A year-long series looking back on the most significant moments of the past 25 years, how they changed our world, and how they will continue to shape the next 25.
Simon Garfield is the author of more than 20 books of non-fiction, including All the Knowledge in the World: The Extraordinary History of the Encyclopedia. His latest, a history of the pen, will be published next year.
The more unstable our times, the more we look for certainties. Here’s one: Every living person who merits an entry on Wikipedia is unhappy with what’s written about them. It’s not the facts, necessarily, but the blandness of it all, the way everyone appears to have lived their life within a template. “That’s what my life amounts to? That’s how I’ll be remembered? But they didn’t get my hilarious side, or my love of striped tropical fish.”
In my case, I don’t like my photo – a full-length, side-on, wholly weird shot of me giving a talk I don’t remember about the Beatles. I have no idea why they chose it, nor why they didn’t select a much more recognizable or handsome one. But I’m powerless to change it, for vanity carries little weight with the Wikipedians in charge.
Screenshot of Simon Garfield's Wikipedia photo.
A few years ago I interviewed Jimmy Wales, who told me that he dearly wished his entry said something about his ravenous passion for cooking, but he couldn’t do anything about that either, despite the fact that he was one of Wikipedia’s two founders, and still sits on the board of the Wikimedia Foundation, the charitable body that hosts the encyclopedia and directs its policy.
My inability to change my photo, and the absence of Mr. Wales’s culinary prowess, are two of the things that make Wikipedia great. As the encyclopedia approaches its 25th anniversary, it may be a good time to acknowledge a few others, beginning with the obvious.
Wikipedia, which launched Jan. 15, 2001, has remained true to its original intention, the establishment of a volunteer-edited, free, live encyclopedia, a resource able to respond immediately and predominantly accurately to changing events. Exceptionally for such a popular resource, Wikipedia does not track you or sell any of your search information. It does not carry advertisements or monetize itself beyond regular appeals to users for small donations. It is fully accountable, with every keystroke credited and dated to a specific user. It is continually trying to improve its accuracy, reach, diversity of content and contributors. And beyond all this, it is a thing relentlessly and reliably useful.
Being both intellectually rigorous and shamelessly trivial, it reflects the world as it sees itself. Its anniversary should be a cause for celebration, an overdue confirmation, I think (due perhaps even from its many early critics), that it has become one of the greatest things online, a rare representative of the internet for good. It is also, I suggest, one of the greatest inventions of our modern age.
Oh, and it’s incredibly entertaining, despite the many rigorous guidelines that define its dull hyperlinked appearance. If you are just 1-per-cent curious about life, you may never emerge from its rabbit holes. I am fond of the Instagram account Depths of Wikipedia, which dredges up obscure gems from the margins. Who would not enjoy the 20-second audio clip of a Panasonic microwave heating a cup of tea (a bit of clattering, then humming, then three beeps at the end)? Or the entry for Osama Vinladen, a Peruvian soccer player who has a brother named Sadam Husein and a sister named Georgia Bush? Or the list of wrong anthem incidents, including the time at the 2012 Shooting Grand Prix in Kuwait when the correct anthem of Kazakhstan was replaced with the parody anthem O Kazakhstan from the Borat movie? And who, finally, can forget Number 16, the spider who lived for an estimated 43 years? As his (her?) entry notes, Number 16 effortlessly outlived a tarantula who died when she (he?) was 28.
Why am I so enthusiastic about such things? Because I am enthusiastic about all entries in all encyclopedias, so keen indeed that I’ve written a book about their history. Like almost everything else, encyclopedias began in ancient Greece, a gathering of mythology and travellers’ tales.
Older readers may recall the golden period in the middle of the last century when many rival printed editions were sold door-to-door on piecemeal payment plans: Grolier, Funk & Wagnalls, World Book, Collier’s, Chambers. Many more will remember Microsoft’s Encarta from 1993, available on a format once considered the last word in technology, the CD-ROM.
I’ve been using encyclopedias for roughly 50 years, since I copied out large sections of them for my school homework (I was aware that teachers might recognize Britannica, so I relied on a lesser publication called the Everyman Encyclopedia; it fooled my history teacher right up to that point when he revealed that he had the exact same set).
In my career as a writer, I continued to refer to print encyclopedias long after most things became quicker to find online. I trusted the old information more, which I always imagined was written primarily by tweedy men smoking pipes (it often was). I liked the formality of the entries, even if many of them had the air of fusty Pathé newsreels. And I continued to refer to Britannica in the early days of Wikipedia, that period where anyone could write an entry titled “My cat Missy is Queen of the Universe!” and it wouldn’t be taken down for months.
