The link 404s. I think the correct link is: http://rationallyspeakingpodcast.org/231-misconceptions-about-china-and-artificial-intelligence-helen-toner/
quanticle
Karma: 2,722
Just going by the standard that you set forth:
The overall impression that I got from the program was that as it proved profitable and expanded,
The program expanded in response to Amazon wanting to collect data about more retailers, not because Amazon was viewing this program as a profit center.
it took on a larger workforce and it became harder for leaders to detect when employees were following their individual incentives to cut corners and gradually accumulate risks of capsizing the whole thing
But that doesn’t seem to have occurred. Until the Wall Street Journal leak, few if any people outside Amazon were aware of this program. It’s not as if any of the retailers that WSJ spoke to said, “Oh yeah, we quickly grew suspicious of Big River Inc, and shut down their account after we smelled something fishy.” On the contrary many of them were surprised that Amazon was accessing their seller marketplace through a shell corporation.
I didn’t see any examples mentioned in the WSJ article of Amazon employees cutting corners or making simple mistakes that might have compromised operations. Instead, they seemed to be pretty careful and conscientious, making sure to not communicate with outside partners with their Amazon.com addresses, being careful to maintain their cover identities at trade conferences, only communicating with fellow Amazon executives with paper documents (and numbered paper documents, at that), etc.
I would argue that the practices used by Amazon to conceal the link between itself and Big River Inc. were at least as good as the operational security practices of the GRU agents who poisoned Sergei Skripal.
The overall impression that I got from the program was that as it proved profitable and expanded, it took on a larger workforce and it became harder for leaders to detect when employees were following their individual incentives to cut corners and gradually accumulate risks of capsizing the whole thing.
That’s not the impression I got. From the article, it says that many of the retailers that the Wall Street Journal had contacted regarding Big River had no idea that the entity was affiliated with Amazon (even despite the rather-obvious-in-hindsight naming, LinkedIn references, company registration data pointing to Seattle, etc). It seems like their operational security was unusually good, good enough that no one at the other retailers bothered looking beyond the surface. Yes, eventually someone talked to the press, but even then, Amazon had a plan in place to handle the program coming to light in a public forum.
In general, it seems like Amazon did this pretty competently from start to finish, and the leaders were pretty well in control of the operation all throughout.
(with an all-in-one solution they just buy one thing and are done)
That’s a very common misconception regarding all-in-one tools. In practice, all-in-one tools always need a significant degree of setup, configuration and customization before they are useful for the customer. Salesforce, for example, requires so much customization, you can make a career out of just doing Salesforce customization. Sharepoint is similar.
Thus the trade-off isn’t between a narrowly focused tool that does one job extremely well versus an all-in-one tool that does a bunch of things somewhat well. The tradeoff is between a narrowly focused tool that does one job extremely well immediately, with little or no setup versus an all-in-one-tool that does many things somewhat well after extensive setup and customization, which itself might require hiring a specialized professional.
To use the Anrok example, is it possible to do VAT calculations in the existing tools that the business already has, such as ERP systems or CRM software? Yes, of course. But that software would need to be customized to handle the specific tax situation for the business, which is something that Anrok might handle out-of-the-box with little setup required.
I also have a “Done” column, which is arguably pointless as I just delete everything off the “Done” column every couple weeks,
Having a “Done” column (or an archive board) can be very useful if you want to see when was the last time you completed a recurring task. It helps prevent tasks with long recurrences (quarterly, biennially, etc) from falling through the cracks. For example: dentist appointments. They’re supposed to happen once a year. And, ideally, you’d create a task to schedule the next one immediately when you get back from the previous one. But let’s say that doesn’t happen. You got distracted, there was some kind of scheduling issue, life got in the way. Then, months later, you wonder, “Wait, how long has it been since I’ve been to the dentist?” Archiving completed tasks instead of deleting them lets you answer that question immediately.
The last answer is especially gross:
He chose to be a super-popular blogger and to have this influence as a psychiatrist. His name—when I sat down to figure out his name, it took me less than five minutes. It’s just obvious what his name is.
Can we apply the same logic to doors? “It took me less than five minutes to pick the lock so...”
