That assumption has come up in almost every conversation I’ve ever had with semi-technical people regarding git, so the confusion is just a fact. It happens so often, I think Linus (or whoever controlled the git trademarks at the time) should have demanded GitHub change their name when it was launched.
What if you have a few local machines you’re using for development, and want to keep them in sync? This method allows that single central repo without having to bounce all the code through a cloud hosting service.
It may have been (probably was) a conscious choice illustrating how new things were (i.e. those people didn’t grow up typing to a level where it was muscle memory). Also, keyboard layouts on early machines were far from standardized (other than the qwerty letters, almost every other symbol was not in a standard location from machine to machine), so even if you knew one machine you might not know others.
Most actors and directors put a lot of thought into small details like this, so when you see something like this it’s often intentional.
Someone responded to a previous comment of mine [0] positing a Peter principle [1] of slopcoding — it will always be easier to tack on a new feature than to understand a whole system and clean it up. The equilibrium will remain at the point of near, but not total, codebase incomprehensibility.
People are often skeptical when I say this, but there's simply no guarantee that it's possible in principle to clean up a bad architecture. If your system is "overfitted" to 10,000 requirements from 1,000 customers, it may be impossible to satisfy requirements 10,001 through 10,100 without starting over from scratch.
It's really not that big of a word. The CAP theorem shows that as few as three reasonable-sounding requirements with no obvious conflicts can be impossible to satisfy simultaneously. (User needs will start more flexible than strict mathematical requirements, of course, but once people start to build production workloads on top of your systems that flexibility is radically reduced.)
I really am surprised that people on a heavy CS themed forum still have trouble grasping this.
Imagine the year is 1995, C exists, but some guy out there is working on essentially what modern Python is. He says to you "check out this language, you can just import stuff, and use it and dynamically modify anything at run time". You can probably come up with hundreds of arguments about things that could go wrong, like memory clean up, threading, e.t.c, but turns out, incrementally, they were all solved and we have the modern Python that basically is good enough to build these large LLM models.
Now imagine modern programming and computing is what C was back in 1995, and AI use is that guy building the Python code.
You can imagine anything you want, but it’s not an argument - you could apply this to anything. “Python was successful after a dubious beginning so NFTs will be successful”
Also, Python does not build or run large language models. It orchestrates C code that does that, and it was probably good enough to do that in 1998.
Highly dynamic languages existed for decades prior to 1995, Python was not particularly innovative in its features at the time. There were also countless languages more feature-rich than C being used for development at the time.
The biggest change that happened was that hardware kept getting better and it became feasible to use garbage-collected languages everywhere including really inefficient implementations like CPython.
That being said, 30 years later Python is still slow as shit even compared to other dynamic languages and runs into all kinds of scaling issues when used for anything serious. And everywhere that performance matters, software continues to be written in typed, compiled languages including C (but also C++, Rust, Go, etc.). Even in ML, Python chiefly acts as a thin wrapper and glue language for high performance CUDA libraries (aka C and C++).
So your historical analogy is mostly anachronistic.
No, you just don't have a grasp on reality. For example, you claim that Python runs into scaling issues for anything serious, but you are blissfully unaware that youtube and uber both run python backends. Nobody cares that its "slow" by whatever metric you consider. Its fast enough. The metric that matters is developer time not compute time, because the former is vastly more expensive. Python and Node are the number one languages on github for a reason. And you are vastly deluded on how many jobs there for C++ and Rust devs lol.
In the future, you won't be dealing with strings, json, or apis. You will be importing agents, and giving them brief instructions, either in plain English or in some intermediate language higher than Python that is more brief. Wanna deal with database reliability ? Import database agent and give it brief instructions on what you want to manage. Just like you mention, right now Python is the wrapper for low level libraries, because everyone who is doing work in ML doesn't want to waste time making sure their C Cuda kernels compile. In the same way, nobody is going to care if they get the API headers right, or if their strings are correctly parsed when you can just invoke a dedicated LLM (which will likely be highly specialized small model able to run on local hardware) to do all that.
You can scream and cry as much as you want how that is bad, how its slow, but nobody is going to care because shit is going to get built faster. Ever notice how despite the massive layoffs across tech, there isn't service degradation in any sector? Good luck trying to sell your Rust skills in the future lol.
