My experience with retro computers is that things start to fail from around the 10-15 year mark, yes. Some things are still good after 30 years, maybe more, but .. capacitors leak, resistors go out of spec, etc, and that means voltages drift, and soon enough you burn something out.
You can replace known likely culprits preemptively, assuming you can get parts. But dendritic growths aren’t yet a problem for most old stuff because the feature sizes are still large enough. No one really knows what the lifetime of modern 5/4/3nm chips is going to be.
Part of why I prefer to use a search engine is that I can see who is saying it, in what context. It might be Wikipedia, but also CIA world fact book. Or some blog but also python.org.
Or (lately) it might be AI SEO slop, reworded across 10 sites but nothing definitive. Which means I need to change my search strategy.
I find it easier (and quicker) to get to a believable result via a search engine than going via ChatGPT and then having to check what it claims.
You could work for a company that open sources your work. You can consult. A very few people can make enough to live off from donations or support contracts.
But … ultimately I think it’s the wrong approach. Open Source isn’t a business, it’s a community commons. You contribute to the commons and use what others have contributed. Making it about money just means it’s now about the money, not doing something you enjoy as a member of a community.
My daily driver is a 2014 MacBook Pro. It's had a few repairs over the years, but in addition to Apple no longer supporting current macOS for it, I'm starting to feel the performance is impacting my work as well.
I'm still debating a replacement, but Linux on Ryzen is a leading option. I don't like Apple's recent work with macOS, and while the hardware seems to continue to be excellent, I'm not sure that's enough.
I've looked at Framework, Tuxedo, Lenovo ... nothing really looks great though, which is why I haven't made the jump. Yet.
Have you spent any time looking at OpenCore Legacy? I briefly was using it on my 2014 13" MBP to install whatever the latest version-1 was. It worked well for a bit but I ran into an issue after an update. I bailed on it but learned later that it was a relatively easy fix. Might give some new feature support to you.
I really think this era of MBPs was the best. I'm kind of shocked how well my 2014 still works.
I've got a 2010 iMac that my kid uses running Sonoma (macOS 14) via OpenCore Legacy and it works great for him. I'm currently waiting on a wireless card to give him Bluetooth 4 so he can use Airdrop.
I've looked, but never played (this being my primary personal device, it's not something I want to have out of service).
I guess that's probably also why I feel like this is the wrong solution for me. I don't want to have to deal with issues on this laptop. If I pull a system update, I want it to install without a problem, automatically, overnight, and not require hours of debugging and reverting and then finding a fix a week later.
I appreciate the effort and skill that goes into making it, and if this was a less critical device for me it might well be the right solution.
Also, I agree re: peak MacBook. The keyboard is good, the ports are good, they're tough, and I really can't believe that the performance has been enough for over 10 years. It'll be a sad day when I move on.
> I don't want to have to deal with issues on this laptop. If I pull a system update, I want it to install without a problem, automatically, overnight, and not require hours of debugging and reverting and then finding a fix a week later.
This is why I never made the jump to Linux. Around 2017-2019 Apple's Laptops had gotten really bad. I was considering moving to Linux, but I've had several situations where I've run normal Ubunutu updates from Canonical and found myself unable to boot. I don't know if this is common, I don't think I was doing anything unusual (stock Ubuntu).
Have you considered kicking tires on Linux installed on your MacBook?
I have a 2018 Mac mini running Ubuntu well. It actually dual boots macos as well using the refind bootloader. Refind can be installed and booted from a USB key as well so pretty straightforward to try. Make a empty partition from macos for Linux. Then boot Linux iso and install there. Boot refind and you can select which os to use. The 2018 mini even supports esxi and windows if you want to quad boot
Apple builds their devices for consumers not developers. With each iteration people are increasingly locked into that ecosystem. And the hardware is designed to be obsolete within 5 years or so.
Laptops in general are built to a price point. Personally, I only find the gaming laptops sufficiently appealing, but the price tags put me off.
Have you considered a desktop / minitower for the bulk of your work? Multiple LCDs are great for productivity. Then a ChromeBook might suffice for on the road use.
It's funny: there was a time when Apple understood that it could attract customers by making its devices the first choice for developers. And it did a good job, with a real Unix underneath that wasn't overly hidden, with a lot of open source support, and just generally being an ok citizen in the developer world.
That strategy, which I think ran from early OSX through until mid-2010's? worked, and you'd see the results, with a sea of glowing Apple logos, even at Linux conferences.
I think they've changed since then. It feels now like developers are not a priority, even as they've re-added some power-user features in recent years.