Or people’s dress choices? “She chose to wear a tight top and a miniskirt so...”
Metz persistently fails to state why it was necessary to publish Scott Alexander’s real name in order to critique his ideas.
The second wheelbarrow example has a protagonist who knows the true value of the wheelbarrow, but still loses out:
At the town fair, a wheelbarrow is up for auction. You think the fair price of the wheelbarrow is around $120 (with some uncertainty), so you submit a bid for $108. You find out that you didn’t win—the winning bidder ends up being some schmuck who bid $180. You don’t exchange any money or wheelbarrows. When you get home, you check online out of curiosity, and indeed the item retails for $120. Your estimate was great, your bid was reasonable, and you exchanged nothing as a result, reaping a profit of zero dollars and zero cents.
But, in my example, Burry wasn’t outbid by “some schmuck” who thought that Avant! was worth vastly more than it ended up being worth. Burry was able to guess not just the true value of Avant!, but also the value that other market participants placed on Avant!, enabling him to buy up shares at a discount compared to what the company ended up selling for.
The implied question in my post was, “How do you know if you’re Michael Burry, or the trader selling Avant! shares for $2?”
That point is contradicted by the wheelbarrow examples in the OP, which seem to imply that either you’ll be the greater fool or you’ll be outbid by the greater fool. Why wasn’t Burry outbid by a fool who thought that Avant! was worth $40 a share?
This is why I disagree with the OP; like you, I believe that it’s possible to gain from informed trading, even in a market filled with adverse selection.
I don’t think the Widgets Inc. example is a good one. Michael Lewis has a good counterpoint in The Big Short, which I will quote at length:
The alarmingly named Avant! Corporation was a good example. He [Michael Burry] had found it searching for the word “accepted” in news stories. He knew, standing at the edge of the playing field, he needed to find unorthodox ways to tilt it to his advantage, and that usually meant finding situations the world might not be fully aware of. “I wasn’t looking for a news report of a scam or fraud per se,” he said. “That would have been too backward-looking, and I was looking to get in front of something. I was looking for something happening in the courts that might lead to an investment thesis.” A court had accepted a plea from a software company called the Avant! Corporation. Avant! had been accused of stealing from a competitor the software code that was the whole foundation of Avant!‘s business. The company had $100 million cash in the bank, was still generating $100 million a year of free cash flow—and had a market value of only $250 million! Michael Burry started digging; by the time he was done, he knew more about the Avant! Corporation than any man on earth. He was able to see that even if the executives went to jail (as they did) and the fines were paid (as they were), Avant! would be worth a lot more than the market then assumed. Most of its engineers were Chinese nationals on work visas, thus trapped—there was no risk that anyone would quit before the lights were out. To make money on Avant!’s stock, however, he’d probably have to stomach short-term losses, as investors puked up shares in horrified response to negative publicity.
Burry bought his first shares of Avant! in June 2001 at $12 a share. Avant!‘s management then appeared on the cover of Business Week, under the headline, “Does Crime Pay?” The stock plunged; Burry bought more. Avant!‘s management went to jail. The stock fell some more. Mike Burry kept on buying it—all the way down to $2 a share. He became Avant!‘s single largest shareholder; he pressed management for changes. “With [the former CEO’s] criminal aura no longer a part of operating management,” he wrote to the new bosses, “Avant! has a chance to demonstrate its concern for shareholders.” In August, in another e-mail, he wrote, “Avant! still makes me feel I’m sleeping with the village slut. No matter how well my needs are met, I doubt I’ll ever brag about it. The ‘creep’ factor is off the charts. I half think that if I pushed Avant! too hard I’d end up being terrorized by the Chinese mafia.” Four months later, Avant! got taken over for $22 a share.
Why should Michael Burry have assumed that he had more insight about Avant! Corporation than the people trading with him? When all of those other traders exited Avant!, driving its share price to $2, Burry stayed in. Would you have? Or would you have thought, “I wonder what that trader selling Avant! for $2 knows that I don’t?”
Why isn’t there a standardized test given by a third party for job relevant skills?