The point is that in the future, AI will be able handle things like missing databases just like the modern high level dynamic languages can import a library to handle whatever you want.
I can't tell if you're being facetious, but a future AI really may be able to fill in a missing database. Like, if it knew some of the entries, it could infer the rest.
Wow - imagine being able to infill a geophysical database with the dullest possible milquetoast totally expected signal derived from the NASVD most common eigen vectors.
The infill will look seamless.
And entirely lack any actual strikes of interest - the outliers are exceptional signal and the entire raison d'etre for building such a database.
Jeez, if AI can just infill where the gold is, why even bother to look in the first place.
>"clean up" dropped databases, compromised computers or leaked personal data?
For each of those things, you can right now build an agent that handles all of that. Or use a large frontier model with enough context to build code that ensures all of those edge cases are handled.
Future coding will essentially be like this. The concepts of dynamic vs compiled language will shift towards having frontier edge models put together code versus small runtime edge models dynamically processing data.
Frankly this is what everyone is counting on whether they know it or not. The question though is not “will the models get good enough?”. The question is does the repo even contain enough accurate information content to determine what the system is even supposed to be doing.
Yes. And as the models get better, it works better. But at one point you do have to understand the code because it's also just guessing as to what your actual intentions are.
It doesn't know what mess you want to clean up. A lot of times AI just starts making up new patterns on top of other patterns and having backwards compatibility between the two. How does it know which one you actually like?
Every frontier model from each major US lab is cheaper than their frontier model this time a year ago with the exception of Anthropic whose pricing has remained exactly the same.
They’re not describing any kind of burnout; just fatigue from working or being overstimulated. Taking a break a the exact remedy for this condition, but many people take breaks in a way that’s not actually restorative (phone scrolling, etc.)
OpenWRT updates are very much discouraged on an ongoing basis primarily because most devices running it use very cheap flash chips which are small and fail quickly after too many writes. They’re nowhere near the level of SSDs, or even SD cards, that can handle many flash cycles.
Almost as important is the fact that updates do not overwrite the original packages, because those are in a read-only partition. Updates are written to an overlay file system, so every updated package uses twice as much flash space. Installing updates weekly would quickly fill the flash.
But as far as vulnerabilities go, what’s the actual exposure? From the outside there’s no ports open, and on the inside only a few for device management, and basic services like dhcp, etc. Those have been around for decades and are pretty well hardened by now.
Pricing for any item is set by one thing: what people are willing to pay for it.
If a business raised prices because of tariffs, and consumers paid the higher price, that was a successful test that consumers are willing to pay that higher price for the item. Once that’s been established, the business has little incentive to lower prices once the tariffs go away. Prices only go down if competition with other companies pushes them down, but every player in a market has little reason to do so when they’re enjoying the higher profits.
Pricing for almost every item is set by the lowest price the producer is willing to accept, not by what people are willing to pay.
It's the "one price rule" in economics.
Everybody is willing to pay different prices. If you're starving, you're probably willing to pay "all my money" for food. But you don't, you pay the same price as everybody else who aren't willing to pay that much. The seller can't set the price to "all your money" because somebody else will be willing to sell for less.
> but every player in a market has little reason to do so when they’re enjoying the higher profits.
In that case any producer willing to defect from this implicit pact and lower their prices slightly will be able to make all the profit. Anti-trust should be ensuring there are enough producers that there's always somebody willing to goose their profits at the expense of their competitors by lowering prices.
In reality these work very well for many of the important things. Ask any farmer who sells a 60 pound bushel of wheat for $6, producer of $10 blue jeans or maker of those $400 60" TV's. They're not swimming in profit.
The exceptions are far fewer, but far more noticeable. Housing and health care don't follow the one price rule. The exceptions dominate our mindshare because they're so painful, but the non-exceptions outnumber the exceptions.
One price rule for commodities isn't super relevant for consumers though? Commodified markets are well understood so long as there is decent competition.
What we are increasingly seeing on the consumer side of the market - even on grocery items - is price segmentation. Grocery stores (moreso their suppliers) learned that many (most?) consumers are willing to pay much more for staple food items that are not commodities but quite common buys. Like chips or soda or branded packaged foods. They set a regular retail price to 50% more than it was a few years ago over time, and then to capture more of the price sensitive consumers they offer incentives like coupons, in-app deals, random sales, etc. to induce those consumers to purchase.
