Arctic World Archive Adds Latest Data Deposit in Svalbard Facility | Hacker News Reader

Arctic World Archive Adds Latest Data Deposit in Svalbard Facility

datacenterdynamics.com

77 pointsinfodocket34 comments

But is a archive not better, when it can be seen?

Imagine humanity is no more, but up there in orbit, a satellite similar in Size to Hubble contains all the data humanity ever produced. It observes the neighborhood, on the look out for likely orbits in the habitable zone, deduces planets. And when it has found a candidate, it beams a laser at the estimated position at arrival time, sending a cultural dump over the ocean of space and time. A lighthouse and a graveyard candle at the same time.

And if hardened against the eventuality of failure properly, somewhere, sometimes in the far future, some alien might look up at the night-sky and see a flickering, traveling disc of light, seemingly flying fast with the speed of planetary rotation. That would be a monument with purpose. Giving others all the knowledge we had, as a gift, to do better.

If you ask an electronics or computer guy to design you an archive that will last 2000 years their answer will be "Well, I can give it a try, but computers are only ~70 year old tech, so there aren't any proven thousand-year designs"

Stick with 2000 year old technology and we know what has a chance of lasting that long.

I heard about this facility previously due to GitHub's Arctic Code Vault[1], but this is the first time that I'm hearing that they're effectively using encoded microfilm[2][3] as their storage medium. Pretty cool stuff!

[1]: https://archiveprogram.github.com/arctic-vault/

[2]: https://en.wikipedia.org/wiki/Microform

[3]: https://www.piql.com/services/long-term-information-storage/

I love the idea of archiving everything we can, but does anyone else feel like the arctic archive was created by people who got wrapped up in too much science fiction? I feel like there’s a lot more practical short-to-mid term archiving of historical physical and electronic data that really needs to get done, that also has a much higher chance of being needed in the future. But perhaps I am just being short-sighted. I am also fully aware that one form of archiving does not mean other forms can not happen. But there does seem to be limited resources for this sort of activity.

I think somebody has to do it. It's like backup, we don't think we'll need it much until we need it.

Imagine if we as human had started preserving our knowledge since Ancient Greek (and, not burning the libraries..), collectively we could have been at a better place.

Or at least 1,000 years from know they will know there was once PPAP (https://www.youtube.com/watch?v=NfuiB52K7X8).

I think the main motivation for doing it in Svalbard is the seed vault, which benefits from the cold, dry environment in the permafrost. And sure, that the remote location lends its self to Foundation-like restore-civilization-after-the-dark-age scenarios, but it works fine for shorter-term back up too. If there's a blight that wipes out corn or something we can use the seed vault to replant. (Assuming we can cure the blight... cf the American Chestnut.)

The Syrian civil war was such a scenario where it was useful as a short-term backup: https://www.croptrust.org/in-the-news/syrian-war-causes-glob...

This thing ain't science fiction:

https://en.wikipedia.org/wiki/Carrington_Event

To go at it from a little different angle, perhaps there is benefit in doing something thats a little sci-fi on the basis that that alone might make it more exciting (and therefor actually get done!). There seems to be limited resources for it because it just isn't a high priority for people, so they don't donate their time or energy. So, maybe this addresses the cause of the fundamental restriction (desire).

"QR code" is a trademark of Denso Corporation, and refers to a specific technology. I'm unsure why news sources continue to misreport that the data is stored in "QR code". The correct term is "2D barcode".

Looking at their github, "QR code" is glaringly elided from their README. I'd guess they're very familiar with the standing QR code trademark. https://github.com/piql/boxing

Piql's 2D barcode technology is called a "boxing barcode". It is not a QR code. https://en.wikipedia.org/wiki/Boxing_barcode

I suspect QR code has become kleenexed. Technically a trademark for a specific brand and group of products, but everyone buying Puff's or their store brand still seems to say, "Pass the kleenex".

Did you Google any of that? :p

The terms 2d or boxing barcode are likely to be unfamiliar to most people, but everyone with a smartphone is likely to have heard of QR codes at some point, and thinks of them as such.

Branding and common vernacular are the enemy of precise language, so... good for Denso, I guess.

I've been on this rant for a while: https://twitter.com/Zamicol/status/1284613190215593984

https://old.reddit.com/r/github/comments/htoq8u/where_to_fin...

I've done a lot of work with QR Codes for work: https://github.com/davidshimjs/qrcodejs/issues/83

OMG is it time to look into tech jobs in Svalbard to take advantage of their unique visa nonrequirement and metropolitan frontier lifestyle???

I've been there couple times during wintertime and it's an absolutely amazing place to visit. Highly recommended.

