The paper offers an interesting signal that heavy gabapentin prescribing tracks with higher dementia/MCI diagnoses, especially in 35-64-year-olds. Yet substantial unmeasured confounding, possible reverse causality, and modest absolute risk increments mean we’re far from proving gabapentin is neurotoxic. It’s a reminder to prescribe purposefully, review regularly, and keep cognition on the monitoring checklist—but not a reason for abrupt discontinuation in patients who benefit.
I was going to post this exact same comment. I was obsessed with Etherpad when it came out and to this day, reminisce about it whenever I fire up a Google Doc. Can’t wait to try this.
Non-removable storage isn't the problem. You have identified one actual problem: privacy. But even this isn't a real problem if you turn on system-level drive encryption:
Soldered-in components make for higher quality, lower cost production. Anecdotally, every Windows machine I've had has failed. Every MacBook machine I have replaced after 4-5 years when I wanted to upgrade to the latest technology.
What does restore mean this context? I’ve downloaded files from their online portal without providing a key. Perhaps restoring in this context means having them mail you a hard drive.
It’s a false narrative that this was about politics and non-work things in the workplace. It was a workplace issue that blew up and employees were annoyed with how the founders handled it.
> The hash rate is currently about six exahashes per seconds. Considering the most efficient ASIC miner with a hash rate of about 13,000 GHS (using the SHA-256 algorithm) being sold for about $2,100, an attacker will require about 500,000 hardware units and this will amount to about $1,005,000,000. When we factor in the cost of electricity and cooling daily, this figure rises to $1,006,000,000.
> To successfully conduct a 51 percent attack on the Bitcoin network would cost an incredible $1.4 billion. This massive network supports over 5 million specialized ASIC mining computers, consuming a total of 29 Terawatt hours of electricity a year—as much as the entire country of Morocco. One of the underpinnings of the Bitcoin network is...
1. The social benefit. There's research showing that if you have a rich social life and friend network, you live longer. The assumption being that the drink is likely to be with other people.
2. The stress management benefit. It's not healthy to be stressed out. The alcohol's mitigating effect on stress may be greater than the alcohol's poisonous effect, in the right dose, for the right person.
Current valuations, Tesla and Apple among the most prominent, are so far disconnected from reality that it echoes the dot com bubble. But this time, the cause is not dot com mania -- but buybacks, driven by reckless financial policy.
1. VR/AR finally become a thing as the tech becomes small enough, powerful enough, and wireless enough to be a good UX. Facebook's investment in Oculus pays off as Oculus becomes their main profit center, and we transition to a Ready Player One reality.
2. The world faces increasingly alarming & extreme weather disasters. The damage will be utterly devastating and start to really hurt economies. Governments will still be inept at managing these crises. People who have the means will move to lower-risk areas like the Pacific Northwest, driving up real estate prices. Those who are poorer will suffer from homelessness. There will be massive encampments like we've never seen before of displaced families.
3. MDMA and psilocybin will be legalized for medical use and will revolutionize mental health care for depression and PTSD. Which is great because we'll all need a lot of mental health care.
4. Trump loses his re-election bid and ultimately flees the country to avoid consequences as the world closes in on him. He loses his US assets but is given safe passage by a foreign power he has had corrupt dealings with.
5. CRISPR is touted as the tech of the decade as it cures a major disease and appears to have incredible potential. At least one nation violates international ethical standards in their experimentation with CRISPR, igniting a debate over whether we have opened a pandora's box. There are protests and our CRISPR anxiety is evident in pop culture.
6. Americans finally catch on to the bidet/washlet craze thanks to an incredibly effective viral ad campaign.
7. Learning Chinese in school becomes more common because China has become the dominant world superpower. India rises to second, and US falls to third.
8. Bitcoin will cost $500k+ per coin, having peaked at 100-150k by 2022 and 500k+ by 2026. There will be a Silicon Valley-esque tv comedy about rich crypto bros that everyone hate-watches.
9. There will be a major pandemic that scares the shit out of everyone because we are woefully unprepared for it.
10. We're all going to become preppers by the end of the decade because there will be so many potential existential threats.
11. Water will become a lot more expensive, driving major innovation around desalinization.
12. Lab meat will have it's first big success a la the Impossible Burger as clean/lab meat tech becomes commercially available. There will be a thriving DIY culture around growing meat at home.
>I think the general theme I see is that folks expect companies to solve problems that their governments must be solving. And when they don't get it uniform, everyone's mad.
People want an outcome and don't particularly care where it comes from, public vs. private. If one fails, they'll push on the other to find a leverage point. Example: A pundit can say some truly awful and damaging shit and it's legal under the law. But if you target their advertisers, that's the leverage point that matters.
>I'd rather live in a world where companies aren't trying to push their morals on me and have a central entity (govt) arbitrate the same (Believe me I see evil in both places).
A company should be free to push its morals on you, given that the market allows it, and their morals aren't illegal. Right?
Personally, in the current environment of profit-at-all-costs capitalism, the major flaws seem to be incentivizing short-term-thinking and negative externalities. When a company flexes morals that appear at odds with short term profit (e.g. Twitter), I tend to assume they are actually acting out of self-interest but are better able to grasp the long vs. short-term incentives, for whatever reason.
You make a good point about how to actually enforce this. Apparently, Washington state did ban political ads on Facebook. As a testing ground for such a policy, the results haven't been great.
HOWEVER, this appears to me as a regulation failure. We know Facebook doesn't want the ban, so their motivation to comply is limited to the clarity and sharpness of the teeth of the legislation. And they aren't very sharp.