This is getting to be extremely aggressive and will continue to do so for the foreseeable future. Uber/Instacart for example have plenty of whistle blower insider types who have written about how price segmentation on an individual basis based on personal information and habits happens. Such as the type of credit card on file (Amex holders get charged more), how much gift card credit balance you have, your trends like accepting higher prices once from a given location/destination pair and time, etc.
If you go to the McDonalds drive-thru and simply order at the window you will be likely paying considerably more than the person who has the app installed and orders through that method.
Airline tickets perhaps follow this model as well - browser history and cookies will present a higher price to one consumer vs. another for the same book at exactly the same time. Some court cases are attempting discovery on this recently, so it will be interesting to see if true.
The price of an individual consumer transaction is absolutely set to what the company charging it believes the market will bear. Increasingly that "market" is the size of exactly one consumer.
I listened to a few earnings calls for fast food and consumer staple companies during COVID. Executives were absolutely incredulous that they could continue to increase prices and have it not impact volume of sales much if at all. What was taught in MBA school simply was not reality on the ground, and COVID times exposed this fact. The US consumer at least as a whole has simply lost the ability to price shop and is not as price sensitive as the textbooks say. This may change, but it's the current state.
About the only thing producer prices set is a pricing floor.
Yeah but, that vagary is literally the exact same wording that can be applied to "prices are set according to what consumers no; not it's value" that sparked this thread.
I don't see why this matters a single bit. You can easily flip it around and say that the businesses were clearly fine with all this because they kept importing, so why shouldn't the entirety of the tarriff refund go to consumers?
Oversimplification. Businesses can exist when the the cost of a good is less than the price they can get. There are many possible prices that this might be true, and there is some price that maximizes net profit.
When an item's margin becomes large, the risk/reward equation becomes favorable for new competition to come in. That puts downward pressure on prices.
For a given good, let's say that tariffs increased the business's cost for that good. If that cost goes away and the price stays constant, then the margin increases. That triggers more competition.
> Pricing for any item is set by one thing: what people are willing to pay for it.
Pricing is set by two things: supply and demand. Tariffs make supply more expensive, less supply is brought in, therefore the consumer must either pay higher prices or go without. Yes, they can just choose not to buy, and then the importer can choose not to import.
Every player has an incentive to lower the price: it attracts customers away from the other players.
They were able to raise their prices all at once because of tariffs. If they'd done that by simply agreeing to raise prices, it would be collusion.
Once the tariffs go away, prices would be naturally expected to fall back to their previous equilibrium because the same forces apply.
It's even more complicated than that, of course. But if there was competition before tariffs then there is competition after tariffs and you'd expect them to act similarly.
Only for optional goods! Exploitative inelasticity for necessary goods usually leads to investigations, regulations, and occasionally jail time. So it’s important to be sure to lower prices in a timely manner if you’re e.g. an egg wholesaler/retailer, otherwise you start having to declare new material risks to the SEC.
It’s not binary. Some customers were willing and some weren’t. Even if the company was able to keep selling the item profitably, it may have reduced its total profits at the higher price point (fewer sales) and would gladly revert once the tariff is gone.
This is why when the price of oil goes up, so does the price of gasoline, but when the price of oil comes down, well... When's the last time you remember gas suddenly costing a whole lot less?
Figuring out how to document stuff for others forces you to think things through at a deeper level yourself, and that’s the main point of the idea being presented here. Forcing yourself to organize your own thoughts is where the personal growth comes from.
The juice is still much less healthy. It’s the act of having your guts extract the nutrients that makes fruit healthy, because it reduces how quickly your body absorbs it. Once you make it into juice (or a smoothie) by mechanically digesting it prior to consumption, you’ve removed the need for that.
You forgot about chewing. Nobody swallows oranges in chunks. You chew and that presses out the juice. Drinking the juice and then eating the pulp is no different although it does sound silly. At that point just eat the damn orange like a normal person.
> Drinking the juice and then eating the pulp is no different although it does sound silly. At that point just eat the damn orange like a normal person.
It's less silly than taking a shot of vodka and eating an orange. Tastier too.
reply