Did you only set up Time Machine? Or did you continue supporting all those users for years and years. If the issue is that eventually the backup store becomes corrupted then you may not see it at all if you're only setting up backups but never dealing with users who actually need to restore something from backup years later.
I'm convinced that this is the fate of all successful software companies. It's not a result of arrogance or hubris or anything else like that. It's the result of turnover.
Take your favourite rock band and turn over all the musicians until no one is left from the original band. Should we expect the band to continue cranking out chart-topping hits?
There's one further factor that makes the situation even worse than the "Rock Band of Theseus." That's the fact that young software engineers are not interested in stewardship. They want to build their own projects, not fix bugs in someone else's. Across the software industry we see this lead to a continual churn, rewrites and redesigns no one wants, and a huge amount of wasted effort reinventing the wheel (and often making a worse wheel).
I have been so irritated by this that I've been considering switching TO the mac ecosystem, BUT this thread is good on my eyes and makes me disinterested now.
It really makes me miss Classic Mac OS 9. I used it from 7.5.1 to 9.2.2. I remember being so excited about Mac OS X when the Public Beta came out that I switched immediately. It really sucked and I went back immediately. But eventually Mac OS X got better and I switched to it, and never looked back.
Now I am looking back and remembering everything I lost. A computer that was so simple and so predictable. It didn't change behind my back all the time. It never shoved upgrades down my throat. It just worked!
"A computer that was so simple and so predictable."
This is what cachyOS + KDE is giving me at the moment. Ok, so it's not totally simple and there are A LOT of updates. But it's by and large predictable. I never had a 1980-90s Mac, but I had an Apple IIe and an Amiga 500. While cachyOS is so much more powerful it doesn't abuse that power like Windows and OSX with so many background processes and telemetry. I have a Mac laptop and I dual boot my PC with Windows + Linux. I don't have hate OSX but CachyOS + KDE is by far my favourite as it's customisable to the extent I want and it just gets out of my way. Highly recommend it if that wasn't obvious!
If you've lived in your house/apartment for a good long while and settled in, you have an idea of what it was like to use Classic Mac OS 9 (and earlier).
It's like flicking a lightswitch or reaching into a drawer and grabbing a spoon without looking. Everything is always right where you left it. Double-click a folder and the window opens in exactly the same state that you left it when it was last closed. All the icons are arranged in the same way, with the same label colours you gave them, and each of the folders inside that folder open the same way as well. One folder might open in list view sorted by Date Modified while another opens in icon view with the exact arrangement you decided on, all according to the way you left them.
All of those folders open their windows in the exact same position, size, and shape they had when you closed them. This lets you quickly drill down through layers of nested folders, moving your mouse to the next one before your eyes can even register it on screen.
The effect of this extreme level of persistence is that you develop muscle memory for the mouse. No other operating system environment I have ever used works like this, or at least this pervasively (modern macOS still has this for the menu bar). Everyone else just gives up and relegates the muscle-memory control to the keyboard only. This is a huge tragedy! A Classic Mac OS power user works with one hand on the mouse, one hand on the keyboard, and uses muscle memory with both to fly around the UI and work very efficiently. This is especially valuable when you're working in software that needs the mouse anyway, such as art or design software.
Having a confidence score isn't as useful as it seems unless you (the user) know a lot about the contents of the training set.
Think of traditional statistics. Suppose I said "80% of those sampled preferred apples to oranges, and my 95% confidence interval is within +/- 2% of that" but then I didn't tell you anything about how I collected the sample. Maybe I was talking to people at an apple pie festival? Who knows! Without more information on the sampling method, it's hard to make any kind of useful claim about a population.
This is why I remain so pessimistic about LLMs as a source of knowledge. Imagine you had a person who was raised from birth in a completely isolated lab environment and taught only how to read books, including the dictionary. They would know how all the words in those books relate to each other but know nothing of how that relates to the world. They could read the line "the killer drew his gun and aimed it at the victim" but what would they really know of it if they'd never seen a gun?
I think your last point raises the following question: how would you change your answer if you know they read all about guns and death and how one causes the other? What if they'd seen pictures of guns? And pictures of victims of guns annotated as such? What if they'd seen videos of people being shot by guns?
I mean I sort of understand what you're trying to say but in fact a great deal of knowledge we get about the world we live in, we get second hand.
There are plenty of people who've never held a gun, or had a gun aimed at them, and.. granted, you could argue they probably wouldn't read that line the same way as people who have, but that doesn't mean that the average Joe who's never been around a gun can't enjoy media that features guns.
Same thing about lots of things. For instance it's not hard for me to think of animals I've never seen with my own eyes. A koala for instance. But I've seen pictures. I assume they exist. I can tell you something about their diet. Does that mean I'm no better than an LLM when it comes to koala knowledge? Probably!
It’s more complicated to think about, but it’s still the same result. Think about the structure of a dictionary: all of the words are defined in terms of other words in the dictionary, but if you’ve never experienced reality as an embodied person then none of those words mean anything to you. They’re as meaningless as some randomly generated graph with a million vertices and a randomly chosen set of edges according to some edge distribution that matches what we might see in an English dictionary.
Bringing pictures into the mix still doesn’t add anything, because the pictures aren’t any more connected to real world experiences. Flooding a bunch of images into the mind of someone who was blind from birth (even if you connect the images to words) isn’t going to make any sense to them, so we shouldn’t expect the LLM to do any better.