Regarding:
> Twitter hasn’t actually banned political ads, they’ve just used a definition that makes it easy for them to claim to have done so.
I don't agree with painting Twitter as just wanting to "claim" they have done so. They appear to be making a true good-faith effort. Check out their policy:
> Is an ad for a climate change organization a political ad? How about an ad for UBI? An ad for a local Catholic Church? An ad about farming subsidies? An ad for birth control?
For each of these examples, there is a clear way to apply their policy based on the content of the message. Is it perfect? Probably not. Will it totally kneecap political ads (by 80%+)? I believe it will.
> All of these are one step away from direct advertisements for candidates and are very political topics for many people.
I would argue that your point is too academic. If political ads are reduced by 80%, even though there are still political-adjacent ads (that aren't funded by a political group and don't reference a candidate or initiative), then the policy would be a wild success.
I was hoping a Facebook engineer would show up in this thread. If you don't mind, I have a couple of questions.
> Something the public doesn't get but engineers do is it's impractical to manually review every action on Facebook, human reviewers aren't necessarily more accurate than ML, and you'll always have some amount of abuse.
I'm curious what your thoughts are on other companies (Twitter, Spotify, etc) disabling political ads for this very reason. Facebook has not. It's a given that you can't manually review every political ad -- so why allow them at all, if their disinformation has negative real-world implications? I don't buy Zuck's argument about free speech.
Secondly, what would Facebook do if a grassroots movement started to put pressure on advertisers until Facebook cancels political ads? What if this movement recruited real users to click on lots of ads in their own feeds, with the goal of disrupting advertiser ROI with difficult to detect garbage clicks? Would advertisers get upset? Would Facebook have any recourse?
Thanks for indulging me on this. The idea came to me in the shower and I'm not sure if it's brilliant or stupid.
> This is, unfortunately, a rational response to the current climate.
Yes! It is rational to have a fear response to the prospect of being held accountable. Because we have not had a culture of accountability, many men are not confident in their ability NOT to behave in a harmful manner, even if they do not intend to harm. That feeling of discomfort is from finally having to think about these things. I hope more men embrace that discomfort instead of resent it.
They never will be. I think people misunderstand cryptocurrencies because of the mania that surrounds them. Bitcoin, for example, is a protocol, and should be judged as such. The most important thing for the protocol is trust, reliability, and scalability. Not UX and convenience. That could/should come from the layers/products that get built on top of the protocol.
Depends on the industry. For very large transactions, like if you are paying a home improvement contractor for something, they will often make you eat the fees if you want to pay with a credit card. So you end up having to figure out where your checkbook is (probably collecting dust in a box somewhere).
I give Bitcoin and LN a continuous pass on UX for now. We're in the infrastructure phase -- growing trust and improving reliability matters far more than UX. Before you know it, there will be 20 years of development and battle-testing behind the tech. The UX can come at any point in the future, while a breach of trust would be fatal. It's a question of ruthless prioritization. Thankfully, the open source devs behind these projects are long-term thinkers, not ponzi-scheme enablers.
Lastly, the UX need not come from Bitcoin/LN. They are a protocol, not a product.
And when the internet was first a thing, shortly after, most traffic was porn, until fairly recently porn was surpassed by social media. And since then, there have been world changing social movements enabled by the internet, and the exposure of horrible abuses (police, clergy, dictators).
Tomorrow, who knows. if North Korea or Russia or China turn half of the US into a nuclear wasteland, the banks are unlikely to function very well for those who remain. There are hundreds of such scenarios that could play out in the next 50 years -- Bitcoin is a great disaster hedge, if nothing else. Everyone should have 1% of their portfolio be crypto, IMO.
The criticisms of its fluctuation are valid, but like you said, there are plenty of people today for whom the fluctuations are tolerable for their use case. As the fluctuations get less, it will become tolerable for even more.
> You cannot look at bitcoin by any measure and say that it's a practical success for any of its goals.
I can memorize 20 words and cross whatever border I want with $100 million in my brain, and full confidence that that money is mine. That's pretty damn cool.
> I do support some form of decentralized currency. Not because people are ever going to be smart enough to be their own bank or because I think some currency will magically topple governments and financial structures, but mainly so people can transact without overly-censorious middle-men who take a cut while adding nearly no value.
It sounds like you're projecting what you think Bitcoin should be.
> We have the internet, now. Payment networks should have died years ago. I'm glad this problem is being worked on. However, it's possible to solve without using a country's-worth of energy to support a pathetic ~10 transactions per second.
10 transactions per second on the L1 network (averaging $10k-20k each), and many thousands possible today on L2. Plus a way to securely store the transacted value.
To argue that a small country's worth of energy being used to secure Bitcoin is a poor use of resources, you really have to consider how much energy the worldwide banking system consumes in all its complexity to do the same thing. I don't know how to do that calculation, but my point is there's more to it than you're seeing.
You have to also consider which energy is being used. Dirty energy is certainly part of it. But a lot of the energy used is through geographic renewables arbitrage. For example, there are hydro power facilities in China that were built and never fully utilized or connected to the main grid. You can set up miners nearby those facilities to use the surplus energy, and as long as you are connected to the internet, you're good to go.
It's also a potentially great incentive for governments to start NEW renewables facilities, because they have a way to use the surplus power until their populations grow to use the full capacity of those plants.
https://chatgpt.com/share/68709753-d2ec-8010-ba8b-e7b57d610e...
TLDR: