Every company I've said that about is still operating smoothly to this day.
12,498 karma · joined March 22, 2015
I think it's very hard to tell the difference between these things ex ante. Everything that succeeds looks like it was pulled. But many things were pushed before they were pulled. Buying things on the internet, for instance. This was laughed at for quite a while, while a few people and companies were committed to the idea that it was the future. They were right of course, so right that it feels like it was effortless and demanded by the public, but it wasn't. Not at first.
That isn't to say that that will happen with VR either. Just to say that there is a long history of things being pushed hard and finally breaking through and becoming a near universal norm, and there is a long history of things being pushed hard and dying and being completely forgotten.
The promise of VR is to unlock a world free of the constraints of physics. Where physical distance prevents certain styles of communication and interaction - where physical real estate is finite and therefore rivalrous, etc. There are many things that VR, in principle, could solve. Whether or not the tech can deliver those things in a way that partially (or totally?) replaces actual physical reality remains to be seen. It is clear though that if you want to weave tech into the fabric of people's lives in the deepest possible way, something like VR is probably your best bet.
This is a political stunt. Nothing more. Some helpful context:
https://twitter.com/superanonymousk/status/16666372273239982...
Gensler is a man who's overtures to work for Binance were spurned. He came into power at the SEC, and then he went after them, literally since the day he entered office. They were originally going to drop this case right when FTX blew up, but they knew how bad the optics of that would be, so they waited. This isn't someone impartial, looking out for investors. This is someone spiteful, looking out for himself.
If he wanted to protect people, he'd have done this when the market was hot. That's when investors need protection. That's when a good regulator acts - when a market is overexcited and people are being exploited. Who, exactly, is being protected now? From what?
The biggest, strongest example is pharmaceuticals. It costs hundreds of millions of dollars to develop a new drug and get it through the approvals process. It costs almost nothing to copy the drug once its developed. Without the protection of copyright, nobody in their right mind would invest the money it takes to develop new pharmaceuticals.
There are lots of important debates to be had about how pharmaceutical companies game that system in practice, and whether or not we should tweak the rules to make that harder. However, it seems undeniable to me that pharmaceutical innovation would completely grind to a halt if copyright were abolished.
That's a fair statement, it could even be true. But the fact remains, what is confidently asserted by many lacks evidence to substantiate it.
> Once you get into the rank and file engineers, where the folks have a lot less to lose by speaking their mind, the praise has been much harder to come by, and you have incidents like at Twitter, where engineers called out Musk's apparent inability to comprehend Twitter's architecture, resulting in at least one heated exchange during a Twitter spaces chat.
I think it is useful here to refer to exactly what it is that you mean. I am aware of a few such instances, but the instances I am aware of are Musk being mostly right and the engineers in question being mostly wrong. Perhaps you are aware of different incidents.
> Also, not for nothing, but if you haven't seen any negative comments about Elon from "celeb" engineers, you should probably look up the stuff Woz has said about Elon, it's not all positive, at all.
Woz hasn't actually worked with Elon. Anyway in a cursory search, I can't find the quotes you seem to be referring to. I'm not disputing their existence, but a link would be helpful to make this concrete.
> Also, it sounds a lot to me like you neglect the relationship between sociability and engineering work. It is a huge thing. It absolutely cannot be ignored to be a successful engineer. Linus, despite criticisms, is quite sociable and socially aware, and it's a big part of his success.
Agree completely. But it is an objective fact that Elon Musk has inspired lots of engineers to work for him, and to do so at an incredibly high level of productivity.
Where is the list of legendary engineers who've worked with him who say the opposite? Seriously, even one example would be interesting to me. I'm not aware of any.
The baseline is the median HN commenter. And Carmack's opinion carries infinitely more weight than that. I would argue it carries substantially more weight than the 99th percentile HN commenter.
The point is: All of the people who have worked directly with him and are also famous for their engineering talent in various domains respect him. I'm literally not aware of a single example in the opposite category. And I'll take the opinion of widely respected engineers who've worked with the guy over the commentariat on an internet forum any day.
I think you can still be a pretty good engineer and believe that a stabbing in SF is probably perpetrated by a random homeless person absent other information. It may have been wrong for him to jump to this conclusion, but I don't think it has any bearing on his technical skills.
> There was another example where he questioned the ethics of a place I've worked at. He didn't have a leg to stand on but he made public pronouncements.
