Disappearing repositories

I do not understand the cartoon Frogfall uploaded. Why not replace the "originals" with uploaded copies? As long as the digital copy has all the information the original had, and it is equally readable, it seems like a good idea to replace it.

Ed Storms sent me 10 or 15 heavy boxes of printed papers. Around 1,600 papers. (The exact number is hard to determine for various reasons.) I scanned them or downloaded clean copies. I sent all of the digital copies back to Ed, and I uploaded several to LENR-CANR.org. I then threw away the paper copies. Neither Ed nor I have any place to store that much paper. I don't see any point to storing it.

If I could wave a magic wand and convert all of the technical books in my house to digital copies, I would love to do that. I have sent out ~90 books to be scanned at https://1dollarscan.com/.

As late as the 1980s, librarians and computer experts often recommended keeping copies of important information on paper. Paper was considered more robust with greater longevity. Computer storage standards were changing rapidly and old copies were often rendered unreadable. Since then, much has changed. First, computer backups are now far more reliable. When you make three copies of a data set, you store one online, the likelihood of losing all three is very small. It is much more likely you will lose a paper copy to fire or water damage. Second, the Acrobat .pdf standard, and the .jpg graphic standard are well-established, because billions of files have been recorded in these formats. I predict they will last for hundreds of years. Possibly thousands. Computers will eventually become so intelligent they will never forget these standards, even if better standards emerge.

I hope that better standards do emerge because, as programmers say, Acrobat is the format where documents go to die.

Quote from Frogfall

Many archivists in the UK still feel pain from the experience of the 1986 BBC Domesday Project.

Interesting. My middle son is a laser disc collector. - hardware and software - I ma also sure he knows how to connect a player to a modern PC. . I can remember seeing (probably in the 70's) a shop window on London stacked with new boxed laser disc platers for £2 each.

Quote from Frogfall

There is certainly nothing wrong with scanning physical items to enable wider access – but to use it as an excuse to destroy the originals can be highly problemati

That depends on what kind of originals you mean. Gutenberg Bibles? Important first printing copies of 19th century novels? Of course they should be preserved! Handwritten papers and first drafts by famous scientists? It would be a crime to destroy them. However, if we are talking about run-of-the-mill physics journals or mass market 20th century books, textbooks and the like, I see no problem destroying the originals.

Quote from Frogfall

But although the cost of digital storage, with the required backups, has come down over the years, it doesn’t mean that the data is always going to be immune from future hardware, software, and format obsolescence.

The data can easily be migrated from one hardware platform to another. As long as the format is something standard such as Acrobat that will never be problem. As far into the future as we have computers, I am confident they will know how to read Acrobat files. They may automatically convert them into a better format. I hope they will soon do that!

Format obsolescence will not be a problem because there are billions of Acrobat documents. There is no way the human race is going to forget how to read them. This is quite different from the situation in the 1980s when they were databases such as the ones at the Census Bureau tapes in unique formats. (They were not all that hard to parse, it turned out.)

The cost of digital storage media for all of the data in the world will soon be zero ($0.00). It will all be migrated to DNA storage. You could fit every byte in the world into a 1 L container. You could leave it in a cool, dry place such as a cave, for 100,000 years and it would be 100% readable. The materials are so abundant that the cost of materials is zero, this being about as much DNA as there is in 20 people. The only cost would be for the gadget that writes the data, or reads it back, and the power to run them. Once it is in that format, DNA can reproduced far more quickly and in larger numbers of copies than any other media. Prof. Church at Harvard who is working on DNA storage copied his own textbook into DNA and then produced 70 billion copies. He said this is the most reprinted textbook in history.

Harvard geneticist stores 70 billion copies of his book in DNA

George Church is a professor of genetics at Harvard University’s Wyss Institute for Biologically Inspired Engineering, and also co-author of the book…

newatlas.com

You could make millions of copies of all of the data in the world and give one to every person on earth who wanted one.

There are other more near-term data storage technologies that are likely to be implemented sooner than DNA. They will be far more data dense and cheaper than today's storage. Here's an example:

Project Silica

Project Silica is developing the first-ever storage technology designed and built from the ground up for the cloud, using femtosecond lasers to store data.

www.microsoft.com

At present we using a tremendous amount of electricity to maintain on-line storage for things like cloud storage and Google. Also a lot of materials, in the various hard disks. This problem will go away in the future. People say that data storage requirements will continue to increase exponentially. I do not think so. Nothing is exponential, or the world would be knee-deep in black-and-white televisions, as my mother used to say. All goods and services have some upper limit of demand. It is true that today's storage is mind-boggling. There are estimates that the world now has more bytes of data than there are grains of sand on all of the beaches in the world. Yet that tremendous number would fit into 1 L of DNA.

The information density of DNA is astounding. The DNA in each of your cells weighs 6 picograms, it has 3 billion base pairs, and it is 2 m long. It has enough information to describe you and control your metabolism and most other bodily functions.

We will never, ever, ever forget how to read DNA, unless civilization collapses and reverts to pre-modern levels.

Quote from Frogfall

None of us know what is around the corner.

I know! I cannot tell you how things will be in 1000 years, but I am confident that for the next 100 years everything will be migratable.

Quote from Frogfall

Many archivists in the UK still feel pain from the experience of the 1986 BBC Domesday Project.

This article says they use media such as Laservision disks. These were soon unreadable because hardware changed. Nothing like that would happen today. They would store videos in formats such as MP4. There are 14 billion videos in this format at YouTube alone. There is no way people will forgot how to read them in 50 years, or 100 years. I would bet that computers will still be able to read them 10,000 years from now. Being able to read documents and view videos is essential to the survival of civilization. Knowledge of such things is seldom forgotten. Even in the so-called dark ages people knew how to make things like Roman roads and buildings. They didn't do it, but they knew how.

When I was young, for a while, I was so jealous of a friend of mine that had Superman The Movie on laserdisc.
I listened to the movie (after seeing it once) on a crystal radio, for weeks on end before bed when it played at a nearby drive-in theatre.

“Everything not saved will be lost”

- Nintendo