Because neither product has any consistency in its results, no predictive behaviour. One day it performs well, another it hallucinates non existing facts and libraries. Those are stochastic machines
I see the hyperbole is the point, but surely what these machines do is to literally predict? The entire prompt engineering endeavour is to get them to predict better and more precisely. Of course, these are not perfect solutions - they are stochastic after all, just not unpredictably.
Prompt engineering is voodoo. There's no sure way to determine how well these models will respond to a question. Of course, giving additional information may be helpful, but even that is not guaranteed.
Also every model update changes how you have to prompt them to get the answers you want. Setting up pre-prompts can help, but with each new version, you have to figure out through trial and error how to get it to respond to your type of queries.
I can't wait to see how bad my finally sort-of-working ChatGPT 5.1 pre-prompts work with 5.2.
It definitely isn’t voodoo, it’s more like forecasting weather. Some forecasts are easier to make, some are harder (it’ll be cold when it’s winter vs the exact location and wind speed of a tornado for an extreme example). The difference is you can try to mix things up in the prompt to maximize the likelihood of getting what you want out and there are feasibility thresholds for use cases, e.g. if you get a good answer 95% of the time it’s qualitatively different than 55%.
They knew who purchased those devices. Did they know that at the moment of detonation only military personnel had those devices on them? Military propaganda of course will nod at “intelligence” to defend any actions in public, as there is no way to prove these statements.
Not only military leadership was killed, there was a significant amount of civilians being harmed.
Even if you drop a bomb to target a military personnel, but you drop it in the middle of busy city, this will be a war crime, as you didn’t do anything to avoid civilian casualties, and disregarded them.
I'm using source code like it's used when referring to source code vs executables. React doesn't simply spit out HTML, nor the JSX used to write said React code, it outputs a mixture of things that's the optimized HTML/CSS/JS version of the React you wrote. This is akin to source code and the optimized binaries we actually use.
Perhaps the wrong usage of "source code". I probably should've been more precise. Forgive my lack of vocabulary to describe the difference I was referring to.
There were no binaries or packages. You wrote the HTML in notepad or maybe you used some "high speed IDE" with syntax highlighting and some buttons like Dreamweaver and then uploaded it via FTP to whatever server you were hosting it on. No muss, no fuss. It was a glorious time and I miss that internet a lot.
> Charging more for publicity transit during peak hours won’t make people use it less, there’s a reason why so many people commute during peak hours
You don't necessarily need all people to use it during peak hours, just "enough" people. There are people who do have flexible schedules, but they may simply may not have had enough motivation to change old habits (yet).
I actually think riding on a crowded train would be more deterrent than a fare increase, so I feel like that would be needlessly punishing people already suffering the full trains because they have to.
See sibling comment by eru: they said there isn't a fare increase congestion charge; instead "Every once in a while they experiment with discounts for off-peak train usage"
The point of a congestion charge (whether on driving or on public transport) is to alleviate the congestion 'punishment'.
As a mental model: congestion works a bit like an auction. Getting from A to B during rush hour brings people some benefits (otherwise, they wouldn't bother). Benefits differ between people. But that travel also costs, both in terms of fares and in terms of annoyance and perhaps delays.
So We can imagine every prospective commuter weighs their benefits against the costs. I say it works like an auction, because there's limited capacity, and the people who are willing to endure the most price + annoyance are going to 'win' the auction and will commute. Everyone else shifts their commute around or stays home.
The people you mention who 'have to' travel during rush hour are just the ones willing to bear the highest costs in our model, and thus they 'win' the auction. (Winning an auction isn't necessarily good..: after all, you have to pay the price.)
Having a congestion charge means different people can bid not just with their tolerance for annoyance and delay, but also with actual money.
So now people compare fare + congestion charge + annoyance against their benefits. Assuming benefits on the right side of the equation have stayed about the same, the breakeven point for 'annoyance tolerance' is going to be lower, just because the other part of the left side grew.
A similar example: if tomorrow your local Walmart was handing out free 20 dollar bills with every purchase, you can bet your hat that pretty soon the queues for the cashier at that shop would grow until the wait-in-line for the marginal customer cancels out the free 20 dollars. Keep in mind that the marginal customer isn't the average customer: the queue would mostly be made up of people who have more time than money.
Conversely, in our congestion charging scenario the winners of our auction will tend to shift towards the crowd that has more money than time (or tolerance for packed trains).
The nice thing about letting people bid with money is that afterwards someone else has the money and can spend it. When you bid with your capacity to endure frustration and delay, no one else gets any benefit from that. It's just a poor waste of society's resources.
reply