Hacker News new | past | comments | ask | show | jobs | submit login
Collapse OS – bootstrap post-collapse technology (collapseos.org)
213 points by rewir 3 months ago | hide | past | favorite | 116 comments



If curious see also

2020 https://news.ycombinator.com/item?id=23450287

2020 (a bit) https://news.ycombinator.com/item?id=22901002

2019 https://news.ycombinator.com/item?id=21182628

Sharp readers of the FAQ* will note that those 2020 posts mean we should really mark this one as a dupe, but there's something so perfect about prepping meeting operating systems and the language being Forth...I just can't do it.

* https://news.ycombinator.com/newsfaq.html


It wasn't posted this year yet. ;)


Genuine question: if we're in a collapse scenario that is so devastating to the global supply chain that we're scavenging microcontrollers, how do we generate electricity to power them? Solar panels and batteries don't last forever and would presumably require the same or similar supply chain as the microcontrollers. This is a cool project regardless, but in such a catastrophic scenario it seems easier to just give up on computing and go back to more primitive technology. If I can't get a single thing delivered, I see myself collecting rainwater and gardening, not compiling Forth to an old Z80.


You can scavenge car's alternators and use a water wheel or windmill to turn it. Alternators can work for a decade with minimum maintenance. Just have five or six stashed, just in case. They can supply 40A at 12V, or using a transformer to up the voltage, 4A at 120V.


Have you successfully run any modern hardware off of the irregular current coming out of something like that? You could charge batteries off of an alternator with a voltage regulator but I doubt any modern electronics are going to work well on current supplied by a jerry rigged water wheel and step up transformer. You’d be reliant on a lot of ICs to do it properly at a small scale.


Would it be viable to use a rig of leyden jars to smooth out power coming from a scrappy alternator/commutator, and to implement cut-outs?


Alternators are also trivial to build by hand.


Sure. If you have the input materials. If you're at the point of having to build an alternator by hand rather than just taking one off a car, I'd like to know where you're getting magnet wire, the machined shaft, the various rotor and stator parts, the end-bells, and the ball bearings.


Consider that you're not necessarily building from raw materials, you will be repurposing things you find around. Finding an alternator probably won't be a problem in the populated areas of USA but might be elsewhere - thankfully building it is still sufficiently simple.


In most reasonable collapse scenarios I could imagine, energy isn't so much a technological concern as it might be a political concern. Slightly scarcer perhaps, but there's a ubiquity of old alternators, plenty of shit to burn, and I guarantee the extractive industries will be among the first to come online, if they ever went out in the first place, and can be coaxed to function in a degraded state without globalized supply chains or too much esoteric knowledge.

Computers are much more difficult to bring back: much of their manufacture and distribution is highly dependent on a globalized supply chain, which in turn is dependent on the security guarantees of an international order backed by military force. Many of their parts are proprietary by nature, and individual components can at their simplest require rare expertise in multiple niche fields of science before their production can even be considered.


> energy isn't so much a technological concern as it might be a political concern

This is what irks me about these kinds of "collapse fantasy". Somebody described the apocalypse novels of John Wyndham (Day of the Triffids etc) as "cosy catastrophe", and that very much applies here. The laser focus on technological detail-hoarding while having blinders on to any consideration of the politics of the situation, or what might be called "material conditions".

Historically few societies collapsed without external pressure. Societies do not collapse, they are collapsed. And the force that's doing the collapsing is the thing you need to worry about.

It also can take a very long time to wane; while Rome may have split from the Roman Empire in 395, the eastern successor Holy Roman Empire lasted until 1806. And the microstates provide even weirder examples, like the Maltese feudal knights with a WW2 air force!


Yeah, the brief thinking I've ever done about the collapse in these scenarios is to potentially band together with some like minded friends and try to setup some subsistence group with hunting, light farming and security.

Firearms are highly regulated where I am from which makes the first and last a little more difficult than they potentially could be.

I really never gave it that much thought to be perfectly frank. Though the point is I feel semi conductors are pretty far down the order of priorities.


Light farming and hunting won’t work unless something like >50% of the population dies off quickly. Even then you’d still need to have everyone spread out pretty evenly across the countryside, be able to learn how to plant and do it on time. You’d be going back to subsistence farming which is extremely hard work. Fertilizer and fuel for tractors would not be replenished. You’d have to learn how to save seeds, irrigate without power and pumps, eventually how to plow with draft animals. If you aren’t already doing this now chances are you wouldn’t make it. That’s the problem I have with collapse fantasies, you’re either all in now or you’re dead. The safer bet is on working to prevent a collapse since that’s something you can do now that probably has immediate payoff unlike going all in on prepping.


In the UK we only just survived the WW2 blockade, with a highly centrally planned push to expand local agriculture to cover missing imports. Loss of imports forever will kill over 50% of the population, slowly.

The problem with transitioning to draft animals is preventing them getting eaten in the collapse. We'd probably also run out of trees in a few years.

(Collapsists should make more study of WW2, such as Beevor's magnificent, horrifying Stalingrad. Or the siege of Leningrad. Surrounded Soviet civilians died by the millions. And that wasn't a collapse in the collapsist sense, the social, legal and technological order remained somewhat intact)


Firearms are highly regulated where I am from which makes the first and last a little more difficult than they potentially could be.

In a collapse they won't be, and those willing to seize them, will be at a great advantage, police dept. military bases, even countries whose population don't have access to an abundance of guns doesn't mean there isn't an abundance in the country.

In almost any scenario where government collapse can occur, people will become the largest threat to survival.

The problem with security against other humans in the long term, is that we are damn clever, and determined. Even with an overwhelming firepower, if you are in a fixed location, people will find a way.

As for Hunting, it can be done without firearms, Crossbows and Traps can be highly effective, and simply constructed. What you have to think about is how much and what sort of wildlife their is around you, and how rapidly they will be depleted by others.

As they say, Hell is other people, best bet since the time of the plague, you want to survive go where others aren't.


Crossbows require a permit as well :) Can't have the population undermining that monoploy on force ;)

But I get your point.

Like I said I hadn't given it much thought, though if one could get a sustainably large commune group with reasonable access to weaponry it would hopefully be a deterrent for most small groups.

I'm in Australia which rather large. Clean water is the likely the most valuable commodity. It would be interesting to see how much of the population would actually survive such a collapse.


The "sustainable" "collapse" (again, what does this mean? end of imports including oil? large scale political turmoil with no clear winner?) level of Australian population would probably be larger than Aboriginal pre-invasion times (estimated 700k?) but not much more than that if you lose access to mechanically pumped water. It's not a very habitable place but the introduction of European crops and animals has irrevocably improved this. You're still looking at the likely death of nineteen out of every twenty Australians in this scenario.

(actually, having thought about this, the "collapse" scenario I'd be most worried about in Australia is average temperatures above human survivable levels plus wildfires. Not so much a collapse as a thorough incineration)


The last 12 months have been quite pleasant :) Plenty of rain and we only just had our first real heat wave, which was unpleasant but nothing insane.

The centre of Australia is positively hot, always has been.

When i was refering to sustainable i meant the size of the "survivor commune", you can only realistically control a small area for sustenance with out significant order and coordination, think more tribe than state. There are millions of kangaroos i suppose :)


If you have a collapse, I don't think permits on weapons are going to be a large concern. And after the collapse, they are one of the more simple to craft, and easy to operate weapons.

MadMax should be your go to guide right? (just kidding)

Large targets, can be whittled away over time, you basically become under siege, your attackers can leave anytime in small groups to gather supplies, you would be stuck at your parameter.

Think castles, after a while, they all fall through starvation, contamination, sabotage, etc..


> Hell is other people, best bet since the time of the plague, you want to survive go where others aren't.

I have played for a while on a Minecraft "anarchy server"[1]. You could consider it a very, very rough approximation of how anarchy would play out in real life. Your statement (if you are in a fixed location, people will find a way) closely mirrors my experience.

Even though, in the Minecraft natural environment, there are things that are dangerous (e.g. wild animals, "monsters", ravines, etc.), nothing comes close to the danger of meeting another player, possibly armed.

If you are a new player, the most challenging part is escaping spawn (the area where new players start from, which is also the only crowded area in the game). This is pretty hard because spawn is a wasteland with no natural resources left (in Minecraft you have a hunger meter so if you can't find food you will eventually die), and in addition spawn is populated by armed players who kill new players as a pastime.

Once you get to about 20 kilometres from spawn (20k blocks in Minecraft) you start finding food (trees from which you can get fruit, wood for building fishing poles, seeds that you can plant to start a small farm). However, you can't stay in the same place for long, because in a few weeks at most, some other player will find your base and steal your stuff and/or destroy the base.

Once you get to about a hundred kilometres from spawn (100k blocks in Minecraft), you can build a temporary base and hope that it will last a few months (mine lasted about 3 months).

Currently I have a base that is over one million blocks from spawn (i.e. more than 1000 kms) and I expect this may last for a longer period, maybe even years, so I'm building larger structures.

Obviously this is just a Minecraft server and doesn't accurately reflect real life.

Some elements are realistic, for example travelling speed[2], or the way natural resources are either renewable (plants, animals) or non renewable (minerals).

Some other elements are not that realistic, for example the number of players on the server is way smaller than the population of a typical city. This means that the "safe distance" from spawn might not reflect the safe distance from a major city in real life.

Also, spawn is the only area with a high population. There are some larger bases across the map, but even the largest ones have a population of a few dozens players at most.

Another element that is not realistic is obviously construction. In real life, building a home or a farm means a lot of hard work, whereas in Minecraft you can do it in a few hours.

[1] https://en.wikipedia.org/wiki/2b2t

[2] walking speed is about 5 km/h, travelling through a "nether highway" you can go at about 45km/h so this could be compared to motorised transport in real life


> If you are a new player, the most challenging part is escaping spawn (the area where new players start from, which is also the only crowded area in the game). This is pretty hard because spawn is a wasteland with no natural resources left (in Minecraft you have a hunger meter so if you can't find food you will eventually die), and in addition spawn is populated by armed players who kill new players as a pastime.

Hmm. Is cannibalism an option?


No because in Minecraft a killed player just disappears in front of your eyes. However you can steal whatever items the player had in their inventory the moment they died (those items are dropped to the ground upon death).


It was Brian Aldiss who coined that description, and I despise it. It's as though he never actually read any of Wyndham's books...


That’s the western HRE. The ERE only survived until the 1400’s.


Nitpicking: If the Holy Roman Empire is seen as a continuation of the Western Roman Empire, then the Russian Empire can be seen as the continuation of the Eastern Roman Empire (e.g. see: https://en.wikipedia.org/wiki/Moscow,_third_Rome), and this lasted until 1917. Both claims are dubious at best though (IMHO).


> Computers are much more difficult to bring back

In all post apocalyptic scenarios I can imagine, there will still be billions of already existing computers.

Rebuilding humanity will fulfill its computing needs by mining for a MacBook. Some will survive whatever natural disasters hit. As the supply of working ones dwindles, repairers will thrive. The skills and equipment necessary to repair waterlogged or crushed computers is still orders of magnitude less than that required to build a new one.

"Mining and repairing macbooks" as a strategy could last for 1000 years or so if necessary. Of the billions out there, at least a few will be working in 1000 years, if there are humans around with the desire to repair them.

As the repairs get harder and the supply dwindles, there will be more and more incentive to reinvent processes to make new ones. It wont be anywhere near as hard 2nd time round, because I'm sure across all of those billions of buried macbooks there are designs of all the machines necessary.


Unless the apocalypse is due to a solar flare similar to the Carrington event, which could destroy a lot of computers.


Rather than a total and permanent collapse, instead imagine a large-scale supply shock. Governments and institutions can access parts but consumers are rationed out of the supply chain for a while. Thailand's tsunami knocked out hard drive manufacturing back in 2011 [0], and crypto-tokens have persistently strained graphics card supplies for several years recently [1].

[0]: https://www.theguardian.com/technology/2011/oct/25/thailand-...

[1]: https://www.forbes.com/sites/erikkain/2021/01/06/bad-news-gr...


Building/refurbishing lead acid batteries wouldn't be all that difficult immediately post collapse.

Solar panels can also be refurbished. For poly silicon cells, it's mostly the contacts that degrade, and you can improve that with a soldering iron. A bit more work, but still doable, would be a forming gas anneal, which can additionally passivate defects.


I suspect you could scrounge materials for a nickel iron battery without too much trouble.


There are lots of ways to generate electricity. You can build a bicycle dynamo or a wood burning steam turbine pretty easily. People had wind turbines charging batteries a hundred years ago.


Singer foot pedal sewing machine style power for electronics would be one possibility. Of course those would have to be scavenged too given the lack of skilled blacksmiths these days.


A top male athlete can sustain 500W power for an hour. I wouldn't have high hopes for your sewing machine generator.


This doesn't really need to be hypothetical. There are plenty of contemporary and recent examples of countries experiencing collapse: Syria + Libya in the 2010s, Yemen now, etc.


microcontrollers are very low power because they are designed to run on batteries


CollapseOS provides the computation substrate, but is anyone thinking about the kind of computations post apocalyptic communities / individuals might be needing to run?

I imagine things like ledgers, basic simulations, farming yield prediction...

I couldn't find design documents besides the technical constraints it is willing to be compatible with.


We will need to calculate ballistics tables for our trebuchets in order to lay siege to neighboring towns.


Are there any works of fiction like this? A broadly competent medieval society that has scarce supplies of recovered present day technology.

There is no end to post apocalyptic fiction but usually it's set fairly soon after the collapse, or the technology uncovered is mysterious and futuristic.

Good chance I'm outing myself as an infrequent reader of science fiction asking this!


Check out /A Canticle for Leibowitz/. Each time CollapseOS comes up, I picture a team of monks cycling alternators around the clock to keep a system powered while another compares parchment to a vacuum-tube display.

I think for its chosen niche, the key qualities are (1) low power draw and (2) the ability to boot and be usable with no on-disk storage. The z80 system has that, but you could build something much more powerful. Burn the OS into a rom, have it copy into ram at boot.


It's not quite a medieval society, but The Windup Girl https://www.goodreads.com/cs/book/show/6597651-the-windup-gi... feature a post-climate change world where most energy comes from beasts of burden, but there are still a few rare computers and staggeringly expensive carbon-fuelled vehicles.


His Shipbreaker series is really good too. other authors: The world made by hand, https://www.goodreads.com/book/show/1689657.World_Made_by_Ha...

Eternity Road, https://www.goodreads.com/book/show/128533.Eternity_Road


Perhaps worth mentioning that Ship Breaker is YA fiction, and it shows IMHO.


The criminally underwatched "See" is one such fiction. It might be somewhat cheesy but I loved the worldbuilding.


https://en.wikipedia.org/wiki/A_Connecticut_Yankee_in_King_A... isn't quite that but is a fun take on the time traveller, by Mark Twain.


You might want to check out Eric Flint’s 1632 series. A small Virginia mining town is sent back in time and translocated to 1631 Germany, the series goes into how they try to build up a more technologically advanced civilization. S. M. Stirling‘s Nantucket series is somewhat similar. Nantucket is sent back in time to 1250 BC and they try to build a society there. David Weber’s Safehold series has a different premise, it’s set in the far future on a planet with medieval level technology. A lot of it is about the protagonists trying to build up their tech level.


I found Souls in the Great Machine by Sean McMullen to be an interesting take on this.


It's an interesting domain to think about. The big one would be re-establishing communication networks, being able to coordinate with groups across a large area is obviously the first thing any group with power (or looking to establish power) will do after ensuring food, water and shelter. It's pretty much a necessity if you want to maintain any power.

I could see embedded style sensors and technology being very practical, tools for measuring temperature/humidity etc and logging it constantly, sensors to measure water levels in rain tanks, basic inventory systems for food, basic automations for things like farming (automatic watering systems/collection systems even?)

Security is another one to think of: Sensors to detect people entering/leaving an area, trail cams for hunting, alarms etc.


Most of the machinery both industrial and in home are now computer controlled. Things like furnaces, CNC machining centers and lathes.


I'm looking around at all my home "machinery," and I don't see anything essential that is computer controlled. I have an ordinary gas furnace heating it; a plain old gas stove for cooking; a refrigerator that's got a plain analog temperature control; a boring, mechanical toilet based on technology that's hundreds of years old; and simple, mechanical valves that control the flow of water through my pipes.

The things I see that are computer controlled are luxury items, at best: TV, camera equipment, security camera, microwave, blender, laptops, tablet, NAS server, air purifier. This does leave out my car, but I don't expect cars are going to be playing a big part in any realistic post-apocalyptic future, anyway.

Most of that stuff could also be replaced by lower tech versions without much loss of functionality: analog TV, film camera, mechanical blender. I think even the microwave could be replaced by a lower tech version of itself without much loss of functionality. You could even have an actual, wireless remote for the TV.

I would miss my DSLR, laptops, tablet, and air purifiers, but I'd gladly give them all up if it would actually be a meaningful thing to do. As it is, getting rid of all the DSLRs and laptops in the world isn't going to save us. Very few individuals in this world have any real power to do anything about the most likely collapse scenario (climate change), and most of those people have a vested interest in not doing anything to help (American energy company CEOs).

Edit: a word


Your thermostat uses a microcontroller to run your furnace.

https://youtu.be/ZZC0SP02PqY


Nope. I’ve got one of the old, mechanical thermostats with an analog dial and a bimetallic strip that completes a circuit to run the heater. My apartment building is about 50 years old, and I’m pretty sure it’s still the original equipment. They’re so simple they just don’t wear out.


Actually, manual lathes and mills tend to keep their value. You can often find century old machines in reasonable condition.


You can find them in unreasonable condition in the middle of woods, etc... and those can be brought back as well.


Organizing information and things around in general will be a nightmare without the proper tools. Doing it using paper is hard enough, then count in that most likely paper will suddenly be considered a luxury for not being an easy replenished commodity. Any social structure over a certain size will greatly benefit from computing, even if only for things like keeping secured inventories and accounting. Data is expensive to handle outside the digital realm (think of making an electronic copy vs. making a copy using other media, probably by hand too). Although computing won't be for sure on the base of Maslow's pyramid for individual hierarchy of needs, it will be a great enabler for the social groups that will have managed to reach stability.


I think on of the main goals of CollapseOS is to keep the ability to program micro-controllers, which could be at least be useful for sensor networks and automation


I like the honesty on https://collapseos.org/civ.html but I get a very John Titor vibe from the author- To be honest if we have the collapse he envisions I think most people would be too worried about food to care about bootstrapping something on random hardware. However cyberpunk or fallout cool the idea is... still, was a fun read.


His expectations ("predictions"?) are predicated on three things: climate-change, peak-oil and "cultural bankruptcy".

I understand the most realistic scenarios for climate-change do indeed suck for everyone involved, but won't lead to mass human extinction - and the predictions of "water wars" were not grounded in realpolitk.

Of peak-oil being a harbinger of imminent societal collapse is immediately bunk: the world's economically leading countries and supranationals have already put in place plans for post-oil sustainability - such as fossil-fuel car sale bans within 10 years. Furthermore Covid has demonstrated that society carries on functioning when air-travel is cut to a fraction of its former volume - leaving only international shipping and road-haulage as the remaining oil consumers without a decent post-oil plan yet - but it's demonstrable that there's plenty of oil left to keep those services running for decades with current proven reserves - considerably more-so with the expected end of gas-powered cars within a couple of decades.

The author's definition of "cultural bankruptcy" seems to be "the US is exporting individualism to the world, which will mean an inevitable decline to non-cooperation and (insert your preferred game-theory metaphor here)". Whatever one may think of that, he doesn't establish a reasoned connection between the two - just hand-waving.

I think the author just has an anxiety disorder that needs better management.


Just looked up John Titor on Wikipedia:

John Titor is a name used on several bulletin boards during 2000 and 2001 by a poster claiming to be an American military time traveler from 2036. Titor made numerous vague and specific predictions regarding calamitous events in 2004 and beyond, including a nuclear war, none of which came true. Subsequent closer examination of Titor's assertions provoked widespread skepticism. Inconsistencies in his explanations, the uniform inaccuracy of his predictions, and a private investigator's findings all led to the general impression that the entire episode was an elaborate hoax.

Sounds more than a bit like QAnon, at least in design if not reach or effect.


The favorite part of this story for me was this:

> ... [various debunked predictions] led to the general impression that the entire episode was an elaborate hoax.

Whereas this

> by a poster claiming to be an American military time traveler from 2036.

Wasn’t the thing. Could be everyone was trolling but some of the commenters really seemed like true believers egged on by the trolls (kind of like flat earthers).

I just fundamentally don’t understand the level of gullibility involved that’s needed to buy into a story like that (like I get it exists but I just can’t understand the mentality that leads there). It’s also likely why astrology, religion, and wild unbelievable conspiracy theories take hold in our society (I like my conspiracy theories to have what I consider a more plausible down-to-earth premise).

I wonder what the evolutionary advantage there is for the duality in society. Is it really just as simple that the primitive “what if it could be true” is good for helping you avoid danger and the “that’s not a plausible line of thinking” helping better locate real world opportunities are the two engines for our success as a species? And that the misapplication is just a trait that’s not useful in a comparatively peaceful time (and the tension ironically brings about the reduction of peace time?).


It isn’t a matter of some people being gullible and others aren’t, everyone is gullible in this sense but a lot depends on prior experience, knowledge, peer groups and a lot of other factors. Trolls and conmen play a big part in building a cohesive narrative for fun and profit. People who are looking for a community gravitate to things like this and are easily influenced due to their lack of a counter balancing force in their lives (friends, community, work, faith, etc.). It becomes the thing they live for, part of the story of their life and a framework they use to understand the world. The scary thing is a lot of the world works this way and only varies in quality and pervasiveness. It’s the air we breath so it’s hard to get perspective on. How can you know for sure that something is true if it is beyond your means to verify? You have to appeal to authority or the majority. We do this with practically everything in our lives out of necessity, it’s impossible to vet everything. Most of it is valid and based on widely distributed independent verification. But there are messy areas where either this isn’t possible or where there is strong motivation to argue for a particular view. Politics and religion are pervaded by unverifiable claims and opinions. But it reaches into every aspect of life where there is uncertainty, money or an axe to grind.


HN thread also worth reading:

https://news.ycombinator.com/item?id=22558182


I was just wondering if people ever claim to be time travelers from the future, and if so what kind of reception they get.


I imagine that if an apocalypse occurs during our era, the internet era, our first main technological endeavor would be to build an internet out of scraps of hardware we would find sprinkled almost homogeneously all over earth (phones, old C64's etc) so that we could once again start organizing ourselves and reach out to our fellow human beings and in this reoccurring daydream of mine I always end up thinking long and hard about the choices I've made in my career that would essentially render me as useless as my mom when it comes to rebuilding the internet, since the only programming language I know really well is C#. Because we would have limited resources a post-apocalyptic internet would almost certainly not be built on something like dotnet.

My plan is to read the news very carefully and quickly throw myself at a C tutorial as soon as I feel an apocalypse coming on.


The most likely networked computers to survive will be banking computers, so time to brush up on your... <checks notes>...Fortran? Oh dear. Nevermind, I'll just be a raider.


With some luck you are going to find a prepper basement with a Raspberry Pi cluster.


There is a nice subplot in After the End - Survival¹ that deals with this. In it a well respected cardiologist finds himself useless post collapse, and is suffering from obvious minor medical problems.

¹ https://www.goodreads.com/book/show/25061009-after-the-end--...