That’s what Triplebyte was trying to do for programming jobs. It didn’t seem to work out very well for them. Last I heard, they’d been acquired by Karat after running out of funding.
My intuition here is “actually fairly good.” Firms typically spend a decent amount on hiring processes—they run screening tests, conduct interviews, look at CVs, and ask for references. It’s fair to say that companies have a reasonable amount of data collected when they make hiring decisions, and generally, the people involved are incentivized to hire well.
Every part of this is false. Companies don’t collect a fair amount of data during the hiring process, and the data they do collect is often irrelevant or biased. How much do you really learn about a candidate by having them demonstrate whether they’ve managed to memorize the tricks to solving programming puzzles on a whiteboard?
The people involved are not incentivized to hire well, either. They’re often engineers or managers dragged away from the tasks that they are incentivized to perform in order to check a box that the participated in the minimum number of interviews necessary to not get in trouble with their managers. If they take hiring seriously, it’s out of an altruistic motivation, not because it benefits their own career.
Furthermore, no company actually goes back and determines whether its hires worked out. If a new hire doesn’t work out, and is let go after a year’s time, does anyone actually go back through their hiring packet and determine if there were any red flags that were missed? No, of course not. And yet, I would argue that that is the minimum necessary to ensure improvement in hiring practices.
The point of a prediction market in hiring is to enforce that last practice. The existence of fixed term contracts with definite criteria and payouts for those criteria forces people to go back and look at their interview feedback and ask themselves, “Was I actually correct in my decision that this person would or would not be a good fit at this company?”
Do you know a healthy kid who will do nothing?
Yes. Many. In fact, I’d go so far as to say that most people in this community, who claim that they’re self-motivated learners who were stunted by school would have been worse off without the structure of a formal education. One only needs to go through the archives and look at all the posts about akrasia to find evidence of this.
What does “lowercase ‘p’ political advocacy” mean, in this context? I’m familiar with similar formulations for “democratic” (“lowercase ‘d’ democratic”) to distinguish matters relating to the system of government from the eponymous American political party. I’m also familiar with “lowercase ‘c’ conservative” to distinguish a reluctance to embrace change over any particular program of traditionalist values. But what does “lowercase ‘p’ politics” mean? How is it different from “uppercase ‘P’ Politics”?
A great example of a product actually changing for the worse is Microsoft Office. Up until 2003, Microsoft Office had the standard “File, Edit, …” menu system that was characteristic of desktop applications in the ’90s and early 2000s. For 2007, though, Microsoft radically changed the menu system. They introduced the ribbon. I was in school at the time, and there was a representative from Microsoft who came and gave a presentation on this bold, new UI. He pointed out how, in focus group studies, new users found it easier to discover functionality with the Ribbon than they did with the old menu system. He pointed out how the Ribbon made commonly used functions more visible, and how, over time, it would adapt to the user’s preferences, hiding functionality that was little used and surfacing functionality that the user had interacted with more often.
Thus, when Microsoft shipped Office 2007 with the Ribbon, it was a great success, and Office gained a reputation for having the gold standard in intuitive UI, right?
Wrong. What Microsoft forgot is that the average user of Office wasn’t some neophyte sitting in a carefully controlled room with a one-way mirror. The average user of Office was upgrading from Office 2003. The average user of Office had web links, books, and hand-written notes detailing how to accomplish the tasks they needed to do. By radically changing the UI like that, Microsoft made all of that tacit knowledge obsolete. Furthermore, by making the Ribbon “adaptive”, they actively prevented new tacit knowledge from being formed.
I was working helpdesk for my university around that time, and I remember just how difficult it was to instruct people with how to do tasks in Office 2007. Instead of writing down (or showing with screenshots) the specific menus they had to click through to access functionality like line or paragraph spacing, and disseminating that, I had to sit with each user, ascertain the current state of their unique special snowflake Ribbon, and then show them how to find the tools to allow them to do whatever it is they wanted to do. And then I had to do it all over again a few weeks later, when the Ribbon adapted to their new behavior and changed again.
This task was further complicated by the fact that Microsoft moved away from having standardized UI controls to making custom UI controls for each separate task.
