The only longterm solution that makes sense is to allow students to use AI tools and to require a log provided by the AI tool to be provided. Adjust the assignment accordingly and use custom system prompts for the AI tools so that the students are both learning about the underlying subject and also learning how to effectively use AI tools.
Drasi does not require Azure to work. You install Drasi on a Kubernetes cluster, configure a Drasi Source to connect to a supported database, write Continuous Queries in Cypher Query Language, and configure a Drasi Reaction to do something with the output of the Continuous Query. For K8S, you can currently use k3s, kind, AKS (docs coming soon), or EKS (requires some work arounds).
The Drasi Team is most active over on discord, where we are happy to answer questions and help get people started using Drasi (https://aka.ms/drasidiscord)
Squiggle is an extension of the OpenAI Realtime API that generates insights (summarization, highlights, structured data, and content flags) in realtime as the conversation occurs. You can use Squiggle with a one line code change, no account setup needed.
The website has a link to a simple demo and an embedded video of the demo in action.
I originally built this for something I'm working on, but figured I should make it available as an independent service that anyone can use.
The insight-generation features do incur additional costs utilizing the OpenAI API, with the same OpenAI API key you are using for the Realtime API, and you can configure how often the insights are generated and thus how many extra API calls are being made.
First money in at a 40M valuation? Assuming this is pre-seed, that seems like the investor is taking on a lot of risk based on the distribution of outcomes.
"Show me the incentive, I'll show you the outcome". I do not know anything of the culture of Tesla and am not commenting on it specifically, and that quote could apply just as much to a company like Boeing. But at companies working on safety-critical products, the problems usually arise from employees acting in their own self interest to do what will get them rewarded for shipping on time, or to avoid punishment for causing delays.
Goodhart's law applies to how inflation measures have ceased to be an accurate measure: "when a measure becomes a target, it ceases to be a good measure."
It's surprising how many parents I know who are in denial about this. It must be because they themselves are constantly using social media and don't want to accept what it is doing to their own mental health.
I, Millennial, am seeing a huge decline in social media usage among my peers. It's reached the point where I didn't even bother to install any of the apps when I got a new phone last year, other than Facebook Messenger which we still use for planning activities.
The vast majority of my friends and "friends" on social media haven't posted anything for years. I think my last contribution was back in 2017.
I thought social media in general is boomer / gen X town nowadays.
Just to clarify, is the scope of social media being referred to here particular, typical apps (eg: Facebook, Twitter, Instagram), or any app/service that has interactive social and media elements (eg: Discord, Reddit)?
I really think we need to make a (large) difference between platforms focused on user interaction and discussion (Discord, Reddit, old school forums, ...) and platforms where the person behind the user is the focus and placed in front of the entire world to be judged (Instagram, Twitter, Facebook, ...).
The former is not focused on you as a person and you are judged by the discussion that you participate in, while the latter is focused on you as a person and you are judged by how interesting you are or can appear to be.
My peers are disappearing from the latter category (Instagram, Facebook, Twitter, ...)
I don't think I know the user name of any of my friends on Reddit even, or who are active there, which is exactly the point - nobody cares about the connection to the real life person.
I agree about more specific definitions of what is being referred to in discussions under the umbrella term 'social media', particularly given those pushing for regulation.
A couple things that stand out in discussions about social media negative consequences are: service-led algorithms and UX which actively fuel addictive patterns, an emphasis on a single and often IRL identity (Zuckerberg famously saying there should only be a single user identity[1]) and audience reach which is far too broad for various content.
On Discord people often become familiar with other users via their pseudonymous handles, even sharing IRL details as they're comfortable. However a few things help with this: no leaderboard-style gamification of posts like there is with Twitter or Reddit, chat is inherently lower stakes and the scope is limited to that community not the wider internet.
Traditional forums are interesting since I know various who were addicted but in a user-led rather than service-led way (ie: they've been addicted through habit of non-gamified checking of content/participation). Even expressing having anxiety posting threads due to the expectations of peers. However the benefit is still an awareness that the primary audience is mostly an in-group of (mostly friendly) peers. Whereas on Twitter for example, a user may say something for a specific intended audience but someone in bad faith re-contextualizes it and initiates dogpiling—which is an inevitability of almost any community but for sites with such broad scope is extremely difficult to moderate.