Most programmers (myself included) are incapable of doing even the most basic things without googling them. We need an offline storage of StackExchange and it's subdomains.


In any plausible doomsday scenario, would scavenged Z80s (with compatible displays and input devices) really be more readily available than old Intel-based laptops, or even Raspberry Pis?

And if your post-collapse computer can’t read PDFs or play videos, how much useful preserved information can you really access?


> Can assemble Z80, AVR and 8086 binaries.

Quick Googling says that most modern x86 chips should be able to run 8086 binaries. Feel free to correct me if I'm wrong.

> And if your post-collapse computer can’t read PDFs or play videos, how much useful preserved information can you really access?

Most websites that don't use JS to load the important content. It's also worth noting that there doesn't seem to be any way to connect normal mass storage devices. It can read from floppies and SD cards. Some people still have floppies, but I would have to imagine there's a lot more useful information laying around on SATA drives.

It doesn't matter if it can read PDFs or play videos if you can't read the storage devices those are on.


Just Wikipedia would be huge.


“Just Wikipedia” exceeds the 16-bit address capacity of a z80 by a factor of 288,391 [0] even when compressed. Heck, you couldn’t even list all the article titles even if they were all an average of one character long.

Plenty of useful things you can still do with a z80, but at that tech level paper and ink is a better storage system than silicon and magnets.

[0] As per Wikipedia’s stats page, the size of current versions of all articles is 18.9 GB compressed


You don’t need the entire dataset to be addressable. Just a working set. How do you think we played 20GB video games on our Pentium 4s?


Much as I enjoy seeing old machines on the modern web [0], they’re still the wrong tool for the job and the wrong job for the tool.

We did useful stuff with computers even back in the 60s when transistors were the technobabble plot device for how Iron Man could fly [1]; we just used them to calculate solutions rather than as electronic libraries.

[0] https://www.c64-wiki.com/wiki/File:ContikiBreadbox64.png

[1] https://comicbooktheblog.blogspot.com/2010/02/iron-man-and-t...


I wasn't thinking a web browser, more like a WikiReader [0]. A tiny microcontroller, a text display and a flash memory card with an offline copy of the Wikipedia database.

[0] https://en.wikipedia.org/wiki/WikiReader


The microcontroller in that is significantly more powerful than a z80 ;)


I think Secure Scuttlebutt would be an awesome technology for a post-collapse world... I see how travellers arriving to a settlement could spread the updates of the network offline using the Gossip protocol, thus bringing news from far away.


On https://collapseos.org/why.html he says:

> I expect our global supply chain to collapse before we reach 2030

What's he going to say if we get to 2030 and his predicted collapse hasn't happened yet? "I was wrong" or "It's just taking a little longer than I expected"?


Doesn't really matter. The project's value doesn't require the collapse to happen by then or any specific time. In any case, the author found value for themselves in building their project even if the collapse never happens:

> But nevertheless, this idea seems too powerful to not try it. And even if it proves futile, it's a lot of fun to try.


I'm not saying the project is without value; I think if the author or others find educational/aesthetic/recreational/etc benefit from it, that's great.

It is like TempleOS. Not useful for its purported purpose of communicating with God, but nonetheless interesting and valuable as an example of a single person developing a programming language and OS, which even has a few novel ideas in it.


Except this one could serve its purported purpose, even if a collapse doesn't happen by 2030.

A collapse doesn't even need to happen ever. As I see it, its purpose is to be available as a safety net.


Probably the latter. Doesn't mean he will be wrong, though.

I really don't get people's optimism here. This is not some doomsday prophecy divined from religious texts. It's a trend in the data - if nothing significantly changes, we're headed for a crash in the coming decades.

It's like watching a program with a memory leak - you see it ticking up in its memory usage, knowing that when it approaches 100%, it will crash. Oh, it didn't, because Haber-Bosch^W^Wthere was a swap drive? Well, the swap space isn't infinite either.

Sure, maybe skyrocketing oil prices won't crash the economy (though people seem to forget that oil isn't just fuel and energy, but also anything made from plastics - i.e. just about everything). But something else will give (likely the habitability of this planet). We need to step off the exponent, or ride it to space. But we can't fit it on Earth much longer.


The thing about these types of events is that if everyone prepares against them they will never happen.


The latter, I'm sure. The historical event known as The Great Disappointment is a fascinating study in cognitive dissonance: https://en.wikipedia.org/wiki/Great_Disappointment


He seems strongly invested in the idea. The more you invest in an idea the more value you attach to it and hence the more resistant you become to evidence that contradicts it. so most likely it will be "It's just taking a little longer than I expected."


He already hedges his bets and makes pretty much the exact excuses you predicted.

I find this whole exercise to be utterly pointless, but hey, its his time, his money, and his effort, so more power to him.


It doesn't really matter what he says - his initiative, in my point of view, is like an insurance policy.