Again you may be perfectly right, but this has nothing to do with engineering talent. Lots of great engineers are famously assholes (in various ways): see, Linus Torvalds, Richard Stallman, etc.
> Somebody who does that is not likely to be a good engineer. A good engineer waits for the facts to come in before reaching conclusions. He's clearly not level headed. I've seen loads of bad engineers who act like Musk in these circumstances. Sometimes they have good reputations in their orgs because overconfident narcissists sometimes get rewarded.
This is just not the case as any kind of rule. Like, sure, a perfectly rational Bayesian optimizer may not do these things. But real engineering at the highest level isn't about that - it's about intuition from experience about which system design choices will be good and which won't, and acting quickly and efficiently on that information.
The very small intersection of incredible engineers who are also very good people have the humility to quickly pivot when they've been proven wrong, admit it, and move on. Elon is obviously not in this category.
But I think you have to admit that it's telling that when pressed for examples of his technical failures in public, all you could come up with were completely non-technical examples.
Yes, this is more or less what I believe. Or to be more precise, I think over the past 5-7 years his fame, attention, and power have corrupted him.
> He's also been caught out since taking over Twitter of broadcasting very inaccurate things about the state of Twitters software stack.
I don't remember ever seeing a valid example of this. I do remember people dunking on things he said at various times, but those people were wrong, and he was right. Nearly all the Musk haters predicted Twitter would collapse after he fired half the staff, that it would fall apart. Former engineers wrote op eds about how it would happen. None of it did.
The particular instance to which I think you're referring was some front-end enginee attempting to falsify Musks's claims about how content was loaded, but the frontend engineer clearly didn't understand that Musk was talking about the micro-services that stitch together the feeds loaded by the front end code. Now, Musk may have been wrong - I don't know what Twitter's architecture looks like inside, but the guy who was criticizing him didn't either.
> Hyperfocus on the software is also definitely the wrong thing to be worrying about when it comes to running a content company and he's only made the product worse in his time there and has shown close to zero understanding what makes the site valuable.
I'm not sure I agree. I did think that at first, but I think he actually does understand what makes the site valuable - the users, and diversified revenue streams. He hasn't fully achieved that, but he has time. A social media site where you actually buy the product instead of being the product is what people have claimed to want here for quite a while.
Twitter is still humming along, and it's still growing: https://www.theverge.com/2022/11/7/23445476/elon-musk-twitte...
He could certainly be lying. He no longer has the kind of obligations not to that a public company does. But the idea that Twitter is dying or dead is just not the case. He certainly overpaid for it, and I don't think every move he's made has been good, but he's iterated rapidly, and been willing to revert unpopular changes quickly.
Err. Sure. It is, in principle, possible for all of these people to be wrong. But where are the well known engineers who've actually worked with him who say he's not? There should be quite a few by now, right?
> I don't have an opinion on Musk's skills as an engineer. I know I've seen him be very arrogantly and confidently wrong about things I know pretty well. That's a bit of a red flag trait that probably bleeds into other areas of his life and work. But also, i don't draw this conclusion out of hate or prejudice for the man. I don't care.
I've seen a lot of people make this claim. But every time I've seen this claim made in a field I know well, it was the person making the claim that was wrong, not Musk. Maybe in your case you're right. But my experience is that every time i've seen him dunked on for something technical, he was mostly right and they were mostly wrong.
These people are not sycophants. These people do not need Musk. They have plenty of their own money. They are not speaking highly of him because they want a promotion, or they want to be in his good graces, or they're afraid of being fired.
You can still hate him for the way he behaves. You can still think he's a bad guy. But if you think your opinion of his talent or engineering depth is better than these people's who've worked with him, you should seriously reconsider your epistemic blind spots.
It's ok for him to be a good engineer and a bad person. Von Braun designed rockets for the Nazis and later the US space program. Nobody doubts his engineering brilliance.
Even when Carmack was Zuckerberg's "subordinate" he was independently wealthy and could do whatever the fuck he wanted, and has been for a long time. You didn't see him going around, even then, talking up Zuckerberg's engineering chops. Despite the fact that we know Zuck did in fact write a lot of the early code for FB.
Carmack is also not a "different kind of engineer". Guy is primarily famous for his software, but he had a rocket company of his own, ran Meta's VR labs which involved quite a bit of hardware, etc.