For example, here is the Office 2003 menu bar:
(Source: https://upload.wikimedia.org/wikipedia/en/5/54/Office2003_screenshot.PNG)
Note how it’s two rows. The top row is text menus. The bottom row is a set of legible buttons and drop-downs which allow the user to access commonly used tasks. The important thing to note is that everything in the bottom row of buttons also exists as menu entries in the top row. If the user is ever unsure of which button to press, they can always fall back to the menus. Furthermore, documents can refer to the fixed menu structure allowing for simple text instructions telling the user how to access obscure controls.
By comparison, this is the Ribbon:
(Source: https://kb.iu.edu/d/auqi)
Note how the Ribbon is multiple rows of differently shaped buttons and dropdowns, without clear labels. The top row is now a set of tabs, and switching tabs now just brings up different panels of equally arcane buttons. Microsoft replaced text with hieroglyphs. Hieroglyphs that don’t even have the decency to stand still over time so you can learn their meaning. It’s impossible to create text instructions to show users how to use this UI; instructions have to include screenshots. Worse, the screenshots may not match what the user sees, because of how items may move around or be hidden.
I suspect that many instances of UIs getting worse are due to the same sort of focus-group induced blindness that caused Microsoft to ship the ribbon. Companies get hung up on how new inexperienced users interact with their software in a tightly controlled lab setting, completely isolated from outside resources, and blind themselves to the vast amount of tacit knowledge they are destroying by revamping their UI to make it more “intuitive”. I think the Ribbon is an especially good example of this, because it avoids the confounding effect of mobile devices. Both Office 2003 and 2007 were strictly desktop products, so one can ignore the further corrosive effect of having to revamp the UI to be legible on a smartphone or tablet.
Websites and applications can definitely become worse after updates, but the company shipping the update will think that things are getting better, because the cost of rebuilding tacit knowledge is borne by the user, not the corporation.
3-year later follow-up: I bought a Hi-Tec C Coleto pen for my brother, who is in a profession where he has to write a lot, and color code forms, etc. He likes it a lot. Thanks for the recommendation.
On the other hand, if plaintiff has already elicited testimony from the engineer to the effect that the conversation happened, could defendant try to imply that it didn’t happen by asking the manager whether he recalled the meeting? I mean, yes, but it’s probably a really bad strategy. Try to think about how you would exploit that as plaintiff: either so many people are mentioning potentially life-threatening risks of your product that you can’t recall them all, in which case the company is negligent, or your memory is so bad it was negligent for you to have your regularly-delete-records policy. It’s like saying I didn’t commit sexual harassment because we would never hire a woman in the first place. Sure, it casts doubt on the opposition’s evidence, but at what cost?
If it’s a criminal trial, where facts have to be proven beyond a reasonable doubt, it’s a common strategy. If the whistleblower doesn’t have evidence of the meeting taking place, and no memos, reports or e-mails documenting that they passed their concerns up the chain, it’s perfectly reasonable for a representative of the corporation to reply, “I don’t recall hearing about this concern.” And that’s that. It’s the engineer’s word against not just one witness, but a whole slew of witnesses, each of whom is going to say, “No, I don’t recall hearing about this concern.”
Indeed, this outcome is so predictable that lawyers won’t even take on these sorts of cases unless the whistleblower can produce written evidence that management was informed of a risk, and made a conscious decision to ignore it and proceed.
Also keep in mind that if we’re going to assume the company will lie on the stand about complex technical points
I’m not assuming anything of the sort. I’m merely saying that, if the whistleblower doesn’t have written evidence that they warned their superior about a given risk, their superiors will be coached by the company’s lawyers to say, “I don’t recall,” or, “I did not receive any written documents informing me of this risk.” Now, at this point, the lawyers for the prosecution can bring up the document retention policy and state that the reason they don’t have any evidence is because of the company’s own document retention policies. But that doesn’t actually prove anything. Absence of evidence is not, in and of itself, evidence of wrongdoing.
One reason not to mess with this is that we have other options. I could keep a journal. If I keep notes like “2023-11-09: warned boss that widgets could explode at 80C. boss said they didn’t have time for redesign and it probably wouldn’t happen. ugh! 2023-11-10: taco day in cafeteria, hell yeah!” then I can introduce these to support my statement.