What do you say if you paid for health insurance, car insurance, but nothing at all happened to you over the course of the decade? Nothing. You count yourself lucky and keep paying.


This site pops up on here from time to time, and I can’t escape the fact that it’s driven entirely by the coolness factor of scavenging for parts and making them work.

And it is cool.

But the underlying arguments he makes about imminent collapse are baffling. Imminent collapse due to our oil supply drying up? That’s been an erroneous prediction made for like a century, because we are apparently really bad at predicting what the economics of oil extraction will be in the near future. As it turns out, finding and extracting oil becomes way cheaper, and different categories of extraction tech emerge. The lights stay on, and more people write about the NEXT energy collapse which is totally going to happen.

Also, am I missing something, or does he really not even consider nuclear power? It just seems like such an odd hyperbolic and ungrounded argument overall.


One of the links in the “why” article gets into more detail and specifically mentions nuclear power: https://dothemath.ucsd.edu/2011/10/the-energy-trap/

There’s also more on the whole phenomenon here: https://www.opendemocracy.net/en/oureconomy/deep-adaptation-...

And lots and lots more collapse literature out there if you care to dig into it.

On the one hand, predictions of doom have so far been beaten by technological advancements. On the other hand, we do live on a finite world. On the gripping hand, it’s quite disconcerting to think how rapidly our civilizations and individual lifestyles have changed compared to the vast sweep of human history and prehistory.


I love the project too, but even if we’re at peak oil (demand) there are still dozens of unexplored or barely touched shale basins around the world. Peak gas is a long way off. And oil & gas infrastructure doesn’t just “rot away” in a year or two due to a pandemic, or it would have already rotted due to the periodic busts that crash the rig count just as much.

It’s interesting how people’s reasons for the impending doom differ with their political leaning. Left leaning people will say we’re running out of resources and soon to be inundated with melted ice caps because we’re too individualistic. Right leaning people say we’re going to turn into Idiocracy because of poor breeding and easy living because there’s too much welfare and immigration.


It's tricky though, right, because these sorts of societal collapses do happen! We've seen them many times throughout history.

I think the pure numbers indicate that the next one probably won't happen in my lifetime (good), but I do think it will happen! What makes us think our society is special?


> What makes us think our society is special?

What is “our society”? A particular country or group of countries, or global society?

Certainly any country or even group of countries could go downhill. But, historically, usually when one country is in decline, somewhere else on the globe another is rising. For global society as a whole to collapse, across all countries and continents - that would be something that has never happened before. I don’t even know if we can put a probability on it - how do you determine the probability of something happening if it has never happened before?


Everything is very tightly connected now. Products and food are produced very far away and does an impressive bit of traveling during assembly. It's a supper efficient approach but not very robust. It is hard to get an idea of the knowledge hidden/locked in proprietary property but one can expect it to be truly advanced. It would take decades to replicate for other companies.


One of the things that has surprised me about the pandemic is how well these systems have held up. There are shortages of some goods, PPE in particular but on the whole our systems have continued working. We haven't been raiding our neighbours for food, if anything it's been the opposite.


In the grand scheme of history, the pandemic just isn't that bad. You can imagine a scenario—a much deadlier disease, or a war—in which we lost 5% of the world's population.


Collapse by 2030 is entirely possible under unsurprising circumstances. I rate it as near 50% likelihood, increasing rapidly after:

As climate disruption increases, crop failures in tropical and subtropical countries result in mass migration, first to cities and then toward wealthy temperate countries.

Temperate countries react by electing fascist governments to repel refugees. Fascist governments start wars (as they do), which spiral and disrupt international trade. Trade failure triggers economic crises, which exacerbate wars, ultimately leading to nuclear weapons exchange, and collapse.

Reason to believe: already starting, with no remedial action visible or planned.


Looking forward to coming back here in 10 years...


I hope you are right. But I don't see anything happening to inspire confidence.


If anyone would prefer to write LISP rather than FORTH, I wrote a similar project that would let you dust the ashes off any post-apocalyptic x86 machine and run John McCarthy's metacircular evaluator in nearly 512 bytes. https://github.com/jart/sectorlisp


Without any access to system resources, you wouldn't be able to bootstrap from a tiny Lisp kernel to a richer system. A sector-sized Lisp (when you eventually get it down to that size) is a neat project, but something of a closed system.

Part of the magic of a minimalist Forth environment is that they come with primitives for accessing raw memory, the stacks, and facilities for extending the language. You can build new parsing words, interface with hardware, and so on.


I don't fully get the reason for this project. Does it consider that you (or your community) will build a computer from scratch? In that case OK, I don't know enough but I guess an 8 bit simple computer makes sense, as long as it's something you can build using cables, LEDs and a breadboard.

If the idea is that you gather already manufactured parts, wouldn't it be better to just use some cheap SOC computers like the Raspberry Pi? They're low on power consumption, so powering them wouldn't be much of an issue, and it has greater, easier to use, software. It would also allow you to have things like a copy of useful Wikipedia articles, videos, photos, etc.

