Say you wanted to host a personal page that can outlive you and be seen by the children of your grandchildren. Other than asking your progeny to keep paying the hosting bills, is there another way?
Get it printed on a physical monument made of metal/stone, put it on property you own, and engage a lawyer to set up the will and property covenants to require its ongoing existence and maintenance.
Our guesses about what the web looks like ten years in advance are likely to be wrong, let alone 500.
You'd have to also include plain English instructions on how to read the QR code, since the QR code format probably won't be around in the future either.
As for reading the instructions: they could probably still interpret modern English, similar to how professionals today can still interpret English from 500 years ago (with some effort).
Write a book and distribute it in high enough numbers. Things that seem too big to ever fall today might not exists in even 10 years, let alone 500, that's valid for every tech companies and even tech paradigms, you'd need people to actively migrate your site every XX years
and I went fishing a few months ago for images to print onto 8″x8″ squares and it's clear that older objects pick up damage over time. There are all kinds of beautiful objects from ancient Egypt but they are not made of paper. A print from the 15ᵗʰ century looks like
but paintings like that get a lot of attention in the form of cleaning and retouching.
Many of my favorite images come from paintings that are about 100 years old, such as the Futurists. These are old enough to have scans in the public domain but young enough that they haven't picked up the damage you see in older art that hasn't been heavily retouched.
Yeah of course, the og medium probably won't last 500 years, but if it's good enough to matter people will put them in safe spaces or reproduce them. Good quality modern books kept in ok storage conditions should last hundreds of years. I have a couple ~100 years old books and they look as new besides slight yellowing.
Prints are different since they're much more exposed and fragile than books
Anything that is on display is going to fade more quickly because of light than something that is kept closed.
When I was a kid in the 1980s I collected many mass market paperbacks (expected to be ephemeral) from as far back in the 1960s, even in the early 2000s I thought these held up pretty well, but circa 2020 I think many of them are getting pretty bad. (Contrast that to trade paperbacks that are sometimes "acid-free" but that frequently break in the first minutes of use because of incorrect and inconsistent construction.)
My house is humid and not a great place to store books, but I went looking in an academic library that follows "good" practices and found that mass market paperbacks from the 1980s and earlier were in bad shape too.
Look at what has lasted 500 years, and use those media. Websites are not among them.
1) stone
2) books if printed on the right kind of paper
3) metal if it's not something subject to rust
Even if it is possible to make a website that lasts 500 years, I expect we haven't figured out how yet. I'm sure it took people a while to figure out how to make long-lasting tombstones; early North American ones were often made out of sandstone or wood, and are illegible or completely gone now. You may be one of the very first people to give thought to how to make a website that will last that long; what are the odds you will get it right on the very first attempt?
It doesn't have to be fair. Things come and go. To host information for X years, statistically the things most likely to last are those that already existed for at least that amount of time.
Or you can take your chances, which, again statistically speaking, is not good if we're talking about websites lasting 500 years.
I recently saw headstones with photos on them. Now I’m thinking about writing a bunch of stuff and encoding it in a QR code on a headstone. The QR code is likely to be replaced with something else, but maybe if it says “QR Code” under it they could snap a picture and figure it out.
An alien civilization of giant lizard men that once roamed the earth whose dietary habits were superior and everyone should emulate (by buying and reading my book - "The Paleo-Plastic Diet for modern man").
Based on fairly common nerd interest in old tech (see: retro gaming, for one thing), I'm pretty sure QR codes will remain a curiosity for the foreseeable future, as they certainly solve a certain use-case/problem
Earth is a terrible place to store things for hundreds of years. I bet you could draw your website into Moon dust and come back hundreds of years later and it would still be there right next to your footprints.
That's a really tricky thought exercise. Legal trusts in most countries have time limits. Businesses come and go all the time. I think you are describing the same dilemma that companies offering freezing/preservation of your body or head upon death/near-death are dealing with. That is probably the business model I would study to find out what countries have the legal structure to support their requirements and thus your requirements. Even then they and their clients are accepting some risk. I suspect you will have to invest some capitol in this if you don't want your future generations to carry on this project on your behalf.
If your time requirements were shorter I would suggest a legal trust and set requirements for trustees to ensure the domain and site are preserved. I would also add instructions in the site itself to have family members preserve the site or even create new domains or methods of presentation. Each generation of family member could then create their own trust and repeat the process through inheritance. Your lineage could essentially leap-frog the system and compensate for businesses going bankrupt or technologies changing assuming they value the site and wish to add to it. I think your future generations would appreciate the ability to update the site. "Keys change, technologies are updated..." -- The Davinci Code
I was gonna suggest, instead of relying on your descendants/lawyers, “simply” avoid physical aging and death yourself, then you can continue to maintain the website.
I've been following the various teams working on halting or reversing aging. They are making great progress but sadly are not producing anything that can be commercially acquired yet. There are human trials on some of the methods being used on specific organs but nothing being tested on the entire body yet as far as I know. If you know of intravenous clinical trials resetting the epigenetic methylation body-wide, I would like to know. I believe they are still trying to find the balance that does not lead to tumors.
Become really famous (or infamous) so you don't have to host anything. Others will do it for you. Think about all the people you know who existed 500+ years ago. Did we just happen to find a book/stone where they wrote their story ? It was more about what they did that created history. So, create history and you will be hosted for ever. Think of names like Julius Caeser etc.
> Keep your information alive, secure and accessible for the future
> Long-term information storage - We can ensure your valuable information is archived appropriately for any length of time, with guaranteed accessibility.
Since there are many variations of "then don't use a website!" Answers here, I'll restate the puzzle.
Say you had to host the information as a set of static HTML documents and you wanted them to remain accessible for as long as possible; what strategy would give you the best odds?
Etch all the dependent technologies in stone so that somebody could replicate it.
Starting maybe from the transistor, if you're insistent on the full experience of loading the website from a browser with HTTP on TCP/IP networking.
Otherwise you can just etch the HTML in stone with a brief explanation of what the tags mean (eg. part of the HTML spec). Generally popular languages can survive 500+ years, so one can safely presume that some people 500+ years later can understand our English if they put in some effort.
.. I don't know why anyone would presume the web in any form would exist in 500 years. It would be fortunate if humanity as we know it still exists in 500 years. We are so capable of wiping out ourselves with various techs (whether intentionally or otherwise) that the odds are not really that great.
Seems like there are no good ways for individuals to keep digital records that long. Even if you found a storage medium that could last, the machine that reads and serves up that content would need replacing.
The Internet Archive would be the easiest way, but a boring answer. You're trusting another organization to maintain your data, but it's about as certain as any other option to be available in the future.
You could trust Cloudflare or AWS to keep an storage bucket alive, but then you've got to continue paying for it, and there will almost certainly be some changes in the next 500 years that would require a human touch. Who knows if AWS will still be around, or if its new parent company, Yahoo-Facebook-Intel, will drop contracts if you don't log on daily to 'Oculus Space'. My point is, The Internet Archive might be your best bet if you rely on another organization to keep the site up.
If you did it yourself, you'd need
- a stable file format (SIRF?)
- multiple backups and copies, even on the same disk
- very stable, simple computer to serve the content
- maybe like, a Raspberry Pi-like system submerged and sealed in mineral oil?
- static IP address to host from
- this might be the dealbreaker for a DIY solution. You'd need to rely on other services
- A local network that can be connected to...even after 500 years of changes to networking protocols.
Even with a DIY approach, you'd be relying on a trust of some sort that could handle replacing parts as needed. You could invent an analog titanium read-only storage disk with your data encoded on it, but even if you have bullet-proof hardware, you still need to allow people to connect to your server...and that is the least predictable part of the problem.
If enough people are interested in preserving content like this, a block-chain storage solution could work. You'd be relying on the system still being active in 500 years, so you'd need multiple generations of people all using the same blockchain to preserve their data. Kinda like a decentralized Internet Archive.
How do you know what the internet will look like in 500 years? Sites from 20 years ago are broken, you can expect that unless you’re using plain text and html that standards will change in 500 years and people will not be able to access your site.
Then there’s issues with domains. You’d have to setup a trust and again assume we will still be using domains in 500 years. If you use something like S3 then you’ll have to ensure they’re around for 500 years.
Even plain text is no guarantee. This assumes that future people will read/understand/use latin characters. ASCII could very well be replaced in the medium-term future
Sure, but that is backwards compatible / a superset of ASCII.
What I meant to say, is that it's totally possible that, in the coming few centuries, even basic ASCII won't be readily understood. As in the character mapping in modern systems will break down, i.e., int 97 is no longer 'a', but some glyph from a language not yet conceived.
We take for granted backwards comparability. Just because ASCII has been readable for the past 60 or so years, doesn't mean it will continue to be for the next 60.
Instead, we design systems to be such because it makes many things much simpler.
This is the reason why UTF-8 has basically "won" over UTF-16 or UCS-4 when it comes to encoding Unicode characters.
If anything, with the amount of data we have today, unless there is a big reason (probably political, but even they exist today for eg. China not to want to use an Unicode transformation based on American Standard Code for Information Interchange) to re-encode all historical data, backwards compatibility will be maintained with the computers of the future (if they still exist). Yes, even if we move their bytes to be 13-qubit qubytes :D
To elaborate on the cost: re-encoding all data from 2050 is probably not going to be too expensive in 2400, but by then you'll need to re-encode data from up to 2400. To me this seems like a reason that backwards compatibility will make sense to be kept because there is not much to be gained. Eg. UTF-8 approach has shown us the best way forward.
The trickiest is going to be to keep all video/audio encoding algorithms, especially as they are patent encumbered.
Shakespeare and the KJV are both closer to 400 years old. Shakespeare wanted to be accessible, though modern readers still need annotations to understand some of the words and most of the references. The KJV was intentionally somewhat archaic. There's a big difference in how accessible KJV is vs Shakespeare
But otherwise yeah, the Norman Conquest did a number on the language
Shakespeare wrote in what is referred to as "Early Modern English". It is no secret that his writings were a big influence in the evolution of the language itself. If you look at some of his contemporaries from the same period, however, their language will be very different and much harder to understand.