Think about the experience of a growing baby, toddler, and child. This person is not having a bunch of training data blasted at them. They’re gradually learning about the world in an interactive, multi-sensory and multi-manipulative manner. The true understanding of words and concepts comes from integrating all of their senses with their own manipulations as well as feedback from their parents.
Children also are not blank slates, as is popularly claimed, but come equipped with built-in brain structures for vision, including facial recognition, voice recognition (the ability to recognize mom’s voice within a day or two of birth), universal grammar, and a program for learning motor coordination through sensory feedback.
Friendly reminder that Google Takeout [1] exists. When I read a story a few years ago about a guy who had his primary Google account banned with no recourse (for reselling Pixel phones) and permanently lost 20 years worth of emails and family photos, I researched and found Takeout and used it to back up all my data, then subsequently stopped using Google services altogether (apart from YouTube).
Unfortunately the service is very buggy in my experience. When I tried to download all of my photos data multiple times it gave me corrupted .zip files and half of the files were just zero bytes. Maybe I can blame Firefox for that though, I dunno. I should probably try again with Chrome before completely blaming Google
I've never had a problem with Google Takeout the multiple times I've used it. Perhaps try making the compressed files smaller (You can choose to make them 1gb or greater, last time I used it), you might need to download 75 files, but it's better than 1 big file.
That's assuming the hydrogen is just loose in the area, like it'd been released from a balloon in a chemistry classroom. That amount of hydrogen is extremely small, from an energy standpoint. Equivalent to a teaspoon of gasoline or so.
If you assume a realistic fuel capacity for a hydrogen vehicle, the hydrogen tank will be both much larger than a gas tank and the hydrogen will be under extreme pressure. A tank like that in your car would be extremely dangerous even if it were filled only with inert gas.
What if instead of allowing the agent to act directly, it writes a simple high-level recipe or script that you can accept (and run) or reject? It should be very high level and declarative, but with the ability to drill down on each of the steps to see what's going on under the covers?
I wonder if there is a viable business model where for each article, readers can pay to unlock it not just for themselves, but for everyone. The price would obviously have to be higher since you aren't just buying it for yourself. But perhaps the sense of "I'm helping build a better-informed world and helping broadcast my values" would encourage people to pay that higher price.
What you're asking for is the system we already have, except at a micro level rather than a macro level. Rich people buy out newspapers to signal-boost their own preferred messages to the public.
I think it's questionable that the "news that people feel is actually valuable" is what really ought to be spread. Some of the most valuable news is local reporting on the daily business of municipal governments. Regular people are notoriously uninterested in local politics, despite the outsized impact it has on their lives. Many of the most mundane decisions made in municipal councils go completely unnoticed yet they can destroy whole communities in the long run.
>Many of the most mundane decisions made in municipal councils go completely unnoticed yet they can destroy whole communities in the long run.
They go unnoticed because of scaling issues, not because people are per se less interested in local politics than national politics. If you write a story about a decision on the local city council, it is of interest to maybe a few hundred thousand, whereas a story about Congress is of interest to tens of millions. Even if people were ten times as interested in local news (as measured by their willingness to subscribe or the amount of ads they are willing to be exposed to), it would still make more sense to send a reporter to the Capitol before City Hall.
I think a lot of our issues today are because people are too engaged in federal politics. It's turned into a massive spectacle on the same level as the NFL.
> Everyone has the same and equal meaningless vote.
There is one vote that is not meaningless: the primaries. A lot of the issues y'all have is that Democrats and Republicans alike don't bother to vote in the primaries. That is how you got people like MTG or Trump, that is how you get people like Chuck Schumer stuck in office for far too long.
AOC/The Squad and Mamdani both proved that it is possible to succeed in a primary and offer voters an actual alternative to the corporate owned shills.
> not because people are per se less interested in local politics than national politics
Actually I believe this is exactly the issue. Most people are interested more in national politics than county or even state politics. Of the people I know who vote in national elections, very few vote in local ones or even go to city council meetings.
> Rich people buy out newspapers to signal-boost their own preferred messages to the public.
Right, but micro level difference matters here. If a middle-class person can help an important story reach an audience, that's helpful for democracy. When a billionaire buys a newspaper, it isn't.
This is also why I think suggesting it work like a kickstarter where multiple people can pool money to unlock an article would be helpful. It naturally collects the will of many people in a democratic way.
This is basically political fundraising, where rich people signal-boost their preferred candidate and help them attract lots of donors from the public.
I think the fundamental piece you're missing is the Pareto principle. In any popularity contest, the most attention accrues to the most popular. This naturally leads to a power law distribution in popularity.
The solution to increasing interest in local government is to strengthen the federal system (repealing the 17th amendment) & even extending it into state government (state senators should be appointed by, say, the city councils of the 24 largest municipalities in said state).
This decentralization of power would bring the peoples' focus back to their own neighborhoods, where they can actually hold government officials accountable.
That’s a recipe for corruption and minority rule. It centralizes, not decentralizes, power, in the hands of fewer people. I would advise looking to effectively any senatorial appointment from the gilded age to see why the 17th was needed: monetary exchanges for senatorial seats was widespread, race-based disenfranchisement was a reality, and in one state, Utah, a theocracy was nearly cemented. A greater focus on local rights, and greater federalist powers, should not preclude senatorial elections.
That’s assuming a developer owns all the codebases they work on. Many don’t, because they work for an employer with a large team and a large existing codebase. If the language evolves and the project manager is 100% keen on adopting new features then it’s going to affect everyone on the team.
reply