If the story isn’t true, then it isn’t just making you think about what the US government is doing. It’s making you fantasize about it based on an unknown person’s political agenda.
Ah the comment was aimed at the person replying to the reason why this was unflagged.
I do appreciate you replying! Thanks!
I realise it's an incredibly difficult question to answer because of the what-ifs, though.
I would also like to think you'd have the same stance... And for what it's worth I do believe you would.
I'd argue that your stance is currently harder to uphold since it makes the whole situation a lot morally murkier, and also because it goes against popular opinion (one can only be immune to it so much).
Presumably not with just anyone who reported that claiming to have a source, but only because it’s Hersh.
As far as I can see your position evaluates to ‘dang trusts Hersh‘.
Since it’s 2023 and we use machines to do our thinking now, I consulted ChatGPT about this. I draw your attention to the last paragraph.
—————
strawpeople: Can Seymour Hersh’s reporting always be trusted?
chatgpt:
Seymour Hersh is a Pulitzer Prize-winning journalist who has had a long and storied career. While he has been praised for his investigative reporting and his ability to uncover important stories, it is important to note that not all of his reporting can always be trusted. Like any journalist, Hersh is human and may make mistakes or have biases that affect the accuracy of his reporting.
Additionally, some of Hersh's reporting has been called into question, and he has faced criticism for making claims that are not supported by evidence or for relying on sources that are not credible. For example, in 2017, Hersh faced criticism for his reporting on the chemical attack in Khan Shaykhun, Syria, which some experts said was based on unreliable sources and was contradicted by a large body of evidence.
In conclusion, it is important to approach Hersh's reporting with a critical eye and to carefully evaluate the evidence and sources he relies on. While some of his reporting has been praised for its accuracy and impact, it is not always reliable and should be independently verified.
No where do I see dang claiming he trusts Hersh. You're just putting words into their mouth now. If anything, they've exlicitly said otherwise in a few threads, but said Hersh shouldn't be so easily dismissed due to his reputation and history with similar reporting. So, his reporting holds more weight than mine or yours.
It should still be approached critically, though.
People here seem largely seem dismissive of this story because they don't like it (or the author).
I've addressed in another thread why the sources are unnamed, but it's plausible it's to protect their safety, and lack of presentable evidence could also be the same reason. Information could be somehow fingerprintes to identify leaks. Hollywood did/does it; printers do it too.
---
> Since it’s 2023 and we use robots now, I consulted ChatGPT about this
On a lighter note, this made me laugh. Somehow makes it seem like we're in 3023, not 2023... but also like it's 2023. What a time to be alive.
> No where do I see dang claiming he trusts Hersh. You're just putting words into their mouth now.
I’m not putting words in anyone’s mouth.
Given that it’s clear he wouldn’t give this post special treatment if it wasn’t from Hersh, we can reasonably infer that dang trusts Hersh more than a random poster as you suggest he should.
I don’t think you represent dang, and at question here is dang’s reasons for giving the story special treatment, which unless you are a dang sock puppet you don’t have special insight into.
I apologise for responding on their behalf. You're right that they can speak for themselves, it was uncalled for on my side.
... And since we're indulging in unnecessary snide comments: They've outlined their reasoning already in a few places. Maybe if you read this thread instead of conversed with ChatGPT, we wouldn't be in this situation to begin with.
> we are seeing things like declines in lifespans and increases in suicide, but I don't think mental illness is increasingly nearly as much as the stats would lead you to believe at face value.
This just seems like a flat out contradiction. How would suicides ‘skyrocket’ if mental illness is not increasing.
Agreed that 2012 may be irrelevant and we may not know the cause.
It's not a contradiction, I'm saying that an underlying increase in mental illness plausibly explains some of the effect, but I'm unconvinced it explains the entire effect.
I'm not sure suicides strictly are the result of mental illness. I, for example, have an uncle that killed himself. He may have been clinically depressed, but mainly, he was an alcoholic. I think it's likely what depression he had would have resolved if he successfully dealt with his substance abuse problem.
Did he die because he was mentally ill, or because drinking can be a pretty bad problem?
I'm drawing a distinction between "mental illness" and "mental illness diagnosis" that I maybe didn't make entirely clear. I think the former is increasing, but the latter is increasing faster than the former.
It seems like only around 10% of suicides are not associated with mental illness. So I agree it seems implausible that this 10% would account for the entire increase, but it could be some of it.
