Historians 100 years hence will have an abundance of source material about how ordinary lives were lived during the 21st century thanks to the unprecedented way we leave traces through websites, email, Twitter and social networks such as Facebook.
Well, that's the theory. In practice, most of this living history will be discarded in digital dustbins unless something is done about it. We are often told that, thanks to startling improvements in technology, all our personal memories will soon be able to be stored on something the size of a sugar cube. But the granules that make up that sugar cube are widely scattered and difficult if not impossible to recover.
It is reckoned that the average life expectancy of a website is less than 75 days and that at least 10% of UK websites are lost or replaced with new material every six months. These figures come from a statement by the British Library at yesterday's launch of the UK Web Archive, which will guarantee access in perpetuity to thousands of hand-picked UK websites – some of which might otherwise have faced oblivion.
They include Antony Gormley's Trafalgar Square Fourth Plinth Project. This involved 2,400 participants, and the live stream by Sky Arts would no longer have existed online from next month had the BL not taken over responsibility for it. Other projects to be preserved for posterity include a record of the Credit Crunch and the 2010 general election.
The BL is doing a marvellous job of preserving key historical events, but what it covers is only a tiny part – about 6,000 sites so far – of the nation's digital memory. Even doing that has proved hugely time-consuming because the BL's small staff has to seek permission every time it takes a copy of anything. This is because of the UK's archaic copyright laws, which will hopefully be partially corrected in the digital bill now going through parliament. Fewer than 25% of the bodies approached by the BL for permissions even bothered to reply.
The issue of copyright is a global nightmare for anyone interested in digital preservation. The problems that Google has encountered in its – utterly praiseworthy – quest to digitise the world's books are nothing compared to the problems of preserving documentary films where the multiple permissions needed for each one from commercial interests will, as Lawrence Lessig brilliantly describes in the New Republic, lead to a situation where " the vast majority of documentary films from the 20th century will be forever buried in a lawyer's thicket inaccessible (legally) because of a set of permissions built into these films at their creation".
Even if these legal problems could be solved – which they won't, because the influence of corporate lobbyists on copyright law is forever tightening – there are still others. Digital files degrade much faster than paper files and have to be upgraded, sometimes as frequently as every 10 years. No one has yet found the digital equivalent of the Rosetta Stone, still intact after well over 2,000 years. You only have to peer into your own digital history to see what a digital black hole looks like. Practically everything I did with my BBC B, Sinclair Spectrum and even Psion computers is lost forever in a jungle of changed formats, obsolete floppy disks and losses from data that was not backed up. But that is nothing to what is happening now. Young people may, or more likely may not, worry that things they say on social networks such as Facebook or Twitter may come back to haunt them but the bigger worry is that in the longer term it won't be there at all as digital dynasties rise and fall. Outpourings on Twitter provide an amazing record of people are doing and even thinking, but they are already history before the end of the day. Does anyone seriously think they will still be there a century hence?
There are some estimable bodies preserving archives that include snapshots of the whole web, such as the wonderful not-for-profit Archive.org which will require funding in perpetuity to match its archival ambitions but it admits that no comprehensive archives of television or radio programs exist. Its sister organisation is the Wayback Machine where you can see, for instance, what the AltaVista search engine looked like in 1996 before Google came over the horizon. But as for someone in the distant future trying to recapture our photo albums scattered among sites such as Flickr.com or Picasa or wherever, long after the subscriptions have run out or the companies sold on, forget it.
It is sometimes argued that if copyright law is standing in the way of a universal archive then maybe the world's collective memories should be placed into some kind of escrow account, not to be opened until copyrights have been sorted out or expired. This sounds plausible, but it would act against the worthy principles espoused by the British Library and others that as much as is humanly possible should not just be available but available now.