Like you mentioned, on Reddit there is almost zero emphasis on the user, whether as an OP or commenter, so there's much less interpersonal community building but OTOH much easier distancing from self 'performance'/anxiety.
I'm not a parent and hardly use social media (well, except HN!), but I am still willing to deny this, because Haidt has never properly addressed the serious demolition of his work presented by Aaron Brown. Have we all forgotten how weak and FP prone social studies are?
To Haidt's credit he did at least try a rebuttal, which is better than most situations in modern academia where outside criticisms are just ignored entirely. But he lost that debate because his responses were bad, and in this article he doubles down on the bad arguments he made last time.
He attempts to suggest that the large number of studies he collected with Twenge means there must be something there. This argument fails because when Brown randomly spot checked some of his cited studies they were all of garbage quality. Often they had nothing to do with social media at all. Haidt's response was the same as in this article: they can't all be wrong! But yes, yes they can all be wrong and Brown has strongly shown that this is likely to be the case.
Then Haidt suggests that you can derive signal from a large pile of bad studies, because if the null hypothesis were true then you'd see random results. But you can't assume that due to publishing biases, spurious correlations and other problems. For example he says that if social media didn't affect mental health, then there'd be no gender signal. That doesn't make sense. It's possible for girls to have worse mental health, and spend more time on social media, and for social media to not be the cause.
After making spurious correlation/causation arguments and denying he's doing so, he moves on to an even worse problem making his entire theory is unfalsifiable:
> much of my book is about the collective action traps that entire communities of adolescents fall into when they move their social lives onto these platforms, such that it becomes costly to abstain. It is at that point that collective mental health declines most sharply, and the individuals who try to quit find that they are socially isolated. The skeptics do not consider the ways that these network or group-level effects may obscure individual-level effects, and may be much larger than the individual-level effects.
In other words you can't even attempt to establish causality by cutting out social media use and seeing if it affects depression, because he will just immediately change his claim to assert that it was the ambient social-media-ness of society causing the depression (a much harder hypothesis to prove), and thus individual interventions can't help. And because this hypothesis is so vague, he can always fall back to claiming that whatever group intervention size is tried it wasn't big enough.
His whole essay is full of problems like this. It's basically a Gish Gallop. Vast reams of unusable paper is presented, along with panicked assertions that it's unreasonable to request valid evidence or a testable hypothesis before making law, due to the nature of the (self declared) emergency: a classic circular argument of the sort that emerges way too often in public health spheres.
There's actually a better supported alternative line of work that shows a completely different and much stronger correlation related to political ideology. But of course Haidt would rather not discuss that, as it's not amenable to legislation.
These types of tools should exclude all files and directories in the .gitignore as standard operating procedure, unless those files are specifically included. Not just because of secrets, but also because these files are not considered to be part of the repository source code and it would be unusual to need to access them for most tasks.
It's amazing to me how often Goodhart's Law can be observed in the wild, and I've come to believe it is an important enough concept that it should probably be something (along with logical fallacies and other concepts related to critical thinking) that should be prioritized above a lot of other things taught in high schools.
So when you tell execs goodhart's law, they ask, ok what can we do about it when managing large organizations? Be it a 200 person department of a 30'000 person organization or even more?
The usual argument is you need a metric and a check metric that can catch gaming the first. Eg unit sales and mean price per sale — the latter preventing a race to the bottom on price point.
You need at least a dozen to prevent gaming of the first. It's not even that hard, but you do need to be aware that you need many distinct metrics and you will bring at least one down at least somewhat, and hopefully by repeated iteration you will end up with everything better.
And what prevents that when it becomes known there are n+1 metrics to game? And your doing this in a large organization, so it is statistically inevitable that people will want to game the metrics, bad actor or not?
The best approach is really hard. (Which is why proxy metrics in the first place.)
The best approach is to have a -very- clear set of goals, and to accurately measure progress to those goals, not to proxy metrics or goals.
Yeah, I know, easier said than done.