Major owners of US companies basically choose the CEOs and have a say on everything they want to.
For instance, msft CEO is on the board of directors of starbucks. And everything is like that in blackrock/vanguard galaxy. It is public info on internet, and you can start on wikipedia.
Then it is more the other way around which is unlikely to have any truth to it: it would be hardly believable that actual owners of big tech companies do not have any say to big tech steering.
There is some kind of exception: amazon, still mostly own my Mr Bezos, then driven by himself. Once he sells his shares, vanguard and blackrock will become the actual owners of amazon (they are right behind Bezos)... but I don't know what kind of relationship between him and vanguard/blackrock (you would have needed heavy investment to build amazon from somewhere and "be chosen" to get the money to be able to do that).
Blackrock and Vanguard didn’t choose Tim Cook, Jobs did. Sundar Pichai was chosen by Larry and Sergei, who still have control of the voting shares in Google.
Even if they did have the power to fire one of these CEOs, that doesn’t give them any leverage at all over small details like browser strategy, because their only option would to replace the CEO with someone who understands much less about the company. I.e. they can really only do anything if the CEO is manifestly failing.
Honestly your argument doesn’t work, and doesn’t reflect shareholder influence.
When I say those people are their own, I meant they do implement what the owners of the company want.
It is actually your argument which does not work: no one would believe the actual owners have no saying about to where and how _theirs_ companies are steered.
msft is using blink already, I would not be surprised to see apple align on that. But they have to care about anti-trust stuff, financing geeko may not be enough.
Actually, it is the other way around, you would have to give evidence that major owners of big techs companies (blackrock/vanguard) don't do anything at steering the companies... which they own... and one of the efficient ways is to push forward a CEO they "own".
I am not the one on the unbelievable side here, mate.
Why do you thing msft CEO is at starbucks board, and you can look deeper via internet public info and this habit of CEOs members of the board of the other companies which are owned by the same companies is kind of smelly and not an sole case. This behavior is not limited to blackrock/vanguard and the USA, you know the cumulative wages are very "motivational", but it gives good hints at who belongs a CEO.
If I recall properly, vanguard is usually the biggest major owner something like between 10% and 20% and blackrock not far behind, but the percentage is stable among most companies I did check, and in most cases (exception with amazon in which Mr Bezos has still control). But they are THE major owners in control and that almost EVERYWHERE.
As I said, this is public info on internet, you can start on wikipedia but you have web sites (which work even without google blink/geeko or apple webkit, aka absurdely and grotesquely massive and complex web engines) which reference the same info (A quick search on google... lol). You can do easily your research (I wish EU companies were as "search-enabled" than the US companies).
Another stinking stuff, hard to find, then it would require to get more verifications, would be some giga-fishy "debts" sucking out a company profits: one day I got a site with starbucks gigantic "debts" sucking out tons of money out via chronic interest rates: some of those debts were most in us currency, but some not (japanese yen as far as I can recall). I will keep that in mind for any further digging.
> If I recall properly, vanguard is usually the biggest major owner something like between 10% and 20% and blackrock not far behind
Wrong.
Just as you are wrong about Blackrock or Vanguard having the ability to fire Google’s CEO, or having chosen Apple’s CEO. I note that you continue to evade this point despite it being central to your reasoning.
Why make these silly conspiratorial arguments if they can be debunked using easily googled facts? Why double down when you’ve been caught out? Are you just hoping people won’t check?
Vanguard and blackrock are the major share holders across the whole of big tech.
As I said, this is public info and people can check that whatever you can post on this forum.
And again, I am not on the unbelievable side here, you are: you have to proove this duo which are the major owners in all of big tech, do not have any say in the steering of the companies they _own_.
Edit: I did a quick re-check, 5 minutes of "googling" (lol), and 1 new site brought something new in 2023 the other sites did not put forward for alphabet (unlike for meta/facebook for instance, which I did not mention for this reason), founders are still in control (like metafacebook) because their shares are actually voting super-shares (first time I see that mention for alphabet).
So vanguard/blackrock, it's apple/intel/msft/amd/nvidia/etc (starbucks/pepsi...) and you can exclude metafacebook/amazon and very probably alphabet(google), waiting for the other sites to sync on this information though.
Funny, in the context of this news: it would mean alphabet(google) founders are going straight against one critical apple software component which is in control of their own major share holders. Wow, the smell of this is... so filthy...