The point of this CFTC lawsuit is to attack crypto, no more, no less. If you don't like crypto, you may think that's a good thing, but I'd argue that it's always a bad thing when regulators leverage technicalities to achieve political ends. And make no mistake, that is exactly what is happening here.
Let's not conflate events here. They did know about some patterns of criminal activity. Its not stated, as far as I know, that they actively aided/coached these people on how to evade anything.
They also knew about the use of their platform by e.g. HFT companies based in the US who accessed it via shell company. They appear to have, to a limited extent, coached this group on what to do. This behavior is the primary substance of the CFTC's suit, not the criminal/terrorist/ransomware behavior.
It's important that these things not be mixed up. The latter is utterly trivial, protects nobody, and it is the real substance of the lawsuit. The allegations of aiding and abetting crime are not the substance of the lawsuit, presumably because there is no significant evidence of it occurring.
Certainly I agree with you that, in principle, this is legitimate enforcement. They do have the authority to regulate this as you are saying they do. But you and I both know that there are no serious concerns within the CFTC about counterparty risk here, and that has nothing to do with their motivations.
They are very clearly not trying to protect anyone - at least not from derivatives counterparty risk. And they are most definitely not trying to protect Citadel, Jump, et al, nor anyone else from them.
> A civil complaint is not required to be, and generally is not, a catalogy of every piece of relevant information the filing party has. And it especially isn't a catalog of everything they know where a civil lawsuit isn't the venue for addressing it.
Cool, so we're back at zero evidence of any of this happening. We agree then.
I don't know, more than the percentage of your average multinational bank? No large financial institution is going to have literally zero unsavory characters using its services.
> Crypto space is chock-full of scammers, hustlers, bad actors, and organized crime. Which of the crimes are worth paying attention to and which aren't? Has the bar slipped that low that some crimes are "small" and aren't worth going after?
The point of my comment and Matt Levine's article is that they are not going after these things. They are superficial artifice on top of the real thing they are going after them for: allowing US HFT firms to trade there.
Here is the actual lead in to him quoting that part:
> Anyway I said that there are only a few accusations of financing crimes or secretly trading against customers in the CFTC complaint, but there are not none, and I should quote them. Here’s this:
Judge for yourself if I am misrepresenting his view.
I don't, but obviously the CFTC doesn't either, otherwise they'd have spelled it out. So, considering the party investigating is ignorant of any significant terrorist financing activity, I see no reason to think otherwise.
> Binance was a major counterparty of Bitzlato, a crypto exchange that existed primarily for Russian money laundering: https://www.reuters.com/business/finance/binance-moved-346-m...
There is nothing in the complaint stating that Binance knowingly facilitated any criminal activity of Bitzlato, afaik. The fact that criminal actors had an account at a financial institution is not prima facie evidence of any crime on the part of the financial institution, especially not for one that isn't registered in the US. Even for US registered entities, plenty of criminals have bank accounts. It's the bank's job to perform a certain level of diligence to stop them, not an infinite level of diligence.
If the CFTC believed Binancing knowing facilitated criminal activity on the part of Bitzlato, that'd be in the complaint. It's not, therefore, they don't. Or at the very least, they have insufficient evidence for it.
So any suggestion that significant terrorist financing is being stopped here is pure speculation, unsupported by anything in the document itself.
> The CFTC’s complaint here gestures at traditional regulatory concerns like retail customer protection and cracking down on money laundering. But it is mostly about cutting off a big international crypto exchange from big sophisticated proprietary market-making firms in the US. I think the market expectation here was that if you are a big trading firm trading with your own money and your own algorithms, and you have enough lawyers and offshore shell entities, you can trade on any crypto exchange in the world from the comfort of your Chicago office: There might be a technical argument that it’s not allowed, but your lawyers are aggressive and sophisticated enough to get around that technicality, and anyway why would the CFTC care? But the CFTC does care, perhaps not because it wants to protect big US high-frequency trading firms from the risks of trading on Binance, but because Binance is the biggest crypto exchange and this is a lever to crack down it. And the CFTC also has lawyers who are sophisticated and not deterred by technicalities — here, for instance, the technicality that Trading Firm B’s account actually belonged to a Jersey company.
This is the actual substance of the complaint. The "terrorist financing" and retail stuff is a smokescreen to seem worthy of the CFTC's attention. Nobody is being protected here. No significant terrorist financing is being stopped.