My specialty in undergrad was 15th and 16th century poetry. His contemporaries' writing (and writings from the early 15th c.) was no harder to read than Shakespeare.¹ The biggest challenge would be the irregularities of spelling—before the printing press and for at least a century afterwards, English spelling was inconsistent and flexible (and frequently was left up to the compositor for printed materials which led to different spellings for a word in the same document to make line breaks work better).
⸻⸻⸻
1. A notable exception would be Edmund Spenser who wrote in a style that was archaic even to his contemporaries.
It’s still intelligible, though. One thousand years is about the accepted timeframe for a language to be no longer mutually intelligible for speakers at either end of that period.
Of course, it’s entirely possible the rate of change within a language is not static over millennia.
...but the world wasn't a global village back then. A thousand years on, the world may look more similar than diverse, and less divergent / more convergent than it did over a similar time frame in the past?
It very well might. The extraordinary ways we’re able to preserve knowledge might slow the rate of change, or perhaps increasing interconnectedness between different cultures will accelerate the rate of change.
a large portion of English, in general, is changes in spelling from prior words, usually traceable to proto-indo-european, which is a catchall group of languages that etymologists are unable to find reliable sources for. It's generally extrapolated backward, and decent extrapolations are used to bootstrap understanding of words that don't quite "fit" with english, and may, in fact, come from other areas of the planet.
There are also scads of words that had a contemporary meaning that changed "overnight", morphing into entirely new meanings, which then brokered entirely new words with different definitions. My current favorite word to use an example of this is "filibuster" - the act of obstructing legislation by talking. The word, as so many in english, came from bastardizing the dutch word for "freebooter" or pirate, through a circuitous route of the French adding an S, and the American English removing an S. If you dig a bit more, you find that the "booty" part of freebooter (which means 'loves plunder' from the original dutch) came from a french word first recorded in the 1300s, "butin", which probably came from some mid-german word meaning "haul from plundering". There's also an implication that for a while in the 1500s-1800s freebooter was also the name of a private entity that engaged in exchanging goods - a "free trader", with the negative connotations falling in and out of style.
So, if you can parse Shakespeare or Chaucer at all, it's because of the mechanism of how English, and other languages derived from the same roots "evolve". Saxon and old High German, as well as Icelandic all play a huge role in the way we speak and write today, to name a few.
Not sure if you intentionally missunderstood, but the point was, that you probably would be able to make some sense of 1000 year old english, but that is about the border.
It depends on the culture of course. There are old cultures with the same references like a bible, that might cover longer timeframes of understanding.
Ah, my apologies. I could be clearer. It was curious for that comment to refer to _mutual intelligibility_ for people on either side of a 1000 year period.
It's a lot easier to ask whether contemporary humans can understand 1000 year old language, than to ask whether humans 1000 years ago can understand contemporary language.
"It's a lot easier to ask whether contemporary humans can understand 1000 year old language, than to ask whether humans 1000 years ago can understand contemporary language."
This is clear. And since we can only look backwards, we can only assume it works the other way around.
It took much less than 1,000 years for various pidgins and creoles to develop, and there are several of such languages that native speakers of the language's parent languages would have difficulty understanding.
This is interesting because your go to was an existing, but non-English language, rather than a completely different language that is an amalgamation of other, existing languages, gone through several generations of memes, in-jokes, meaning reversals, etc, to the point of being unintelligible from the former.
Indeed. As I commented separately I think the only practical solution is to use a simple medium (USB stick) and simple, standard file formats, then to copy from medium to medium, and to convert obsolete formats as time passes.
This requires descendants to keep at it over time, and it not really a "web site", but IMHO is the only way to keep the date accessible and useable over time.
Broken but not illegible. I feel like you're arguing that because the world may forget how to read some data 500 from now, storing the data is worthless. But that's just not true. Archeologists find meaning in writings they've never seen before all the time.
Sites which were carefully designed are not broken, and I think that is a good starting point when designing with longevity in mind.
Five hundred years is a long time, but I think it's reasonable to try to design a site that could last, for example, 25 years, because you can already write something which COULD HAVE worked for the PREVIOUS 25 years by testing with older browsers.
You'll want to restrict yourself to a subset of HTML which is supported by all of them, perhaps with some progressive enhancement.
Hm. I think your best bet is to make it easy for archive.org to archive.
archive.org will work on making your legacy technology work, as they are doing for flash, for example. Or they will find projects to make that work. That has a higher probability to work, opposed to finding a silver bullet now.
Though, the silver bullet there would be to minimize technological complexity. Make a simple static site with hugo, for example. That's easy to archive entirely.
Regarding archive.org, it’s worth to note that you can submit an URL for archival through their site. Here: http://web.archive.org/save
Another thing to note is that in the past they would retroactively apply robots.txt, such that if a previously archived URL was matched by a disallow directive in a later crawl of robots.txt, the page would be removed from public view. Fortunately they began reconsidering this behavior in 2017 though, and started not applying later robots.txt for some domains. Not sure about the current status of that though. Here’s a a blog post they wrote about it, from 2017: http://blog.archive.org/2017/04/17/robots-txt-meant-for-sear...
So hopefully crawled robots.txt etc will not prevent public access to archived pages in the future the way that it used to.
The main problem, aside from original owner putting directives in robots.txt that would cause archive.org to remove a page from public view, was that if a domain expired and someone else picked up the domain and made a robots.txt, then that one would be retroactively applied as well. And even if the new owner did not intend to remove anything from public view on archive.org, they could do so unintentionally simply by having a strict robots.txt and not being aware of what this would mean to archive.org when they crawled the domain again.
Another question though is, how are ancestors 500 years into the future going to know to look at the Internet Archive for the pages that OP made? And how will they know what URLs to look for? Though this same thing applies for most of the other solutions as well anyhow.
IPFS caches popular sites but forgets those that are unused. Sure link rot won't happen, but it's highly likely no-one will bother caching for 500 years. Of course, IPFS almost certainly won't be around then and what we're using today tech-wise will be as ancient to them as the loom is to us.
Probably the only way that has a chance of success is some kind of trust fund that will accumulate funds to keep the website in operation, including converting it several times to whatever's in fashion in 100, 200... 500 years.
Or, as some other post has already said, become so famous that people will record everything you've said. Although the first option only requires becoming moderately rich, which may be easier.
500 years is too long for modern tech. You will definitely need many copies of different types of storage, down to atomic-level records such as DNA. GitHub's Arctic Code or Amazon vault looks like the first Godzilla computers, but you're asking for a "500-years PC".
Perhaps there will be a technology that allows you to write site files in the DNA directly of your children and run a micro-DNA web server in the body, and your grandchildren can surf on your Internet data in the brain.
This DNA writing idea is premised on the concept that most of our DNA is "junk", which is an increasingly unlikely idea. Most of it does not code for proteins, that's true. We are slowly uncovering more and more functions for the rest.
Who said anything about modifying our existing genome? Perhaps in the future we’ll be able to add extra biologically inert genetic material solely for the purpose of permanently storing information, passed from generation to generation.
It doesn't need to be human DNA, indeed the part chosen may not 'make' it 500 years. While I have no desire to do this, choosing a variety of other life seems more viable, and more easily rediscovered if necessary.
probably write it as a virus that jumps from webserver to webserver, doing nothing malicious but adding your HTML to a pre-determined route
it will have to be a strong AI virus so it can keep rewriting itself as new software updates come along so you’re risking a skynet situation - oh! reminds me of the cowboy bebop episode where a hacked satellite entertains itself by drawing geoglyphs in the desert-definitely do that.
Encode the knowledge into the DNA/RNA of a highly contagious and non-deadly human virus and release it into the population. Hopefully, mutations won't overwrite your information.
Hosting isn't so much the issue. You need to create something worth keeping around for 500 years. If you succeed, people will make sure it's available somehow.
The thing is there can't be more than one internet, ever. The moment someone invents a new internet, someone will create a gateway to the old internet, and the internet will live on.
The internet can be highly regulated or the new one can be highly centralized and controlled.
The old internet could disappear as long as the economical incentives are enough to transition to a new centralized form of internet.
We are far from it, thanks, but don’t neglect that Facebook made a ton of websites become useless (to their owner at least) and disappear. Also, networks operators have made low cost plans limited to some of the internet. This is far from enough to kill the internet but a more restricted network is totally amongst the possibilities. And I’m not even talking about state control like in China.
This would be true if we only make incremental changes to communications over the next 500 years. I highly suspect we’ll see revolutionary changes that are fundamentally incompatible.
Assuming we can always hook up another gateway to the next “internet” is like someone 500 years ago assuming that they could send a message via horse to the internet.
That's a variation on the Ship of Odysseus. Years pass, nternets come and go, gateways open and close. Does "the Internet" still exist? (Rhetorical question.)
“The thing is there can't be more than one phone system, ever. The moment someone invents a new phone system, someone will create a gateway to the old phone system, and the phone system will live on.”
Thus far, this is true. The global phone system has persisted for almost 100 years, evolving via numerous “gateways” to older iterations of system (e.g. land lines -> cell phones, copper wires -> microwave links -> fiber optic lines, in-band signaling -> out-of-band signaling, individual lines -> multiplexing, etc.)
Yet despite the same phone system evolving and persisting for almost a century, dial-in BBSes, Minitel [0], and other outdated technologies that use phone lines are completely dead. Just because a communications medium may persist for a long time doesn't mean protocols utilizing the medium will.
The claim was not about the medium, but about the content ("a 500 year old seeded-torrent"). Phone system has never stored any data (other than the mapping between phone numbers and addresses or, lately, people), and was usually upgraded in a backwards-compatible way.
I think that's the important nuance: humans have always had a keen interest in keeping records of history. We are at an early age of electronic computers, but we've already got things like archive.org — that's likely to persist in some shape or form, just like we are actively trying to persist books and movies from different eras.
Other than natural or civilizational catastrophes, I only see the risk in the amount of data needing storage surpassing any one's entity ability to archive it, but I am sure "humans" would deal with that in due time too.
Edit: I do not necessarily believe the torrent claim, but wanted to clarify why I see a point in it.
Some (small) amount of content from those BBS systems has been archived elsewhere, but the BBS servers themselves are long dead and cannot be connected to.
By analogy, the contents of a few torrents might be archived somewhere 500 years from now, but nobody will be seeding the torrents themselves.
I'm emptying out a storage bin with 25+ year old stuff that begs to differ. Storing things has a cost, actually storing things so they last, far more so.
But they won’t need to delete anything. Imagine the storage and data science in 500 years. Probably the whole internet of 2021 will fit into a 2521 usb stick
My 600 GB of Photos begs to differ (I've worn out camera shutters!), you can't find anything without a long search. If you can't find it, you don't own it... so yes, in theory it's there, but nobody will access it, ever... so does it really exist any more?
It's back to the same meta that everyone else says... you have to make something that people want to have around in 500 years as a first step, or you're just pushing against very poor odds.
If you want it live-hosted for 500 years, you need people to host it. Some company. I don't know anyone that will even take your money and lie to you about that.
Your best shot is a simple set of files on a few redundant medium. If you want to get fancy, put a browser on there that will run (presumably emulated) with no network dependencies. Hopefully with a 64 bit time_t.
20x overspec'd solar panel. No battery. Broadcast a wifi hotspot at minimum radio power. Keep the components cold just above freezing and never allow them to freeze. You can accomplish this by using geothermal engineering to your benefit and burying the device with an appropriate heat exchanger so the planet can regulate the server temperature. Add in high availability and failover by using multiple buried devices connected over ethernet shielded with corrosion and abrasion resistant material at the lowest power and speed settings, then using raft algorithm and heartbeat. Seal everything with heaps of epoxy resin. Cut cable insulation off near where the cable enters the epoxy cube and reseal exposed wire with epoxy. This way water cannot wick into the main epoxy cube.
As much as I call cryptocurrencies a cancer, I admit storing on Bitcoin blockchain is probably the most resilient way, as its constantly being replicated across a huge amount of machines around the planet...
I don't know if the decay is linear or exponential with a long tail, but I assume solar panels don't fail outright unless they experience a short circuit or extreme temperature swings. Instead their output progressively declines. I bet a very large solar panel in the shade will last a very long time.
10-year-old solar panels are already deteriorating, and producing diminished output.
All these technological (digital) solutions are pissing in the wind; digital technology has lasted just 50 years so far, and people will discard it in an instant when something better comes along.
But think of all the websites! Yeah, OK. They're nearly all going to be gone in a decade or two.
Conserving bits has already proved to be difficult. How would you set about conserving an 8" floppy disk you found in your Dad's belongings when he died? Yeah, I know, you could probably source an antique 8" floppy drive. But 8" floppies only went obsolete in about 1985.
Internet archive? So you want to rely on a website to preserve your website? You have to be quite young, to be susceptible to the notion that the web will last.
As people up-thread have suggested, if you want your writings conserved, write something that the people of the future will spend effort to conserve. And forget about your descendants 500 years in the future; you have no reason to think your bloodline will survive. Wars have become progressively more destructive. And there's a slim chance that if your descendants do survive, they will know you're their ancestor anyway. I know about my ancestors back to 4 generations, beyond that I only know their names (and without supporting documentation).
People have spoken of using gravestones. Good luck with that. Gravestones as young as 50 years are being knocked down by councils because they are dangerous. And graveyards are full; they are being re-used, old monments removed and replaced with new ones. And have you ever tried to get detail off even a 200-year-old monument? Instead, since you're in the churchyard, go inside and read the parish register.
Love this question! Books are obviously gret. The Bitcoin blockchain looks promising as well. This is a great problem, looking forward to reading the comments. There must be some projects which tries to solve this.
Store it on the Bitcoin blockchain in the OP_RETURN field [1][2]. Note that this is considered an abuse of the system and discouraged, but of all the systems available now, Bitcoin is most likely to still be around in some form in 500 years IMO.
I work for a university that is over 800 years old, with a library and press (book and journal publishing) that are about 500 years old. The institutions have changed a lot over the centuries, especially the last 200 years.
Probably the best way to get people to preserve your work that long is to make it really outstanding, so future generations will want to preserve it even though technology and institutional governance keep changing.
put your page on a decentralized platform like Ethereum or IPFS and hope it gets maintained for 500 years. With Ethereum at least, there is a monetary incentive to keep the platform alive.
824 comments
[ 2.7 ms ] story [ 186 ms ] threadOur guesses about what the web looks like ten years in advance are likely to be wrong, let alone 500.
As for reading the instructions: they could probably still interpret modern English, similar to how professionals today can still interpret English from 500 years ago (with some effort).
The collection of the Metropolitan Museum has an API
https://metmuseum.github.io/
and I went fishing a few months ago for images to print onto 8″x8″ squares and it's clear that older objects pick up damage over time. There are all kinds of beautiful objects from ancient Egypt but they are not made of paper. A print from the 15ᵗʰ century looks like
https://www.metmuseum.org/art/collection/search/367024
It's frequently said that oil paintings hold up well over time such as
https://www.metmuseum.org/art/collection/search/816514
but paintings like that get a lot of attention in the form of cleaning and retouching.
Many of my favorite images come from paintings that are about 100 years old, such as the Futurists. These are old enough to have scans in the public domain but young enough that they haven't picked up the damage you see in older art that hasn't been heavily retouched.
Prints are different since they're much more exposed and fragile than books
When I was a kid in the 1980s I collected many mass market paperbacks (expected to be ephemeral) from as far back in the 1960s, even in the early 2000s I thought these held up pretty well, but circa 2020 I think many of them are getting pretty bad. (Contrast that to trade paperbacks that are sometimes "acid-free" but that frequently break in the first minutes of use because of incorrect and inconsistent construction.)
My house is humid and not a great place to store books, but I went looking in an academic library that follows "good" practices and found that mass market paperbacks from the 1980s and earlier were in bad shape too.
It says "ca. 1435–1491" - the 15th C.
Sounds interesting! Mind sharing links to some of your favourites?
1) stone 2) books if printed on the right kind of paper 3) metal if it's not something subject to rust
Even if it is possible to make a website that lasts 500 years, I expect we haven't figured out how yet. I'm sure it took people a while to figure out how to make long-lasting tombstones; early North American ones were often made out of sandstone or wood, and are illegible or completely gone now. You may be one of the very first people to give thought to how to make a website that will last that long; what are the odds you will get it right on the very first attempt?
Or you can take your chances, which, again statistically speaking, is not good if we're talking about websites lasting 500 years.
Somewhere a crackpot conspiracy theorist group will be trying to claim that the QR Codes contain messages from this long-lost civilisation…
If your time requirements were shorter I would suggest a legal trust and set requirements for trustees to ensure the domain and site are preserved. I would also add instructions in the site itself to have family members preserve the site or even create new domains or methods of presentation. Each generation of family member could then create their own trust and repeat the process through inheritance. Your lineage could essentially leap-frog the system and compensate for businesses going bankrupt or technologies changing assuming they value the site and wish to add to it. I think your future generations would appreciate the ability to update the site. "Keys change, technologies are updated..." -- The Davinci Code
Good starter resource: https://www.amazon.com/Abolition-Aging-forthcoming-extension...
"Anything really important is also worth doing anonymously."
https://www.piql.com/
> Keep your information alive, secure and accessible for the future
> Long-term information storage - We can ensure your valuable information is archived appropriately for any length of time, with guaranteed accessibility.
Say you had to host the information as a set of static HTML documents and you wanted them to remain accessible for as long as possible; what strategy would give you the best odds?
Starting maybe from the transistor, if you're insistent on the full experience of loading the website from a browser with HTTP on TCP/IP networking.
Otherwise you can just etch the HTML in stone with a brief explanation of what the tags mean (eg. part of the HTML spec). Generally popular languages can survive 500+ years, so one can safely presume that some people 500+ years later can understand our English if they put in some effort.
.. I don't know why anyone would presume the web in any form would exist in 500 years. It would be fortunate if humanity as we know it still exists in 500 years. We are so capable of wiping out ourselves with various techs (whether intentionally or otherwise) that the odds are not really that great.
The Internet Archive would be the easiest way, but a boring answer. You're trusting another organization to maintain your data, but it's about as certain as any other option to be available in the future.
You could trust Cloudflare or AWS to keep an storage bucket alive, but then you've got to continue paying for it, and there will almost certainly be some changes in the next 500 years that would require a human touch. Who knows if AWS will still be around, or if its new parent company, Yahoo-Facebook-Intel, will drop contracts if you don't log on daily to 'Oculus Space'. My point is, The Internet Archive might be your best bet if you rely on another organization to keep the site up.
If you did it yourself, you'd need - a stable file format (SIRF?) - multiple backups and copies, even on the same disk - very stable, simple computer to serve the content - maybe like, a Raspberry Pi-like system submerged and sealed in mineral oil? - static IP address to host from - this might be the dealbreaker for a DIY solution. You'd need to rely on other services - A local network that can be connected to...even after 500 years of changes to networking protocols.
Even with a DIY approach, you'd be relying on a trust of some sort that could handle replacing parts as needed. You could invent an analog titanium read-only storage disk with your data encoded on it, but even if you have bullet-proof hardware, you still need to allow people to connect to your server...and that is the least predictable part of the problem.
If enough people are interested in preserving content like this, a block-chain storage solution could work. You'd be relying on the system still being active in 500 years, so you'd need multiple generations of people all using the same blockchain to preserve their data. Kinda like a decentralized Internet Archive.
Then there’s issues with domains. You’d have to setup a trust and again assume we will still be using domains in 500 years. If you use something like S3 then you’ll have to ensure they’re around for 500 years.
My perspective, this is entirely unrealistic.
What I meant to say, is that it's totally possible that, in the coming few centuries, even basic ASCII won't be readily understood. As in the character mapping in modern systems will break down, i.e., int 97 is no longer 'a', but some glyph from a language not yet conceived.
We take for granted backwards comparability. Just because ASCII has been readable for the past 60 or so years, doesn't mean it will continue to be for the next 60.
Instead, we design systems to be such because it makes many things much simpler.
This is the reason why UTF-8 has basically "won" over UTF-16 or UCS-4 when it comes to encoding Unicode characters.
If anything, with the amount of data we have today, unless there is a big reason (probably political, but even they exist today for eg. China not to want to use an Unicode transformation based on American Standard Code for Information Interchange) to re-encode all historical data, backwards compatibility will be maintained with the computers of the future (if they still exist). Yes, even if we move their bytes to be 13-qubit qubytes :D
To elaborate on the cost: re-encoding all data from 2050 is probably not going to be too expensive in 2400, but by then you'll need to re-encode data from up to 2400. To me this seems like a reason that backwards compatibility will make sense to be kept because there is not much to be gained. Eg. UTF-8 approach has shown us the best way forward.
The trickiest is going to be to keep all video/audio encoding algorithms, especially as they are patent encumbered.
500 year old English is nothing like English today
Now if we go back a thousand years…
But otherwise yeah, the Norman Conquest did a number on the language
⸻⸻⸻
1. A notable exception would be Edmund Spenser who wrote in a style that was archaic even to his contemporaries.
Of course, it’s entirely possible the rate of change within a language is not static over millennia.
If it's not, I would love to know more about this.
There are also scads of words that had a contemporary meaning that changed "overnight", morphing into entirely new meanings, which then brokered entirely new words with different definitions. My current favorite word to use an example of this is "filibuster" - the act of obstructing legislation by talking. The word, as so many in english, came from bastardizing the dutch word for "freebooter" or pirate, through a circuitous route of the French adding an S, and the American English removing an S. If you dig a bit more, you find that the "booty" part of freebooter (which means 'loves plunder' from the original dutch) came from a french word first recorded in the 1300s, "butin", which probably came from some mid-german word meaning "haul from plundering". There's also an implication that for a while in the 1500s-1800s freebooter was also the name of a private entity that engaged in exchanging goods - a "free trader", with the negative connotations falling in and out of style.
So, if you can parse Shakespeare or Chaucer at all, it's because of the mechanism of how English, and other languages derived from the same roots "evolve". Saxon and old High German, as well as Icelandic all play a huge role in the way we speak and write today, to name a few.
It depends on the culture of course. There are old cultures with the same references like a bible, that might cover longer timeframes of understanding.
It's a lot easier to ask whether contemporary humans can understand 1000 year old language, than to ask whether humans 1000 years ago can understand contemporary language.
This is clear. And since we can only look backwards, we can only assume it works the other way around.
Yes, they’re long gone so it doesn’t really matter.
This requires descendants to keep at it over time, and it not really a "web site", but IMHO is the only way to keep the date accessible and useable over time.
Five hundred years is a long time, but I think it's reasonable to try to design a site that could last, for example, 25 years, because you can already write something which COULD HAVE worked for the PREVIOUS 25 years by testing with older browsers.
You'll want to restrict yourself to a subset of HTML which is supported by all of them, perhaps with some progressive enhancement.
archive.org will work on making your legacy technology work, as they are doing for flash, for example. Or they will find projects to make that work. That has a higher probability to work, opposed to finding a silver bullet now.
Though, the silver bullet there would be to minimize technological complexity. Make a simple static site with hugo, for example. That's easy to archive entirely.
Another thing to note is that in the past they would retroactively apply robots.txt, such that if a previously archived URL was matched by a disallow directive in a later crawl of robots.txt, the page would be removed from public view. Fortunately they began reconsidering this behavior in 2017 though, and started not applying later robots.txt for some domains. Not sure about the current status of that though. Here’s a a blog post they wrote about it, from 2017: http://blog.archive.org/2017/04/17/robots-txt-meant-for-sear...
Meanwhile, even Google does not interpret robots.txt the way they used to: https://developers.google.com/search/docs/advanced/robots/in...
So hopefully crawled robots.txt etc will not prevent public access to archived pages in the future the way that it used to.
The main problem, aside from original owner putting directives in robots.txt that would cause archive.org to remove a page from public view, was that if a domain expired and someone else picked up the domain and made a robots.txt, then that one would be retroactively applied as well. And even if the new owner did not intend to remove anything from public view on archive.org, they could do so unintentionally simply by having a strict robots.txt and not being aware of what this would mean to archive.org when they crawled the domain again.
Another question though is, how are ancestors 500 years into the future going to know to look at the Internet Archive for the pages that OP made? And how will they know what URLs to look for? Though this same thing applies for most of the other solutions as well anyhow.
Edit: Perhaps, of course, JavaScript :P
Or, as some other post has already said, become so famous that people will record everything you've said. Although the first option only requires becoming moderately rich, which may be easier.
Perhaps there will be a technology that allows you to write site files in the DNA directly of your children and run a micro-DNA web server in the body, and your grandchildren can surf on your Internet data in the brain.
> Perhaps there will be a technology that allows you to write site files in the DNA directly of your children
- Matter (paper, stones and eggs)
- Electromagnetic (light and drives)
- Quantum (a time-travel machine is must read the html code from the time-space by a Twitter-2500 hashtag #webarchive_2021_file_date)
- Promises of others (peoples, companies that can and can recreate the space) . ?
it will have to be a strong AI virus so it can keep rewriting itself as new software updates come along so you’re risking a skynet situation - oh! reminds me of the cowboy bebop episode where a hacked satellite entertains itself by drawing geoglyphs in the desert-definitely do that.
You would probably be best served by backing things up to tape and instilling a culture of copying these tapes every 15-20 years by your decendents :)
—someone in 1985, probably
The old internet could disappear as long as the economical incentives are enough to transition to a new centralized form of internet.
We are far from it, thanks, but don’t neglect that Facebook made a ton of websites become useless (to their owner at least) and disappear. Also, networks operators have made low cost plans limited to some of the internet. This is far from enough to kill the internet but a more restricted network is totally amongst the possibilities. And I’m not even talking about state control like in China.
Assuming we can always hook up another gateway to the next “internet” is like someone 500 years ago assuming that they could send a message via horse to the internet.
Thus far, this is true. The global phone system has persisted for almost 100 years, evolving via numerous “gateways” to older iterations of system (e.g. land lines -> cell phones, copper wires -> microwave links -> fiber optic lines, in-band signaling -> out-of-band signaling, individual lines -> multiplexing, etc.)
Yet despite the same phone system evolving and persisting for almost a century, dial-in BBSes, Minitel [0], and other outdated technologies that use phone lines are completely dead. Just because a communications medium may persist for a long time doesn't mean protocols utilizing the medium will.
[0] https://en.wikipedia.org/wiki/Minitel
I think that's the important nuance: humans have always had a keen interest in keeping records of history. We are at an early age of electronic computers, but we've already got things like archive.org — that's likely to persist in some shape or form, just like we are actively trying to persist books and movies from different eras.
Other than natural or civilizational catastrophes, I only see the risk in the amount of data needing storage surpassing any one's entity ability to archive it, but I am sure "humans" would deal with that in due time too.
Edit: I do not necessarily believe the torrent claim, but wanted to clarify why I see a point in it.
By analogy, the contents of a few torrents might be archived somewhere 500 years from now, but nobody will be seeding the torrents themselves.
While I agree they are interested in everything, I think they are more interested in sampling insignificant works, as opposed to archiving all of them
Even historians curate and don't keep everything.
It's back to the same meta that everyone else says... you have to make something that people want to have around in 500 years as a first step, or you're just pushing against very poor odds.
Your best shot is a simple set of files on a few redundant medium. If you want to get fancy, put a browser on there that will run (presumably emulated) with no network dependencies. Hopefully with a 64 bit time_t.
https://en.wikipedia.org/wiki/Bathtub_curve
As much as I call cryptocurrencies a cancer, I admit storing on Bitcoin blockchain is probably the most resilient way, as its constantly being replicated across a huge amount of machines around the planet...
All these technological (digital) solutions are pissing in the wind; digital technology has lasted just 50 years so far, and people will discard it in an instant when something better comes along.
But think of all the websites! Yeah, OK. They're nearly all going to be gone in a decade or two.
Conserving bits has already proved to be difficult. How would you set about conserving an 8" floppy disk you found in your Dad's belongings when he died? Yeah, I know, you could probably source an antique 8" floppy drive. But 8" floppies only went obsolete in about 1985.
Internet archive? So you want to rely on a website to preserve your website? You have to be quite young, to be susceptible to the notion that the web will last.
As people up-thread have suggested, if you want your writings conserved, write something that the people of the future will spend effort to conserve. And forget about your descendants 500 years in the future; you have no reason to think your bloodline will survive. Wars have become progressively more destructive. And there's a slim chance that if your descendants do survive, they will know you're their ancestor anyway. I know about my ancestors back to 4 generations, beyond that I only know their names (and without supporting documentation).
People have spoken of using gravestones. Good luck with that. Gravestones as young as 50 years are being knocked down by councils because they are dangerous. And graveyards are full; they are being re-used, old monments removed and replaced with new ones. And have you ever tried to get detail off even a 200-year-old monument? Instead, since you're in the churchyard, go inside and read the parish register.
[1] https://bitcoin.stackexchange.com/questions/39347/how-to-sto...
[2] http://www.righto.com/2014/02/ascii-bernanke-wikileaks-photo...
Probably the best way to get people to preserve your work that long is to make it really outstanding, so future generations will want to preserve it even though technology and institutional governance keep changing.