> Vanguard and blackrock are the major share holders across the whole of big tech.
This is simply a false statement. Holding a single digit percentage of a company doesn’t make you “the major shareholder”.
> As I said, this is public info and people can check that whatever you can post on this forum.
They should check, and so should you.
> And again, I am not on the unbelievable side here, you are: you have to proove this duo which are the major owners in all of big tech,
They aren’t ‘major owners’. They own a minority of shares, whether considered together or separately.
> do not have any say in the steering of the companies they _own_
They don’t own the companies.
They own a small percentage of the shares of each of the companies, completely insufficient to win any votes.
I see you’ve walked back your claim about Google now that you’ve checked the facts. At some point hopefully you’ll see that wherever you got this whole argument from, you have been misinformed.
The only thing which _may_ change (need more sources to confirm, since only 1 new site did pop up very recently with that new info) is about alphabet(google) because the other sites were missing the alphabet(google) super voting shares, but I knew it was the case about metafacebook though.
As I said in my previous post, that would make the context of the news even more awkward and weird.
And you are wrong, it does not change anything for all the other companies without "super voting shares" from big tech, actually, it expanded (unless the super voting shares information is missing too for amd, intel, nvidia, etc) and it does not change the fact, again, that you are pushing something that nobody sane would believe: major owners of a company would have no say in the company steering and management?? I am not the stubborn guy here mate, actually, I start to feel sorry for you.
And again and again and again (and it is over), you are the one who is not in the believable zone , _YOU_ are to prove _YOUR_ claim that the major owners (which takes 5 minutes of googling to check on several different sites, and which is public info, even the debts as it is significant info for "investors") do NOT have any say in the companies they own.
> Scams have been around as long as technology, and even the original Apple iPhone demo was completely faked. The mock-up didn’t work at all; Steve Jobs followed a carefully constructed choreography to mislead his audience into thinking it did
This is totally false. The iPhone demo wasn’t a mock-up. It was alpha software that worked to do exactly the things that were demonstrated. Of course there was some choreography to avoid known bugs as there is in any demo, but in fact Jobs did deviate from the script, scaring the developers who were present, but it didn’t crash.
If the author is willing to lie outright about this story or unwilling to do even basic fact checking, then nothing in the story can be trusted.
I'd say more importantly, a mocked-up software demo is not fraud. If I show you some new ux element or layout or workflow, who cares what's behind it?
The fraud comes when it's something like theranos that lies about the existence of a technology their product depends on. If their blood test tech was proven and rock solid and they mocked up a UI for it, that would be fine, as there is no magic that needs to come true for it to work.
That's why tech has a big element of "fake it till you make it". Say, Airbnb or uber were obviously doable software-wise, and so if you have an idea like that there's little harm in presenting the software as being as far along as you can, so you can get the signups you need to scale. That's ok in my books (it's obviously still fraud to trick investors, but confidence in low-risk software that's not finished is not really an issue).
The problem occurs when people don't understand the distinction between software tech debt and fundamental tech or research risk, and treat them the same. This is what the author of the article appears to do
1. It's not an Electron "app", it's a native, compiled regular good old program.
2. From the VS Code website: "By downloading and using Visual Studio Code, you agree to the license terms and privacy statement. VS Code automatically sends telemetry data and crash dumps to help us improve the product."
I mean, you could swap in any editor for comparison. A quick skim of the homepage seems to say that it’s capable of everything an editor-that doesn’t-claim-to-be-an-IDE has to do. Some folks might value the lack of Microsoft involvement.
Since when does VSCode cost an arm and a leg to operate? Also saying it’s one of the best doesn’t explain why. I’m asking whether there is something distinctive about it.
There is no evidence that management problems are harder to solve than coding problems. Not only that, but when coders make mistakes the impact can be large and the responsibility clearly falls on the coder, whereas when managers make mistakes, the responsibility is diffused to their subordinates. Managers should either be paid a lot less, or be fireable by their reports when they make mistakes.
It has nothing to do with difficulty. It’s not a leveling up game of leetcode.
> whereas when managers make mistakes, the responsibility is diffused to their subordinates.
No it’s not. They will roast their subordinates but unless the org is completely dysfunctional they are responsible for the ultimate deliverables.
The higher up you move, the more responsible for outcomes that are less and less in your control. You even gain legal liability as you enter the “officer of the company” levels.
