Internet Archive Fire Shows Vulnerability Of The World's Online Memory

The Internet Archive is a jewel of the digital world: 

On  Techdirt.

Why Piracy Is Indispensable For The Survival Of Our Culture

Last Year Techdirt wrote about the case of the huge collection of historic jazz recordings that had been acquired by the US National Jazz Museum. The central problem is that even if the recordings can be digitized before they deteriorate, very few people will hear them because of their complicated copyright status. 

On Techdirt.

How Can Your Content Live After You Die?

The current computer scene is notable for the role played by user-generated content (UGC): Facebook, Twitter, Flickr, YouTube etc. are all driven by people's urge to create and share.

Most of this is done by relatively young people; this means death is unlikely to be high on their list of preoccupations. Which also implies that they are probably not thinking about what will happen to all the content they create when they do die.

So we find ourselves in a situation where more and more content is being produced - not all of it great, by any means, by certainly characteristic of our time and important to the people that create it and their family, friends and users. Despite that rapid accumulation, no one is really trying to address the issue of what is going to happen to it all as users die.

This is quite separate from the more immediate problem of services shutting down, as is happening with Google Video. At least in these cases, you generally have the option to transfer it to some other site. But what happens when you - the creator, the uploader, the one that is nominally responsible for that content - are no longer around to do that?

You might hope that your heirs, whoever they might be, would carry on with things. But that presupposes that you leave all your passwords with them - in your will, perhaps? There are probably also issues to do with changing over the ownership of accounts - again, something that has not needed tackling much yet.

But is it really realistic to expect your family and friends to carry on caring for your content? After all, they will probably have their own to worry about. And what happens when they die? Will they then pass on not only their own UGC, but yours too? Won't that create a huge digital ball and chain that grows as it is passed on to the unlucky recipient? Hardly a recipe for sustainability.

Doubtless at some point some sharp entrepreneur will interpret this coming need as an opportunity. Just as you can pay a company to keep your cryogenically-preserved body against the day when a cure will be found for whatever ailment you eventually die of, so there will be companies offering digital immortality for your content.

The key question - as for those cryogenic preservation companies - is: will they really be around in hundreds of years' time? Of course, that's not really a problem for those sharp entrepreneurs that have your money *now*; and there's also not much you will be able to do about it if they don't make good on their side of the bargain...

What we need are repositories where content can be stored safely with a very particular audience in mind: posterity. To a certain extent, the Internet Archive already does that, but as I know from my own blog posts, its coverage is very patchy. And that's to be expected: a single organisation cannot hope to archive the entire Internet, including its second-by-second changes.

Moreover, depending on on one organisation is like putting all of the world's knowledge in the Library of Alexandria and nowhere else: after a good fire or two, you have lost everything. No, the solution is clearly to store the world's digital heritage in a distributed fashion.

We could start with national repositories, like the great deposit libraries that have a copy of every book published in their land. Those national Net holdings might also be national - after all, if every country did this, the world's output would be covered.

But clearly that's not a safe option either: ideally, you want multiple backups of national material to build in redundancy. You'd also want vertical markets to be stored by relevant organisations - every architectural site by some architectural body, every fishing site by some suitable organisation. You might have even more local stores of data in local libraries, or in local universities. Obviously the more the merrier (although it would be good to have some protocol so that they could all signal their existence and what they held to each other.)

Of course, none of this is going to happen, because the intellectual monopolists would be squawking their heads off about the inclusion of "their" content· This would have knock-on consequences for UGC, since, as we know, the boundaries between what is fair use and copyright infringement is ill-defined without hugely-expensive court cases. No organisation is going to take the risk of getting it wrong given the insanely litigious nature of the content companies.

And so we must sit back and contemplate not only the inevitability of our own demise - however far off that might be - but also the inevitable destruction of all that really ace content we have created and will create. Because, you know, maintaining that 18th-century intellectual monopoly is just so much more important than preserving the unparalleled global explosion of human creativity we are currently witnessing online.

Follow me @glynmoody on Twitter or identi.ca.

DVD Copying Software is "Illegal", but Copying?

In a decision that shows how ridiculously unclear the situation around copying DVDs is:

A federal judge ruled here late Tuesday that it was unlawful to traffic in goods to copy DVDs.

U.S. District Judge Marilyn Hall Patel’s ruling came in a decision in which she declared RealNetworks’ DVD copying software was illegal. She barred it from being distributed.

Patel said the RealDVD software violates the Digital Millennium Copyright Act of 1998 that prohibits the circumvention of encryption technology. DVDs are encrypted with what is known as the Content Scramble System, and DVD players must secure a license to play discs. RealDVD, she ruled, circumvents technology designed to prevent copying.

But the decision, although mixed, left open the door that copying DVD’s for personal use “may well be” lawful under the fair use doctrine of the Copyright Act, although trafficking in such goods was illegal.

Making backup copies etc. is so eminently reasonable that it needs spelling out by the courts. Paradoxically, not spelling it out is worse for the film industry, since the boundary of what is and is not (morally) acceptable are ill-defined, letting people make it up as they go along.

The Elephant in the Library

As I read about the incredible riches of content stored on the Internet, one thing worries me increasingly: who's doing the off-site backups? Too many of the current stores seem to have single points of failure, but nobody's really talking about this serious issue - call it the elephant in the library.

So it's good to hear of new projects that aim to back up content independently of others. Things like HathiTrust:

HathiTrust is a bold idea with big plans.

As a digital repository for the nation’s great research libraries, HathiTrust (pronounced hah-tee) brings together the immense collections of partner institutions.

HathiTrust was conceived as a collaboration of the thirteen universities of the Committee on Institutional Cooperation and the University of California system to establish a repository for these universities to archive and share their digitized collections. Partnership is open to all who share this grand vision.

HathiTrust is a solution.

To prospective partners, HathiTrust offers leadership and reliability.

It provides a no-worry, pain-free solution to archiving vast amounts of digital content. You can rely on the expertise of other librarians and information technologists who understand your needs and who will address the issues of servers, storage, migration, and long-term preservation.

Not all of this content will be freely available to all, although that will be the main emphasis - here's the current stats:

2,123,209 volumes
743,123,150 pages
79 terabytes
25 miles
1,725 tons
335,300 volumes (~16% of total)
in the public domain

Still, it's good to have backups for proprietary content too: if in the coming apocalypse it's lost because the primary stores go down permanently, there's no hope of ever opening it up.

And if you were wondering:

What does the name HathiTrust mean?

Hathi (pronounced hah-tee) is the Hindi word for elephant, an animal highly regarded for its memory, wisdom, and strength. Trust is a core value of research libraries and one of their greatest assets. In combination, the words convey the key benefits researchers can expect from a first-of-its-kind shared digital repository.