> have to hand-crank a tiny air compressor just to start a small emergency generator
Similarly, the US Navy maintains banks of pressurized air flasks to air-start emergency diesels. Total Capacity being some multiple of the required single-start capacity
Random fact: Those starters are a plot point in the 1965 film The Flight of the Phoenix, where the protagonists are trying to start a plane that’s stranded in the Sahara, but only have a small supply of starter cartridges left.
I lived for a while on a sailboat equipped with an ancient Saab tractor engine (8 whole horsepower!). Was designed for cartridge starts in cold weather, though someone had fitted an electric starter by the time I saw it
Zuck is learning theres a difference between shallow short term engagement and deeper long term engagement. Who could have seen this coming, except literally everyone?
It's like a tragedy of the commons, except there's only one party destroying all resources for themself
In Zuck's defense, it's not just him, it's the entire American school of business.
They never learn. GM, GE, RCA, you name it. They always want to make more money now now NOW. They don't understand they're taking on a metaphorical loan. They don't understand the interest they have to pay.
It's the ultimate greedy algorithm. Just make the decision that makes the most money right now, every time, over and over and over again. Don't look at anything else.
They know, it's just that most of the people will be gone before the negative effects become apparent. Most senior people are only going to be around for 7.2 years so if they optimize for short/medium term benefits and cash out, the long term consequences won't affect them.
What makes you think “they don’t understand the interest they have to pay”?
They are optimizing for short-medium term profits. The people there in the early days pull the ejection code when the “interest” is due. The company coasts until some private equity runs the numbers and realizes the parts are worth more than the whole.
This is capitalism. You are using “interest” (a finance term) seemingly in a moral / ethical critique. If so, use a moral / ethical term instead.
Corporate valuation isn't about short-term thinking. It's actually all very long-term. Plenty of companies are not paying out all their profits to shareholders, and their valuation is entirely based on expectation that it'll happen in the distant future and the discounted perpetuity value will equal the initial investment, probably after the current investors are dead.
There are still plenty of vulture investors who find a way to trick the market in the short-to-medium term. I'm not convinced Facebook is a case of that, even though I hate what they do.
They need not learn, they do as they’re primed, to go for profit, squeeze and profit, profit and profit some more. Then profit even from the dead husk on the way out. That’s the hyper capitalist lifecycle of a business product.
All I want is nice, non-toxic, non-addictive place to share photos and birthdays and life events with my family and close friends.
I understand that's not going to net hundreds of billions in revenue, but surely a site like that could keep the lights on and the engineers paid at scale.
All those photos and videos cost bandwidth, and that ain't free.
But the number of people willing to pay for their accounts on this stuff is vanishingly small.
So either you run this as a side project and accept that it's losing money, or you start running ads. And the moment you start running ads is the moment your most profitable choice becomes slowly turning your site more and more addictive, so that people spend more and more time on it and see more and more ads.
(Or you can keep the place small and constrained to people who have a high chance of being able to kick some money in for the bills, I'm only paying about half my Mastodon instance's fees because of making this choice.)
Or you can create a huge societal shift where we decide that having non-profit social sites is a good thing, and that they should be funded by the state, even if many of the views on them contradict the views of the giant bags of money pretending to be humans who are currently in control of the country. Ideally this societal shift would make it much harder for these giant bags of money to exist, as well.
Oh also getting people to stick around on a site that's not built to be addictive is surprisingly hard.
Wikipedia runs on donations. Most of FB is a massively bloated interface to maximize engagement, shove as much “content” as they can anywhere and everywhere, track everything you do, and add more “features” to find the next mechanism to get people more addicted.
For over a decade, I used Facebook lite messenger app which was built for countries with spotty, slow internet. It was less than a tenth of the size of the US messenger (of course it was unavailable in the app store and had to be installed via apk), was fast and easy to use (no stories, feeds, money sharing, animations), and was much better at doing the one thing it was supposed to be for, messaging people. It finally stopped working a couple of years ago and the regular app is a bloated mess where chats are an afterthought.
And why? Ads. You need more engagement so you can show people more ads. You need more content, so you have more things to attach ads to. You to autoplay videos to get people to watch more and see more ads. You have to run trackers so you can better target your ads. It’s the ads, not the functions, that make the modern internet too expensive to be funded by individuals.
2000s Facebook was able to run just fine on 2000s internet and storage. It would take a trivial amount of modern data and a fraction of modern storage to run now.
As positive social networking disappears, the market demand for one you can pay for with no ads increases. Pricing would be difficult but every year the average consumer learns more and more about how much "free" costs.
I agree a non-profit approach might be the only option to avoid the same long term problems we've seen time and again.
You're technically correct - you can't force people to give consent for targeted advertising (since it would no longer be consent). But you're absolutely allowed to show people ads if they don't want to pay for ad-free.
Generally, trying to directly convert a free service to a subscription service can be much harder than starting out as a subscription service. Just look at all the resentful conspiracies about Facebook planning to charge money that would go viral back in the day.
Users don't like a contract radically changing from under them, and shifting from free to paid is breaking a contract in an immediately understandable way.
No one was forced to buy the plan nor was the free Facebook going to go away. You just would have had the option to pay to not have targeted ads. And that was vetoed by the EU, the very thing many here claim they'd like to do.