Regarding your username, would it be rife with spooks e.g. your CIA counterparts —-a la Casablanca?

Maybe. Maybe not.

And if you like guns: It's the place administered by Norway with the by far most liberal weapons regulations (indeed you can be fined if you don't have adequate deterrents against polar bears outside of town, though it could be just a flare gun)

> However, future generations would need at least some level of technology - a camera and some compute capabilities.

It does kind of make you wonder - what is the data that is worth backing up in a really futureproof manner? If we keep our tech level, we’re probably not going to need to relearn how to make semiconductors. If we do not, we probably have bigger concerns than what a user with the nick rebuilder said on a nerdy forum in the year 2021.

> what is the data that is worth backing up in a really futureproof manner?

A futureproof encoding would have to be something like images at a scale visible to the naked eye engraved in a substance that cannot be corroded or destroyed. Stainless steel is probably fine for the material, but you run into limits on the amount of space (and steel) you're willing to devote to the project. And most information just can't be represented pictorially anyway, so the theoretical possibility of futureproof storage isn't even there.

I wonder if progressively shrinking the scale from human-visible to as small as we like would be a better way. After people see the obvious data, it would be clear that there's always more just out of sight and create the incentive to develop the technology to access it.

Stainless steel still corrodes, and even 316 can't even be used for many marine applications because of the salt water corroding it. Maybe fancier alloys are OK but for 1000s of years, it might pay to use something more history-tested.

I always thought Svalbard was a made up place in His Dark Materials by Philip Pullman.

> The same mine complex also features the Global Seed Vault which preserves some 400,000 plant varieties to maintain diversity in crops.

Isn't this the mine complex that is flooding due to melting permafrost?

well at least with data, I'm sure they have some durable format that will survive flooding

> Piql takes data and converts it into a QR code that is put onto 35mm film and stored in reels.

or not.

It seems insane putting this stuff in an icy hole in the ground on a warming planet.

Why not find some geologically stable mountain or plateau that has a slow erosion rate in a desert and build a bunker?

Cooling takes energy, which is the first thing to go in a doomsday scenario. But you get it very much for free 1300km beyond the Arctic Circle.

> Isn't this the mine complex that is flooding due to melting permafrost?

The entrance hall was flooded in 2017, the seeds were not endangered. The entrance has been upgraded since.

> It seems insane putting this stuff in an icy hole in the ground on a warming planet.

As it is customary, all the concerns you come up with in first 5 minutes after reading an article on the Internet were likely considered by people working on the project.

> As it is customary, all the concerns you come up with in first 5 minutes after reading an article on the Internet were likely considered by people working on the project.

This is beautiful.

They should put vaults in Antarctica though.

Much safer, and the mainland Antarctica will not likely melt.

Norway however only has territory in the Arctic (which is an island that is not going to melt as well), and there were not any other takers.

Norway claims territory in Antarctica too, but these claims are only recognized by a handful of states who also claim land in Antarctica...

The lack of recognition doesn't really preclude construction that can be justified as for scientific purposes, though.

Convenience is more important.

Norway has claim to a huge area of Antarctica. Queen Maud's land. Subject to the Antarctic Treaty and also recognized by UK, France Australia, etc.

The odds of the Arctic becoming ice-free, including continental ice, are really disturbing. See the reports on the probable collapse of the Greenland ice.

Whereas there are no such concern in Antarctica, only its ice shelves are threatened.

Antarctica's ice is continually sliding into the sea. Anything embedded in it will eventually get dumped on the sea-floor as it melts out the bottom of an ice sheet. We have already lost things that were left lying around there.

There are mountains in Antarctica, ice-free spots that can be the location of vaults.

The point of "ice not melting" is not to use ice, but to show you that the land will keep cool even in summer.

Apparently they were not considered if they had to do major retrofitting.

> In October 2016, the Seed Vault experienced an unusually large degree of water intrusion due to higher than average temperatures and heavy rainfall. While it is common for some water to seep into the Seed Vault's 100 m (328 ft) entrance tunnel during the warmer spring months, in this case the water encroached 15 m (49 ft) into the tunnel before freezing.[21] The Seed Vault was designed for water intrusion and as such the seeds were not at risk.[21] As a result, however, the Norwegian public works agency Statsbygg completed improvements to the tunnel in 2019 to prevent any such intrusion in the future, including waterproofing the tunnel walls, removing heat sources from the tunnel, and digging exterior drainage ditches.[22][23]

Your own quote says it was designed for water intrusion, and so the seeds were not at risk. Seems like they'd accounted for actual conditions well enough, but decided to take addition precautions because they could.

Reply on news.ycombinator.com