As for longevity, I think my current laptop speaks for itself. It's now 11 years old, and going strong. I've carried it around the world, dropped it numerous times, and it's been the best laptop I've every owned. They do artificially force an end-of-life, usually at around 7 years old, for their operating system. I don't really have a problem with that: I appreciate that it's difficult to continue to support old hardware in new releases, and I think that timeframe is reasonable. I'd prefer that it was explicitly stated, and I'd prefer it was longer, but ... it's not bad, really.
Over the years I have had many devices: desktop PCs, Unix workstations, Linux workstations, and laptops from 6" to 17". In the early 2000s I bought a 12" Powerbook, which was my first Mac -- the motivation was to get seamless Microsoft Office document support on a Unix system, and it did a fine job of that. Eventually I spilled water on it, and replaced it with a 13" unibody MacBook, which I later upgraded to the current 15" MacBook Pro.
I currently have a work-supplied dual Xeon 56 core, 192GB RAM, 8TB SSD monster Dell with 2x27" screens on my desk which is used for my day job compiling a large proprietary C++ application. That's the right kind of machine for that job, and I wouldn't do it on a laptop by choice.
But for my personal stuff, while I'd previously set up my laptops as a secondary device to be used when travelling, when I got the 2014 MBP, I deliberately upgraded to the 15" screen, maxed out the CPU, and got a big SSD so I could make it my primary device. And that's been a model I have liked a lot.
I don't travel anywhere near as much these days, and could probably revert to a desktop instead and just access that remotely via Tailscale when I travel, but ... I like having everything in the device in my hand.
But it's a good point: I should probably reconsider.
I think this is both true and yet completely insufficient.
There’s a whole system that produces the people and materials that enable every field of endeavor. Having the books and the eg. sub-nanometer scale silicon wafer etching machine does not get you a new iPhone.
Yeah, I agree. The whole chain is long and no one really knows everything. Even China doesn't produce everything albeit it is the manufacturing powerhouse of the world right now.
Since I live in Canada, a country that needs to import a lot of manufactured goods from foreign countries, I'm worried that a reboot is not possible. Adding the US into the picture improves quite a bit, but again US has transformed from a manufacturing giant to a financial one for many decades.
I was surprised a couple of years ago to read that the US no longer made the aluminum tubing used to make bicycle frames. Despite having quite a significant custom frame building scene and many craftspeople, such a simple raw material was no longer made in the US.
I'm sure there are many, many things like this, for most countries.
I agree that it looks grim, but I do think preserving a full set of documentation for every industry helps the future generation to reboot. The resource part is not solvable, I have to admit that. The only hope is that with much less population and a forever mind of scarcity we do not need that much resources.
Actually, taking away the virus part, looks like we are dead if we can't get out of earth to mine other planets economically before the resources are depleted. But as far as I heard, that is super expensive. Maybe colonization with terraformation is the only answer? But we are so far from it.
The current generation however will struggle: there won't be an Internet, so all the information will be in paper books. And finding what you want when presented with a library without a paper catalog ... there's a whole bunch of people needed to rebuild indexes that had moved online.
I'm confident that small numbers of people can survive with the resources available. Maybe tens or even hundreds of thousands in total, in medium-sized groups on fertile land and in good climates. But I don't think rebooting to 21st century technology would be possible inside hundreds of years.
Isn’t the Dewey decimal system universal so that the same book will have the same reference no matter which library it’s in? So all it would take is one analog index somewhere and it would apply to every library.
Even without it, there are sections that are labeled in libraries that require no index.
Definitely true for coal and petroleum (and those are big ones), but are other resources much more available? (e.g. metals could be "mined" from excess infrastructure)
I'm absolutely no expert here, but tin comes to mind. It was a critical ore with limited sources during the bronze age, and required a substantial trading economy to exist to allow bronze-based technology to flourish.
Extracting resources from the detritus of the 20th and 21st centuries would surely be a major activity. But to efficiently do so requires a bunch of power and chemistry and reagents/catalysts that are themselves probably a heap of work to acquire.
Getting iron and mild steel from rebar, fallen bridges, rail lines, cars, etc, would likely be plausible though. It wouldn't be high-quality steel, but for blacksmithing-type stuff, it'd work ok. Probably sufficient for sand-casting too? Aluminium and copper are probably recyclable too, although I'm not sure how aluminium would work with different source alloys?
It would be an interesting research project to focus on one of them, say coal, and gather as much public information about each major coal mines, as possible.
You can replace known likely culprits preemptively, assuming you can get parts. But dendritic growths aren’t yet a problem for most old stuff because the feature sizes are still large enough. No one really knows what the lifetime of modern 5/4/3nm chips is going to be.
reply