> ...unless the org is completely dysfunctional they are responsible for the ultimate deliverables.
This is true in theory if everyone has perfect knowledge. In practice I've never seen it play out. A large part of management is framing (i.e. playing off imperfect knowledge). It happens before and after success or failure is decided. Smart managers commit, but frame their responsibilities or outcomes in favorable terms. They find ways to de-risk. They publicly share risky deliverables with other teams or increase scope to push decision making up the ladder. They commit to only what their team can slam dunk. They make sure to only lead rockstar teams (this is huge). They cherry pick metrics to make failure look reasonable and success look incredible. They leave before major failures are realized or hand failures off before they sour. Remember: In large human systems, feedback cycles can easily take years.
It's a game and there's lots of clever ways to play. But the only way to lose is to publicly accept loss. When you do, it's very honorable and moving. But I've never seen the hit to perceived competency offset by perceived integrity. Because at the end of the day, companies make money. There's no company metric for honor. It's politics, whether we choose to see it or not doesn't really matter. For an extreme example, take a look at the presidency. Incumbent presidents lose in election year recessions. Period. Never mind most economic crises are a decade or so in the making.
People in aggregate are much simpler than we like to think. And so leadership is much more sleight of hand than our nobler ideals would have us believe.
> It has nothing to do with difficulty. It’s not a leveling up game of leetcode.
Then you should reply to the parent comment and not mine.
> No it’s not. They will roast their subordinates but unless the org is completely dysfunctional they are responsible for the ultimate deliverables.
Please. If you’ve never seen incompetent managers bumbling along making their staff into scapegoats, etc., then you don’t have a lot of experience.
Nothing changes at the higher levels either. Projects are cancelled and teams fired etc, not the executive in charge. The time execs get fired is when they do deliver and there is some kind of liability.
I guess I’ve been mainly working at successful companies? The biggest company I worked at was Google 7+ years ago and managers that couldn’t deliver and blamed employees would get exited or demoted quite quickly.
It’s been even more aggressive in the startups I’ve been at since.
> There is no evidence that management problems are harder to solve than coding problems.
As someone who has done both in small and medium companies, I find people problems to almost always be harder. Most coding problems outside cutting edge companies tend to at least have solutions. People and leadership often doesn't have a right answer.
And yes, a key coders mistake can be large. But a management mistake 'build feature X over Y' or 'implement policy B' can just as easily cost the company dearly and put small companies out of business.
> There is no evidence that management problems are harder to solve than coding problems.
I think that's the sort of thinking some forms of schooling drill into our heads: harder problems = greater reward/value.
That's not really how it is. We get rewarded for the value we supposedly generate by solving some problem or at least some portion of that value is rewarded to us. So if there are some low hanging fruits that generate a ton of value, then that's how you will be rewarded. I'm sure Ph.D.s solve far harder problems than me all day long but societally we don't tend to value many of those problems or their solutions.
> We get rewarded for the value we supposedly generate by solving some problem or at least some portion of that value is rewarded to us.
And that's the sort of thinking other forms of schooling drill into our heads: value generated ~= value compensated.
That's not really how it is. We get rewarded for the position we are at in the class hierarchy. If you're "in charge of" someone, you have to be making more money than them.
Is that how it should be? No. It's very much not. But it's largely how it is.
Totally agree with you. I think your point and the one I made earlier are at least two of the naive beliefs often held by a lot of engineers in tech -- myself included. The really pernicious part is that the value generated by engineers are sometimes quite big and the reward they get is a small fraction of it but big enough to make the engineer happy that they never stop to ask for more or find a better deal.
There is! Management would be replaced by cheaper management if it were possible to do so at the same level of quality.
Either corporations are insane immoral pathological paperclip/profit maximizers, in which case any manager would be replaced with a cheaper version or even a algorithm, or they're not and managers aren't worth what they're being paid.
If the CEO or upper management is so fungible, why aren't they being replaced with staff who will work just as hard and effectively for a tenth of the price?
You’re confusing leverage and political positioning with competence and capability.
Not that all managers are incompetent. Some are indeed inspired and or excellent. However these are rare. Just as there are many average engineers who aren’t very productive, so there are many managers who aren’t very capable and are just in their position because they were in the right place at the right time.
Obtaining leverage and positioning oneself politically are indeed competencies and capabilities that are required to excel in medium to large orgs. They're not mutually exclusive.