That case was about forcing users to choose between personalized ads or a paid subscription. I can understand why the EU would reject that.
A case like that is outside of the scope of my argument. My proposal is a site that offers subscriptions with no free ad supported option at all, which the EU wouldn't have an issue with.
> forcing users to choose between personalized ads or a paid subscription. I can understand why the EU would reject that.
Why do you understand why that should be rejected? I don't personally understand it at all. How can it be possible for users to get free Facebook and not give up any personal data to it? There would be no money coming in to keep the site running...
If social media were paid, it would effectively be another barrier between people with different means connecting with each other.
From the perspective of the EU and their regulatory environment (vis a vis GDPR) and given Facebook's reach and size, it fits with how they approach big tech and privacy.
| another barrier between people
It's been said enough before that cheap is always better than free. If the costs can be kept low enough, the benefits of removing ads and data-mining from the equation can be worth it. And there's always the option of regional pricing where that makes sense.
I'm still using Facebook for this, which works for the very few of my friends who are on it. It's actually nice if you aggressively report and unfollow everything you don't want to see.
Does anybody here know of an alternative that works like 2010 Facebook?
At this point, why would you trust anything? I certainly don't. Any platform that exists could get bought up by another company that just uses all the content to train AI.
There are structures that are more immune to this such as non-profits or cooperatives, but otherwise that distrust is warranted given the way it's all gone.
Im just a layman here, but i don't think this is true. Language is an abstraction, an interpreative mechanism of reality. A reproduction of reality, like a picture, by definition holds more information than it's abstraction does.
I think his point is that LLMs are pre-trained transformers. And pre-trained transformers are general sequence predictors. Those sequences started out as text or language only but by no means is the architecture constrained to text or language alone. You can train a transformer that embeds and predicts sound and images as well as text.
A picture is also an abstraction. If you take a picture of a tree, you have more details than the word "tree". What i think the parent is saying, is that all the information in a picture of a tree can be encoded in language, for example a description of a tree, using words. Both are abstractions but if you describe the tree well enough with text(and comprehend the description) it might have the same "value" as a picture(not for a human, but for a machine). Also, the size of the text describing the tree might be smaller than the picture.
I read this and I thought it was a bad writing, and I assumed (right or wrong) that (metaphor ensues):
The airplane's pilot could see where the ground was and he estimated it was 1000m-1200m below. But the instrument of the airplane were malfunctioning and reported that the ground was at 500m (or 5000m). This caused the plane to override/resist the pilot's manual landing efforts causing the plane to ultimately crash but not allowing the wheels to come down and the flaps to do turn.
So two different mechanisms, providing contradicting evidence, and the lack of a 'higher' deciding authority to use 'judgment' and/or other inputs to 'decide' which reported measurements are more accurate, and thus make the viable decision/take the viable action.
I get that for such devices every-gram-counts, and perhaps from now on some devices/mechanisms that are truly single-points-of-failure, will get 2x or 3x for redundancy/contingencies. If your ABC-camera fails, it's 'ok' you can still get images from your DEF and GHI will give you some data, reducing to 80% the output of the mission, but still not making it 5%. Your 'landing equipment' fails, and you get out only 5% of the value (flying/trajectory/telemetry of flight/perhaps some eclipse photos).
HR likely deals with health info related to disability or fmla claims, or work-related injuries that is shared with health care providers and/or insurance companies; this makes them a covered entity subject to the requirements under hipaa.
My understanding about hot water is that it causes localized histamine release, which then depletes the available histamine to trigger the itch reflex.
Certainly works great for poison oak and eczema, regardless of the actual mechanism.
Many sea-animal toxins are protein based, and heat will denature the protein making it harmless. Similarly, acid (like vinegar) can help, e.g. for jellyfish stings.
(And if the toxin has penetrated your skin by e.g. the sharp tail of a stingray, heat will be better than vinegar since you can't get the vinegar "inside" your body... whereas jellyfish toxin tends to sit on the surface of your skin)
I love the dialectical equivalency here, Feels exactly accurate. At least 4chan plays you laugh you lose, federal politics seems to omit the first half.
> dominance, including groups of chimps
Indeed one of the most fascinating lessons. "Alpha" by force is almost tautologically impossible; leadership is built on inspiration and willfullness.
+1, the question of origin in general is politicized. I don’t think individual theories are taboo, it’s just very difficult to truly evaluate the evidence at this point, unless perhaps you are an expert and can do all the fact checking yourself.
Heliocentrism was politicized. The Galileo affair had a lot more to do with Church authority during the Reformation and Counter-Reformation than about cosmology. Galileo went out of his way to antagonize certain key players, and his punishment was primarily a political matter, not a scientific or even theological one.
The thing is that unlike political science, virology is an actual science. Scientists have found little evidence of coronavirus at the labs and there is record of them failing even to culture the virus which would be required for any research including gain of function research. In dramatic contrast samples from the wet market were easily cultured into multiple related strains which strongly points to the wet market as the source.
Similarly, the US Navy maintains banks of pressurized air flasks to air-start emergency diesels. Total Capacity being some multiple of the required single-start capacity
reply