A set of the Encyclopaedia Britannica at the New York Public Library in 2012.Angel Franco/The New York Times
The inspiration for my book came from one of those annoying banner appeals that appear on Wikipedia occasionally, asking for a paltry amount to keep things going. I saw one of these and did something extraordinary – I contributed. At that moment I could think of no better cause. As a writer, I use Wikipedia many times a day, and wonder how I ever operated without it. In addition to its usefulness, it is the most reliable time suck I know. I am particularly partial to the entry which contains, among a great deal other things, lists of fictional presidents of the United States in books and movies, lists of Scooby-Doo episodes, and lists of Danish football transfers 2008-09. As the first paragraph explains, “this list of lists of lists is a list of articles that are lists of other list articles.”
These days, old encyclopedias are incredibly difficult to dispose of; charity shops are reluctant to take them, and they are almost impossible to bequeath, as young people cannot see the point of them, even if they know what they are. Their interest to us is anthropological, not least their mistakes and errors of judgment. They are a repository of what we knew when, and how we regarded ourselves. The homophobia and racism that exists in the early editions of Britannica is stomach-turning, as is its begrudging support of Hitler in the 1930s. This history-in-amber is not an affliction that troubles Wikipedia, which is committed to correcting known errors as soon as they are discovered. (Its staff admit that there are many thousands of undiscovered ones.)
First Person: I had to say goodbye to my old Encyclopedia Britannica set
And of course Wikipedia is now the only serious player in the game. Britannica still exists online, but according to the analytics company Semrush, in April, 2025, it attracted just over 60 million visitors, compared with Wikipedia’s 4.55 billion. For this reason alone, Wikipedia carries immense responsibility alongside its influence. It does not carry a monopoly of information, but it is the dominant supplier at a time when factual reliability is increasingly subjective and seductively corruptible. If we don’t trust Wikipedia, where else do we go?
To better understand its scope, we need to go a little further into its weeds. At the beginning of May, Wikipedia contained 6,992,651 individual articles. New articles appear at an average rate of 500 per day. On average, it receives two edits and 4,000 page views every second. That’s the English Wikipedia. There are also 355 editions in other languages, including Cebuano, Waray, Yiddish and Chuvash, which collectively host 64,908,921 articles. The source of these figures? Wikipedia, of course, and there is no feasible way of double-checking (one could count each one, I suppose, but by the time you were finished, the stats would have long multiplied.)
English Wikipedia has been edited by about 15 million users, and although the majority of these have just amended a few small things, others appear to have forgone sleep to patrol its pages. People calling themselves Ser Amantio di Nicolao and Tom.Reding, for example, have respectively made 6,517,829 and 4,341,692 edits, some of them as small as a comma. Tom.Reding has qualifications in astronomy and mechanical engineering; he describes his editing specialisms as “researching primary sources, adding missing information, fixing errors, making things look pleasing, ordered, and complete,” among other things.
Wikipedia founder Jimmy Wales speaks during “Wikimania 2012” – an international Wikimedia conference in Washington, D.C.MANDEL NGAN/Getty Images
It is both obvious and extraordinary that this enormous project began with just one entry, then a few, and then a large information-dump from the out-of-copyright 29-volume 11th edition of Britannica published in 1910/11 (30 pages on Glass, 47 on Greece). Barely regulated at its start, mischief was rife. On Jan. 17, 2001, just two days after its launch, a didactic dog-lover made the following judgment for the entry entitled Standard Poodle: “A dog by which all others are measured.”
On Dec. 20, 2001, not quite a year since launch, and still evidently an obscure enterprise, the following message appeared on its landing page: “Welcome to Wikipedia, a collaborative project to produce a complete encyclopedia from scratch. We started in January 2001 and already have over 19,000 articles. We want to make over 100,000, so let’s get to work – anyone can edit any article – copyedit, expand an article, write a little, write a lot.”
Seven years later, the writer Nicholson Baker observed in the New York Review of Books that the enterprise was still like “a giant community leaf-raking project” in which everyone could participate, but there was a wide range of expertise. “Some brought very fancy professional metal rakes, or even back-mounted leaf-blowing systems, and some were just kids thrashing away with the sides of their feet or stuffing handfuls in the pockets of their sweatshirts, but all the leaves they brought to the pile were appreciated.” The pile grew, Mr. Nicolson noted, “and spilled out all over, and everyone jumped up and down in it having a wonderful time.”
The pile grew quite fast. By March, 2006, the English-language Wikipedia amounted to one million articles. Eighteen months later, the total was double that. The five million milestone was reached by November, 2015, and by January, 2020, there were six million.
Editing rules have tightened up alongside this expansion. It is no longer possible for a novice to change entries about famous people or articles judged politically sensitive. Gone, alas, are the days when it was possible to write an entry called “1933 in Video Gaming” or “Albert Pooholes” (yes, both genuine and “live” for a while.)
The different editing responsibilities of Wikipedia’s custodians follow a strict hierarchy. As of this writing, there are 841 administrators (who are able to block user accounts and edit fully protected pages), 6,916 rollbackers (able to revert the last edit of a page with a single click) and 43 oversighters (able to immediately suppress defamatory material or copyright violations). There are also editors blessed with the title autopatrolled. If you are autopatrolled (and 3,633 people are), your fine reputation attained by your previous contributions have granted you a free pass for your future entries, eliminating the need for initial peer review.
I am registered as a user, which grants me the most basic editing privileges. I can make an edit on an unprotected article, but unless I adhere to all the established guidelines it is likely to be reviewed and amended. This even – or especially – applies to any changes I might make to the article titled “Simon Garfield,” despite my advanced knowledge of the subject. As the rules state, even changing my photograph might be considered “unduly self-serving.”
Further, as the Wikimedia Foundation’s former chief executive Katherine Maher told me in 2020, its philosophy is not quite what one might expect. “I don’t think Wikipedia represents truth. I think it represents what we know or can agree on at any point in time.” Ms. Maher ran the organization for five years, and is now CEO of NPR. “What Wikipedia offers is context,” she believed, claiming it is less an encyclopedia than “an information ecosystem.” At no time was this better demonstrated than during the pandemic, a period in which Wikipedia became the global landing ground for real-time experience and expertise, and demonstrated its strength and purpose like never before. The mischievous “vandalism” that had been present in its first years was in most cases swiftly corrected.
Wikipedia’s most misunderstood rule seems counterintuitive. Unlike every encyclopedia that preceded it in print, Wikipedia prides itself on containing “no original material,” relying instead on information that has been published before by reputable sources and can be cross-referenced (hence the frequent appearance of the phrase “citation needed” within its pages). According to its foundational principles, “Wikipedia summarizes accepted knowledge. As a rule, the more accepted knowledge it contains, the better.”
Occasionally this rule gets it into bother, most famously in 2018, when anyone seeking a biographical sketch of the Canadian scientist Donna Strickland immediately after she had won the Nobel Prize in Physics found no mention of her at all. An entry had been prepared, but was deemed insufficiently referenced. Her fame lay largely hidden behind the firewalls of scientific journals and had yet to penetrate the broader media. It led to two embarrassing criticisms: Wikipedia tended to favour the trivial over the serious, and its content greatly under-represented women (its biographical entries skewed 82 per cent male). In the past 10 years, worldwide “editathons” and a project named Women in Red have attempted to improve this, the gains gradual but small: women’s representation now stands at about 20 per cent.
The list of senior staff at Wikimedia Foundation offers a healthier mix, with more than half of its senior posts occupied by women. There are about 700 staff in all, their salaries accounting for the majority of the annual US$170-million operating costs. Its staff have been known to favour a single confounding phrase: “Thank God our enterprise works in practice, because it could never work in theory.”
Wikipedia’s greatest asset of all? The fact that it’s a towering work in progress. “Perfection is not required,” one of its many instructional pages explains. “Collaborative editing means that incomplete or poorly written first drafts can evolve over time into excellent articles. Even poor articles, if they can be improved, are welcome.” This inclusiveness achieves two overlapping things: it reinforces the value of esoteric topics that wouldn’t have made it into Britannica, and it encourages people who aren’t professors to write about them. In its print adverts of the 1970s, Britannica was particularly proud of the number of its contributors with PhDs, and today the lack of academic “experts” is one of things that infuriates traditionalists. But it is this feature that confirms Wikipedia’s excellence: a huge pool of talent continues to create something brilliant in an era of democratic retreat.
There is, of course, a cloud on the horizon. Even the most primitive form of AI appropriates Wikipedia for its knowledge base, and outpaces it in response to any inquiry. When I asked Google “What is the Globe and Mail?”, an “AI Overview” augmented information from Wikipedia with the newspaper’s own “About Us” section and The Trust Project consortium dedicated to accountability in news reporting. Wikipedia’s stand-alone usefulness may thus be waning, a threat it is attempting to counteract by using AI for its own ends, not least for simple editing tasks and the elimination of rogue bots. As recently as 2021, Jimmy Wales could claim on NPR that, “At Wikipedia, there is no algorithm – humans wrote it all.” Alas, this is no longer the case.
I asked Claude, one of the newer AI systems, whether it was condemning Wikipedia to obscurity. It replied that it didn’t believe AI would completely “take over” Wikipedia, especially given the “strong governance structures” that values human involvement. Users would continue to use Claude for short answers and Wikipedia for more detailed and intricately sourced information.
“Let’s hope so, Claude,” I typed.
“I share that hope,” my generative knowledge assistant replied. “Wikipedia is genuinely one of the most remarkable achievements of the internet age – a massive, collaborative effort to make human knowledge freely accessible to everyone. It would be a real loss if that were to disappear.”