, so it's all in competition. On a different level, a Twitch livestream or TikTok feels like an entirely different category of media from a scripted, high-production-value TV series or movie, and I want both in the world.
While traditional publishers may be losing % of daily media consumption - especially in younger age brackets - it's unclear to me where this trend asymptotes. My intuition is that most people will spend some time on "reels" or livestreams (or whatever), some time on blockbuster movies, some on Broadway plays, and some time on scripted produced "TV style" content. Some will expand their denominator of total time to accommodate additional media sources, others will pick one over the other.
It seems there will be a degree of loss of market share as you allude to, but it's unclear how dramatic it will be and where it stabilizes.
One thing is absolutely 100% for sure though in my opinion: media preservation should be deeply prioritized, and this news seems like a blow to that.
Yesterday was MTV News deleting their archive, right? But on your points: brace yourselves for AI-generated content, with increasing technical quality and probably also increasing entertainment quality. That will be The Flood and might wipe out the small human creators by their lack of discoverability, and replace the studios output because hey AI is cheaper than employing real humans. So what do we do then??? Also no archives because shareholder value (I think I use this term already too much)...
I think their numbers are BS, just like Hollywood's numbers have always been BS. Remember when Lord of the Rings lost money for New Line somehow?
Paramount's supposedly got 71,000,000 subscribers. At ~$10/month, they're making something like $8,500,000,000 / yr in subscriber fees. You gotta be daft to lose money on that.
At like a $1,000,000 / episode (last numbers I heard, might be old) that's 8,500 episodes of television a year. Pretty sure they didn't make anywhere near that much content last year. Seems like mostly all they're doing is taking away content and then charging you for it again. Disney already plays this game pretty extensively.
-2.8% annual return since Jan 2006, 4.25% annual return since Jun 2009, and their 5 and 10 year annual returns are way worse (dqdyj total return calculator doesn’t even provide a percentage return).
I think you're right. People never had all the content we can access today. So the real competitors of those few big providers are now the millions of people creating new content every day on social media, YouTube, etc.
> But instead they live in fantasy of becoming next Netflix and burn in billions of dollars in debt.
Hell, even ~15+ years ago I knew not to fall for cargo-culting arguments of "come work for us; when we're as valuable as Facebook the stock we'll pay you with will make you a billionaire". Yeah, ... sorry, not moving to Silicon Valley and try to haggle for coffee with imaginary invaluable stocks.
So, I can't believe that ... for all their money's worth, these companies can only come up with "when we have as much users as Netflix, we'll be billionaires".
This is complete incompetence from their leadership who show no value to their own content. They could have easily auctioned off/sold very old content to someone else but that kind of thinking would be beyond their competency. It's no wonder they go in such huge losses despite of having loyal audience and monopoly over unparalleled content. To this day I cannot get over the fact that there are literally millions documentaries out there made with a lot of love and hard work but only available through DVD or mailing in a check to some dude. Similarly, lot of my favorite music albums are still on cassette tapes and never digitized online by their creators. Fortunately, audience did digitized them nicely and uploaded over to torrents and that's the only way to get them today. Same goes of out of print books and magazines. The producers of this content could have easily digitized it and uploaded over to some marketplace and made at least free coffee money for rest of their lives but surprisingly they just never get around doing it. IMO, it just expresses complete naivety and disregard to importance of their own content. They sure spent days and months of blood and sweat but can't get around to do a last mile of uploading files.
There is a huge startup opportunity here for folks who are willing to chase these content and do the last mile on their behalf.
They could have easily auctioned off/sold very old content to someone else..
That assumes they own the exclusive rights to the content. A lot of media has many rights holders (writers, music, etc), and you need to get them all to agree a sale or waive their right in order to sell. That could be expensive because it's involve lots and lots of lawyers. For a bunch of old comedy clips it might not make any commercial sense.
This sounds like regulatory failure creating a deadweight loss. IP rights were invented to further the creation and distribution of works in the arts and sciences; if they're making it prohibitively difficult we're doing it wrong.
> IP rights were invented to further the creation and distribution of works in the arts and sciences; if they're making it prohibitively difficult we're doing it wrong
They definitely further the creation, as we can't see the old stuff! I find music to be particularly bad here. People on Youtube can clip videos with fair use and talk over them, but any audio with music in needs to be muted, even if it's part of the fair use, because music is enforced so stringently.
If the copyright system were changed so that rightsholders were guaranteed to get paid, but didn't have veto over publication, how bad would that be? The reasons people usually justify copyright protection usually centre on rewarding the creators, and I think the right to stop people seeing your work is a harder sell.
There is of course a new question of how to set the price, but you could e.g. have an auction of some kind where the highest bid must be accepted.
(There are certainly notable cases like Mein Kampf where copyright has been conspicuously used to prevent further distribution.)
> If the copyright system were changed so that rightsholders were guaranteed to get paid, [...]
How much would they get paid? What if they wanted to hold out for more? Would every piece of copyrighted material be paid the same? Or would you normalise it per sentence or per letter? Or per frame in a movie?
Also to a first approximation we can treat all music as roughly interchangeable and we can measure the 'amount' of music eg by runtime or by song, so compulsory licensing can set some arbitrary fees.
You can perhaps do something similar for video, but it's hard to do that for all copyrighted material. Eg for a video game a single sprite has a very different value than some modules in the game's engine.
I dunno man, did you see what I wrote a couple of lines down? There are ways of selling things that must be sold, I mentioned auctions, but also you can get independent assessments like in the case of compulsory purchase/eminent domain.