Yes, that’s certainly something you can do. But it’s a much weaker sort of evidence than a printout of an e-mail that you sent, with your name on the
fromline and your boss’s name on thetoline. At the very least, you’re going to be asked, “If this was such a concern for you, why didn’t you bring it up with your boss?” And if you say you did, you’ll be asked, “Well, do you have any evidence of this meeting?” And if your excuse is, “Well, the corporation’s data retention policies erased that evidence,” it weakens your case.
The thing I said that the defendant would not dispute is the fact that the engineer said something to them, not whether they should have believed him.
I still disagree. If it wasn’t written down, it didn’t happen, as far as the organization is concerned. The engineer’s manager can (and probably will) claim that they didn’t recall the conversation, or dispute the wording, or argue that while the engineer may have said something, it wasn’t at all apparent that the problem was a serious concern.
There’s a reason that whistleblowers focus so hard on generating and maintaining a paper trail of their actions and conversations, to the point that they will often knowingly and willfully subvert retention policies by keeping their own copies of crucial communications. They know that, without documentation (e-mails, screenshots, etc), it’ll just be a he-said-she-said argument between themselves and an organization that is far more powerful than them. The documentation establishes hard facts, and makes it much more difficult for people higher up in the chain of command to say they didn’t know or weren’t informed.
If you notice something risky, say something. If the thing you predicted happens, point out the fact that you communicated it.
I think this needs to be emphasized more. If a catastrophe happens, corporations often try to pin blame on individual low-level employees while deflecting blame from the broader organization. Having a documented paper trail indicating that you communicated your concerns up the chain of command prevents that same chain from labeling you as a “rogue employee” or “bad apple” who was acting outside the system to further your personal reputation or financial goals.
Plaintiff wants to prove that an engineer told the CEO that the widgets were dangerous. So he introduces testimony from the engineer that the engineer told the CEO that the widgets were dangerous. Defendant does not dispute this.
Why wouldn’t the defendant dispute this? In every legal proceeding I’ve seen, the defendant has always produced witnesses and evidence supporting their analysis. In this case, I would expect the defendant to produce analyses showing that the widgets were expected to be safe, and if they caused harm, it was due to unforeseen circumstances that were entirely beyond the company’s control. I rarely speak in absolutes, but in this case, I’m willing to state that there’s always going to be some analysis disagreeing with the engineer’s claims regarding safety.
If I say I want you to turn over your email records to me in discovery to establish that an engineer had told you that your widgets were dangerous, but you instead destroy those records, the court will instruct the jury to assume that those records did contain that evidence.
Only if you do so after you were instructed to preserve records by the court. If you destroyed records, per your normal documented retention policies prior to any court case being filed, there’s no grounds for adverse inference.
Plaintiff responds by showing that defendant had a policy designed to prevent such records from being created, so defendant knows that records would not exist whether the meeting took place or not, and thus his argument is disingenuous. Would you follow defendant’s strategy here? I wouldn’t.
Every company I’ve worked for has had retention policies that call for the automatic deletion of e-mails after a period of time (5-7 years). Furthermore, as I alluded to in my other post, Google had an explicit policy of disabling permanent chat records for certain sensitive conversations:
At trial, the DOJ also presented evidence and testimony about Google’s policy called “Communicate with Care.” Under that policy, Google employees are trained “to have sensitive conversations over chat with history off,” the DOJ said, ensuring that the conversation would be auto-deleted in 24 hours.
This policy has created much tension between the DOJ and Google before the trial. The DOJ has argued that “Google’s daily destruction of written records prejudiced the United States by depriving it of a rich source of candid discussions between Google’s executives, including likely trial witnesses.” Google has defended the policy, claiming that the DOJ has “not been denied access to material information needed to prosecute these cases and they have offered no evidence that Google intentionally destroyed such evidence.”
And while this does look bad for Google, one can very easily argue that the alternative, the release of a “smoking gun” memo like the “embrace, extend, innovate” document would be far worse.
The Firefox problem on Claude was fixed after I sent them an e-mail about it.