Usually a business has multiple goals, which are complementary. Enumerating, and measuring these make the system harder to game. Or more accurately make it harder to get better metrics without also getting closer to the actual goals.
If you want sales,then measure sales. If you want growth measure that. If you want customer satisfaction then measure that. If you want happy employees then measure that. List -everything-, measure everything. Understand the tensions, improvement in one area can drop the metrics in another. So work to keep improving them all.
All of this is -hard-. So most places don't bother. Find a simple measure (say stock price or turnover or profit) and ignore sustainability etc.
In practice the government will mis-interpret any theory and attempt to build a financial perpetual motion machine that revolves around printing money. Then everyone will feel very poor while being told they are rich.
Fairly similar to how at present they have done exactly the same thing and generally try to call it Keynesianism from what I can tell. They skipped the parts of Keynesianism that would involve spending less which I suspect is a contributor to the progression of 2000-2007-2023 where every 10 years the crisis gets bigger.
The situation is not that complex. We need people to, by and large, create at least as much value as they consume. There are enormous efforts to find an alternative to that basic balance by creating lots of money and they by and large aren't actually working.
Also the rich don't consume that much stuff in absolute terms, so taking stuff away from them can't help other people to an great extent. The taxing has to be on the middle class who do most of the consuming. There are no alternatives. Can't tax the poor because they have no money, and ironically can't tax "the rich" because although they nominally have money they don't own that much real stuff as a group.
That actually sounds closer to Austrian concepts of inflation than MMT. Reducing the money supply in itself doesn't affect consumer prices, if it only affects money that was not in circulation anyway.
In order to use taxes to counter inflation (when it is defined as increas in consumer prices), you need to tax those who would otherwise spend most of the money. The most efficient way is to tax the upper middle class.
I save most of my money, but supposedly I was going to spend it anyway so actually it should be taxed? Infuriating take.
Edit: Forgot to mention the obvious, if this is how you feel, why isn't a higher sales tax on certain items ideal? People scoff at sales tax proposals because they'd target the poor, but sales taxes could be implemented progressively too. Tax new cars, smartphones, etc. More expensive items get a higher tax, reducing the impact on poorer people.
I'm not arguing for or against higher taxes. Just stating what I see about facts about the relationship between taxes and inflation under MMT thinking. (And I'm not a fan of MMT, btw.)
Inflation happes when there's an imbalance between supply and demand. MMT argues for increasing deficit spending to fund all sorts of social programs. It also argues that the purpose of taxes is not to "balance the budget", but rather to tamper the inflation that tends to follow overspending.
The alternative (increasing interest rates) doesn't work properly under MMT, since treasury bonds cannot have a real interest rate, or the deficit spending would eventually break down. That means that under MMT, purchasing power of consumers must be pulled back using taxation of various sorts, which only works if you tax the consumers.
The problem with MMT (the way I see it) is that adjusting taxes to regulate inflation is probably an even more painful approach for most people than using the interest rate.
Anyway, the fact you need to tax the actual consumers to fund social programs remains the same, even in more traditional social democratic/keynsian approaches, even if they use the interest rate to regulate inflation.
If you look at northern Europe (where I live), taxes on the middle class are much higher than in the US. They're also higher for the poor, but with the increased social benefits, free healthcare, etc, it matters less for them.
Oh, btw, a lot of the taxes collected ARE, as you propose, collected as sales or luxery taxes, as well as employers taxes (hidden income taxes) Where I live, the direct income tax is "only" around 45%. However, the base sales tax is 25%. The employers tax (tax on employers for paying salaries, in other words a hidden income tax) is 25% (if you make more than $100k). For some items, such as cars, fuel, alcohol, tobacco, cosmetics, sugar/candy, etc, there are additional taxes, often well over 100% of the base cost.
All-in-all, I suppose, out of the salaray budget that my employer allocates to me, 75-80% ends up as taxes, somehow.
And I'm middle class, not rich and not poor. If the left in the US really wants to have a European style welfare state, this is about the level that allows for that.
If given the facts about this, I doubt many American's would want to switch. On the other hand, few people over here would want to switch in the other direction, too, since they're used to the taxes and the benefits they buy.