I'm sure you could object to some particular solution I propose, but people much smarter than me have studied this kind of game theory extensively and there are a lot of options.
The normalization can be the same as licensing is done now, "per work", negotiating specific usages needn't change, except that the seller has to allow that each covered work be subject to at least one must-sell auction per e.g. year. They can even win the auction themselves if all the bids are too low.
Maybe I should have said so explicitly, but in that case the seller would have to pay someone else, e.g. a common fund for promotion of the arts, or general taxation, or to the other bidders in portion to their bids to compensate them for not having any way to access the work.
Are you arguing that the idea is impossible in principle? The details will depend on what sort of incentives you want to end up with, but I can't see yet that there isn't some reasonable solution.
You can probably come up with some messy compromise. But I'm not sure it's really much better than the existing system which already comes with fair-use constraints.
Or you could eg charge people a certain tax as a proportion of their self-declared value of the copyrighted material. (With the provision that they need to sell the rights to that material at the self-declared value to any comer or something like that.)
There is some system like this for music in Austria. I may record a new version of your song but it entitles you to a writers credit and some guaranteed amount of all revenue.
But is this content available somewhere else then? Should it not have been archived then or given to those who hold those rights so they could publish?
> They sure spent days and months of blood and sweat but can't get around to do a last mile of uploading files.
The people making these decisions didn't spend any time, blood, or sweat on producing this content. That's why it's so easy for them to discard it: they're only concerned with making (big) money, not figuring out how they can preserve content without incurring a loss. Which, IMO, should be the goal for older, historical content.
I'm pretty sure the money we're paying countries for wars would cover historical content preservation costs a gazillion times over.
> lot of my favorite music albums are still on cassette tapes and never digitized online by their creators.
Or worse. Rust In Peace was completely rerecorded by studio musicians and that's what you get if you look for it on Spotify or even buy a new CD today! To actually hear the album as recorded originally, you need to find a thirty-year old disc and just deal with the scratches.
GP is not quite correct. It wasn't completely re-recorded, but Dave Mustaine re-recorded many of the guitar solos on the album, and the whole album was remixed.
At any rate, it's not the same (classic, imo) album that was released in 1990.
Exactly, I'm not going to argue that artists don't have the right to re-release their material. It's the destruction of the original that to me is unconscionable.
Imagine this in a historical context, what if the publisher of Poe or Hemingway just decided to burn all their manuscripts and stop publishing because it was a better tax write off than their accountants though they would make on the lifetime sales of the work.
Because this is exactly what we are doing to future generations, lighting art on fire.
> This is complete incompetence from their leadership who show no value to their own content
Have you not seen the U.S.? It doesn't matter. The system is orchestrated to existing money-people making money simply for having it in the first place.
Even "bankruptcy" doesn't mean anything anymore.
Your interest or legitimate use cases do. not matter. At all. Ever. For entertainment or technology.
They can still sell it. We shouldn't assume because the webservers were shut down that everything is deleted and gone. When the article says it's "gone" they mean it is no longer available to the public.
What proofs do you have that he is doing that? I mean I'm not him and my first instinct if I saw your previous comment "What are you even talking about. When is the last time you went to comedycentral.com to watch something... This is giant self-righteous paragraph of junk." would be to downvote it.
I'd downvote it because of "self-righteous paragraph of junk." and because I strongly believe that historical content should be maintained.
I don't really care to recover my older account on my MBP with a dead logic board. Do you want me to recover a really old account for a measurement-contest
This is the same as Facebook/Instagram, TikTok, Twitter, etc. taking something down for "hate speech" or "community guidelines". OP reported my original comment the same way Paramount removed content. Bit of a double-standard. I advocate for torrent, to believe in a large corporation to up-hold content (like a comment on the internet, or a TV series) who is surprised it moved to a paid streaming service.
I don't think it's even possible for a user to down-vote a direct reply to their own comment.
I can't see the post you're complaining about; but based on the tone of the rest of your messages, maybe you were just off topic or rude-atop-the-soapbox and others flagged you?
So, NanoGPT took 1.8 days on 8xA100 for 124M model training on 30.7B tokens using flash attention. This would translate to 14.4hr for 10B tokens. With llm.c it is ~1.5 hr which is almost 10X speedup!
Does this look ballpark correct? Is there any summary of where majority of this improvement comes from?
This doesn't make sense. There are tons of other business which will have higher day to day return and property appreciation. I don't think anyone was craving for car wash subscription and I don't know of anyone having car wash in their checklist monthly item. Those people exist for sure but not in that kind of demand that a town needs even two of them.
I think this is just another instance where private equity is creating imaginary business model of subscriptions and selling it to its investors. The money managers will get huge management fees and will disappear before the whole thing collapses in coming years. Meanwhile, they will run a lot of good and necessary businesses out of town.
It's hard to think of many businesses that require lower capital investment and maintenance than a car wash. The ones I see in my city that look like land banking are a big plot with asphalt covering it (don't even need to mow lawns), a few awnings with some plumbing for hoses and soap and other gear, and nothing else. Even a laundromat would cost more to set up.
These car washes are not simple, nor cheap. They aren't talking about places where you wash your own car, they're talking about the increase in automated ones.
In reality, intuition is simply a pattern recognition mechanism that sometimes work. Entire science is basically testimony of how our intuition lead us astray and why we need to be disciplined about looking at data, evidence and crafting experiments. Our intuition has always said Earth is flat, Sun rotates around us, time is constant... Virtually every single thing in science is how our intuition (aka the primitive pattern recognition) was so magnificently wrong.