Imagine a disaster causes widespread collapse, where resources are no longer easily available, power and internet access is erratic or non-existent, and then things calm down enough to where the necessities of life are available and society tries to resume functioning.
Now, suppose you'd like to make computers work. What sort of things would you need?Power, obviously, or a way to charge batteries. What else?
What if you wanted to have communication with other people, or perhaps data communication? Would you want packet radio? HAM gear?
What if you want to repair equipment? I don't know if the computers in cars "break", for example [not that I imagine fuel being available]. I wonder about what sort of needs there might be to repurpose gear.
What if you need to keep an application running? Maybe you have a mesh network and people want a web app for some purpose. But, without documentation, it'd be incredibly tedious.
I can't help but think I've seen a linux distro designed for this scenario, where you have binaries, and all the code, and all the toolchains, and all the documentation.
The right answer might be "don't bother", but I'm still curious.
Basically, the massive prevalence of USB over the past couple decades is your friend.
I would store local copies of the standard fare like Project Gutenberg, Wikipedia, instruction/repair manuals, and textbooks (especially math and science ones). For particularly important stuff, like survival guides, I'd even look at paper copies. Perhaps even laminate a few maps and navigational charts.
I'd also try to have several copies of complete Linux package repositories or at least all of the packages that I know I'll need. If/when backbone infrastructure starts crumbling, it'll cascade. Even downloading a text-based browser could prove an impossible task.
Even just having a hand radio could prove vital. Just be aware of how you might power it (and any audio equipment it might need).
https://www.amazon.co.uk/Knowledge-Rebuild-World-After-Apoca...
I’m guessing repair knowledge would be more important in the short to medium term.
Outside of that I think you need to work your way back up the tech chain starting with the wheel, leapfrogging as much as you can. But you don’t want to leapfrog over early tech like industrial machines, transportation, etc. and start working on web apps too early.
https://100r.co/site/working_offgrid_efficiently.html
Also, sneaker net has been surprisingly effective in Cuba.
Remember when Maersk, the shipping line, shut down for over a week due to a cyber attack? The pipeline shutdown? Several hospital shutdowns. Expect that all at once.
Internet enabling power grid protection relays now looks like a big mistake.
Presuming hardware's running and communications are just spotty; then having your own local archive of sources is a good idea. mirror gutenburg and the wikipedia dump and stuff like that as well as the full repositories of some of your favorite distributions (source packages too!)
Access to internet-level knowledge in a pocket-sized device when the internet is gone is going to be a massive advantage.
Could go with a specialised solar-powered e-ink device for power concerns.
Micro: ham radio
Since you mention computers in cars breaking, I am guessing your preferred flavor is nuclear winter.
Check out this study by the EMP Commission.[0]
Also I am having trouble imagining the average Joe in a post apocalyptic scenario wanting a web app for anything.
[0] http://www.empcommission.org/docs/A2473-EMP_Commission-7MB.p...
Given that, i would think the big factor in building anything post-apocalypse is communication. So i guess my answer would be paper and pens. Hard paper copies of as much information as possible and blank paper to write letters.
Computers are the end product in a long long tech tree. So the question just seems sort of non-sensical to me. Like if we get to the point where we rebuild society enough to even consider computers, we have already solved food, shelter, heating, manufacturing, mining etc. At that point its mostly a matter of information - just archiving as much info as possible on stuff like silicon manufacturing, linux source code, electronics etc. would be enough.
[1] https://www.econlib.org/library/Essays/rdPncl.html?chapter_n...
Have any simple-ish tech stack, and have many copies of the SAME thing.
ie: Get N-ipads, N-rasperrys, etc.
Then N copies of whatever you need to power, connect, protect, etc (cables, routers, ups, batteries, etc).
Then N copies of the data in N storages
--
This is probably the best course: REPAIR anything will be near impossible.
They were projects like decentralized WiFi networks, mostly on cities (I remember one in my city), maybe people need to remember them
Specifically why? Paint me a scenario.