Maybe I'm misunderstanding the project goal though.


Love this sort of thing, though less for the post-collapse aspect and more for the dependency-free nature, which provides a certain amount of freedom.

I don’t like that seemingly all accessible bootstrapped projects still rely on non-bootstrappable silicon (like the 8086 here). Does anyone know of an effort to bootstrap the actual computational substrate (i.e. the silicon, or something which serves the same purpose but isn’t silicon)?


Could you explain why the 8086 is "non-bootstrappable silicon" please? Or what you mean by "bootstrap the actual computational substrate"? Thanks.


I think those are both referring to the fact that without some very sophisticated manufacturing processes and equipment, it's not possible to build an 8086 to begin with. Once we're at a point of societal collapse, we would, presumably, need to begin re-climbing the tech tree at a lower level and work ourselves up to 8086s again. So, we would need to start with computers we could actually build in order to get there.


This could be useful even outside of the global collapse that the creator fears. Imagine a sanctioned country or separatist region that was both completely cut off from the internet and also had all of their computers remotely bricked by remote management engine, malware, or EMP. Maybe the chips in their weapons and radios are shielded and working for now, but how would they bootstrap a new development environment?


>how would they bootstrap a new development environment?

The worst part of the collapse is going to be remembering everything in my vimrc file


Thinking about this is interesting if you go with premise the author uses. (to the extent I understand it. Which might be entirely wrong)

The farther we progress techologically the less can be salvaged in the aftermath of a collapse.

I have done a bit of manual soldering earlier in my life. This was simple stuff but I switched out memory chips. Switched out ROMS (only way to update some machines)

Primarily, zx81 (that you could get as a kit) zx spectrum and then what I consider my first real computer the Atari ST.

I have not performed any such surgery since my ST.

Looking at motherboards if various types we have now requires a lot more precision to “fix”

If you found 3 broken motherboards for the latest version of an I9 it would I think be difficult to cannibalize. Parts to get one working motherboard.

One these old computers it would be possible.

I presume that is one of the reasons the author talks about the z80. I am very familiar with the generations Sinclair used.

And there are genuinely massive amounts of them in various less sexy roles as controllers for electric appliances.

I wonder how far in the Intel PC evolution one could realistically salvage and make Frankenstein computers out of.

Storage would be a problem. Hard disks are complicated and prone to fail more so after a collapse I am sure.

Getting production of floppy drives going is not trivial either

Say the collapse hapoens in 20 years there will not be easy to find huge collections of 3.5 5 or 8 disks

Casette takes are also getting rare. Though some appear to think it retro and it is making a comeback now.

VHS tapes would be nice.


I like it. It's like having a small fire extinguisher in your kitchen. Hopefully you'll never need it. Most likely you wont. But if you ever do, you'll be glad it's there.

Are there any other projects out there like this?


Feels a bit like something of becoming. A new successor to ham radio for the tech and internet. Love the idea.


I'm not sure about the predictions, but I believe that devices have the right to exist, and that saying "your browser is not good enough" is akin to saying "your car is too old for this road" or "your wheelchair is incompatible with these stairs".

I test my website with every browser I can get my hands on. Yes, hundreds of them. Yes, even Netscape 2 and IE 3. I think Web has taken a bloaty turn, but we still have a chance to retain and preserve the knowledge.

The nice part about it is that for every browser I tweak for, the site gets more and more compatible and accessible. As someone with experience, I've been able to access it under the strangest circumstances and methods.


Makes me wonder what's the cheapest most efficient application class chip out there. Esp8266 ? I assume it's not the best but I'm not too aware of the MCU world.


It's always a multi dimensional spectrum of cost, performance, peripherals, development support, availability, family reach, etc. I personally really like STM8 microcontrollers for their simplicity and very low cost (can be less than 30 cents). There's actually another project that brings Forth to STM8: https://github.com/TG9541/stm8ef It has very good documentation and I recommend anyone to take a look


Today, probably the ESP32 is the most bang for the buck - unless there's something more cost competitive and equally widespread, which didn't come to my mind.

Notably, an ESP32 based on RISC-V is due shortly, with an expected price of $1. Wifi, BT5+BLE, 400KB ram (which may not be enough for everybody, but it's close), 22 GPIO's, a single 32-bit RISC-V core running at 160Mhz.

https://www.cnx-software.com/2020/12/20/esp32-c3-wifi-ble-ri...


No love for the Motorola 68k series?


Is there any hyperscaler that publicly disclose that they run centos?


Can't wait to use it.


[flagged]


It's not a product designed to support large numbers of web visitors, given its goals, your statement that it is trash shows that you might not understand what its creator was going for - it's quirky and weird, but it's very clear and adheres to the project’s intended goals.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: