A Paean to Wikipedia
I’m doing blogmas — a blog a day, til Christmas — and I’ve already screwed the pooch and missed a day. Ah well, keep me motivated by subscribing to my (free) newsletter, Future Proof.
I’ve been thinking about Wikipedia a lot lately.
I remember, many years ago when I was at university, being part of a Marxist reading group (as all university students must be, at some point). We were asked by the convener to go around and name the invention or innovation which we each believed might represent the apotheosis of human accomplishment. Penicillin, air travel, eradication of smallpox, nuclear energy: these are the sort of things that people were saying. And then the conch arrived at one of our number, and he said, “Wikipedia.”
I hadn’t thought, for a minute, before then, about what Wikipedia means as a human accomplishment. When he said it, I think it raised titters from the group. After all, Wikipedia is distinctly unglamorous. On university campuses around the world it is considered akin to SparkNotes — a superficial, and often incorrect, skim through a subject matter. All the same, we were totally reliant upon it for parsing biographical and historical information. And what made the whole undertaking all the more unique was its not-for-profit nature and its community construction. Unlike penicillin or air travel or the A-bomb, Wikipedia is not the exclusive domain of a single great mind or an elite body. It is something by human civilization for human civilization. And therefore, under the criteria of the question, it works (I bumped into this same guy recently at a party and mentioned this memory to him — “I stand by that,” he told me).
But, of late, my thoughts about Wikipedia have been less about its conception but its survival. It has survived everything that the Internet Era has thrown at it. Founded in 2001 by Jimmy Wales and Larry Sanger, Wikipedia is now a venerable old man of the World Wide Web. When it launched, the biggest website in the world was Yahoo.com (the №2 was AOL.com and №3 was MSN.com). Remember that time? It has outlived the cultural currency of all of them. In fact, it has risen with titans like Google and Amazon; companies that have gone on to be leviathans of the modern business world. Now, in 2023, it ranks, in traffic, behind only Google, YouTube (a Google subsidiary) and Facebook — all three of which are amongst the most valuable companies in the world.
Listen below to my recent podcast interview with Alex Hollender, who led the 2023 Wikipedia re-design and talks about the challenges facing the organisation in the coming months…
Through all this, Wikipedia has maintained its independence and, therefore, its integrity. But most remarkable is how it has fared in the age of misinformation and disinformation, when there has been intense scrutiny on the dissemination of knowledge and counter-knowledge. “Facts” and “alternative facts”, as Kellyanne Conway might say. All of social media has been put through the wringer, subjected to the wild currents of political polarisation that have characterised the past decade. News sources, meanwhile, have been constantly bombared by accusations of bias and bad faith. The BBC, for example, has come the world’s largest online news source: but here in the UK it is routinely abused by the right for being excessively left-wing, and by the left for being inordinately right-wing. And that’s the BBC, a corporation that has neutrality hard coded into its constitution. Other big media organisations, from the “failing” New York Times and CNN on one side, to the Murdoch empire on the other, have been routinely lambasted for their role as mouthpieces, client journalists and propaganda machines. Nobody is happy with social media and nobody is happy with the media.
In a way, Wikipedia is a synthesis of social media and traditional media, which makes its unchallenged objectivity all the more remarkable. It combines the ability to edit and create (as with sites like Tumblr, Reddit or Substack) with being an old school knowledge receptacle. It balances a statement of neutrality with open access to create, collaborate and contaminate. In many ways, it should be more vulnerable to these torrents of partisan bunfighting than your average internet company, and yet… it has, at time of writing, survived these trends. Wikipedia has, remarkably, not pissed off the left or the right. It does not face serious or systemic accusations of infiltration or partiality. It has enraged nobody (other than Elon Musk, who doesn’t like Wales’ liberal politics or the company’s non-business business model) other than a few university tutors. It has survived, just like the library of Alexandria didn’t.
But, of course, it is under threat. As one of the world’s most exhaustive corpuses of knowledge and human-created writing, it is a natural resource for Large Language Models, which are used to train AI. And ChatGPT — the energy intensive chatbot launched by OpenAI — is already eating into its market. Why read through a lengthy Wikipedia entry on the Roman Catholic church, when you can just ask ChatGPT: “is the Pope a Catholic?”
Does it matter if ChatGPT, or some encyclopaedia specific AI-service, overtakes Wikipedia? Possibly not. Possibly, indeed, Wikipedia will work with an AI company in order to make just that: a question and answer service whose responses are drawn from Wikipedia’s grand archive, rather than the folly-strewn excesses of the entire World Wide Web. What is worrying about the use of AI-sourced knowledge is the lack of control. I have grown to trust Wikipedia’s army of editors and curators, its community of writers and researchers. They are all volunteers, all slightly obsessive, with a passion for knowledge, and for equalising and democratising access to it. If they are subject to internal algorithms, I believe them to be fundamentally benign. If you leave acquiring and parsing knowledge to a computer, it is harder to discern the preferences and prejudices at play there. Because, after all, these Artificial Intelligences are all built on human curated knowledge: they aren’t themselves going out to the Vatican and making notes on St Peter’s Basilica, crawling through the dusty catacombs, asking Francis, man to man, whether he’s a Catholic. They are simply acquiring this knowledge from their corporeal (for now) overlords.
Trust is, ultimately, the most important factor in the distribution of knowledge. The fact that a service run by thousands of imperfect people, the sum product of all these mismatched interests and intellects, has maintained that trust through the turbulence of the past few years — the drowning waves of distrust that have swept over the media at large — gives me enormous confidence in it as an undertaking. Whether it is humanity’s greatest accomplishment might depend on how it weathers this next chapter, as we start to delegate knowledge curation and acquisition to mysterious algorithms built by mysterious companies housed in mysterious servers, making money for mysterious investors. For now, I’d rather trust Wikipedia’s joyous, benevolent, generous nerds.
Follow me on Twitter, and subscribe to my newsletter, Future Proof.
