Cath and i were having a heated discussion this morning: one of our friends reported that out of 5,700 episodes of PlaySchool (a favourite childhood TV programme) only 1,900 are known to survive. Her comment was that nobody would ever watch them again and it was a waste of space.
My response was that they would, one day, form an invaluable archive giving us insight into things as diverse as gender equality in the 1960’s, developments in technology and even the evolution of the spoken language. Much of this won’t be apparent in our lifetime, but it prompted me into a broader thought.
The default position of history has always been transience: books, scrolls, buildings, bodies, all decay. But today, in the digital realm, the default position of history is permanence.
The default writer of history has always been the solitary historian. In the future, it will be semantic engines and layers of meta analysis that can cope with 1,900 or 1.9 million episodes, chewing through them to discern the meaning, creating digital histories that we consume but don’t create.
If history is always subjective, this will be the ultimate objectivity: written by nobody. Sure, it will simply be an averaging out and weighting of all the sources, but it will be increasingly hard to determine what’s ‘fact’ from ‘interpretation‘.
As computers develop the ability to recognise not just inputs but ‘things‘, as they learn to discern what makes a ‘chair’ a ‘chair’ and to spot racism in recorded voices, or sarcasm, or lies, they will learn to create histories, stories that are broader than one person could ever hope to write. I’d struggle to even watch 1,900 episodes, let alone analyse them for balance in gender, or tracking new words (and colour) entering the picture.
Clay Shirky’s ‘cognitive surplus‘ gives us one model where distributed networks may manage this type of work on a human level, but technology will inevitably triumph. Even with all our shared interests and surpluses, we simply don’t have the capacity to compete.
Chris shared an article with me about ‘decision making‘ software, used by Banks and other businesses to help determine the dispassionate ‘right‘ thing to do. The article indicated that things were not always clear cut, that the software only worked in some cases, but that wasn’t the point: the point is that it’s real right now. We are already enhancing our capability with digitally created narratives of what’s right and what’s wrong. We are adopting the technology and adapting accordingly. Just as we did with Wikipedia: first we fretted that it would dumb us down, but now it’s the first place we turn (and, again, Shirky’s cognitive surplus has seen to it that it’s about as reliable as Encyclopaedia Britannica for hard science facts).
In an earlier life, i trained as an archaeologist: the challenge, to sift through the strata, to analyse patterns and determine the best place to dig a trench because, with limited resources, you had to make some lucky guesses where to deploy your effort.
The challenge for digital historians is not scarcity of resource: it’s abundance. It’s beyond our comprehension to make sense of it, even a tiny part of it. Material is being created, shared, reviewed and reformatted far faster than we can track. If i read two books a month and live for another forty years, that’s 960 books (let’s call it 300 pages each, that’s 288,000 pages of text (if i’m optimistic…). It’s a drop in the ocean. I doubt i could even read all the books that philosophise about what history actually is, let alone all the actual history books.
Technology has created shifts in knowledge before: quills allows us to write, capturing thoughts on expensive vellum, then Gutenberg nailed the word to the page in volume and allowed us to replicate those words rapidly. But these were as nothing to the age of Digital History and the emergence of digital historians.
Just because we’re not quite there yet, don’t dismiss it: within out lives we will see a ‘definitive‘ history of the First World War created by a computer, parsed down from all the primary and secondary sources, every one of them weighted for accuracy against the political bias of the author and the number of copies it sold on Amazon (and probably a convoluted correlation to the number of times the Author Tweeted about it).
And it will last forever.
Our understanding of permanence is poor: we are used to decay, the slow erosion of the rocks that built stonehenge (and the total loss of meaning around that structure as the entire civilisation that built it had no written words). We are less used to permanence: worrying about how we memorialise Facebook pages, how we delete decade old blog posts or try to hide our credit history online. ‘Nothing lasts forever‘, we used to cheerfully say, except that now it does. Or it may do anyway.
Those 1,900 episodes that survive? I bet you that no more will be lost: now they’re here to stay.