When Mathew Modine’s character first shows up in Christopher Nolan’s Oppenheimer, I figured the rest of the cinema audience wouldn’t have appreciated me shouting out “VANNEVAR BUSH IN THE HOUSE!” so I screamed it on the inside.
The Manhattan Project was not his only claim to fame or infamy. When it comes to the world we now live in, Bush’s idea of the memex has been almost equally influential. His article As We May Think became a touchstone for Douglas Engelbart and later Tim Berners-Lee.
…the device he describes does not resemble the internet or anything I’ve ever found on it.
Then he says:
What Bush was describing sounds to me like what you might get if you turned a browser history — the most neglected piece of the software — into a robust and fully featured machine of its own. It would help you map the path you charted through a web of knowledge, refine those maps, order them, and share them
Yes! This!! I 100% agree with the description of browser history as “the most neglected piece of the software.” While I wouldn’t go as far as Chris when he says web browsers kind of suck, I’m kind of amazed that there hasn’t been more innovation and competition in this space.
If anything we’ve outsourced the management of our browsing history to services like Delicious and Pinboard, or to tools like Obsidian and Roam Research. Heck, the links section of my website is my attempt to manage and annotate my own associative trails.
Imagine if that were baked right into a web browser. Then imagine how beautiful such a rich source of data might look.
I woke up today to a very annoying new bug in Firefox. The browser shits the bed in an unpredictable fashion when rounding up single pixel line widths in SVG. That’s quite a problem on The Session where all the sheet music is rendered in SVG. Those thin lines in sheet music are kind of important.
Browser bugs like these are very frustrating. There’s nothing you can do from your side other than filing a bug. The locus of control is very much with the developers of the browser.
Still, the occasional regression in a browser is a price I’m willing to pay for a plurality of rendering engines. Call me old-fashioned but I still value the ecological impact of browser diversity.
That said, I understand the argument for converging on a single rendering engine. I don’t agree with it but I understand it. It’s like this…
Back in the bad old days of the original browser wars, the browser companies just made shit up. That made life a misery for web developers. The Web Standards Project knocked some heads together. Netscape and Microsoft would agree to support standards.
So that’s where the bar was set: browsers agreed to work to the same standards, but competed by having different rendering engines.
There’s an argument to be made for raising that bar: browsers agree to work to the same standards, and have the same shared rendering engine, but compete by innovating in all other areas—the browser chrome, personalisation, privacy, and so on.
Like I said, I understand the argument. I just don’t agree with it.
One reason for zeroing in a single rendering engine is that it’s just too damned hard to create or maintain an entirely different rendering engine now that web standards are incredibly powerful and complex. Only a very large company with very deep pockets can hope to be a rendering engine player. Google. Apple. Heck, even Microsoft threw in the towel and abandoned their rendering engine in favour of Blink and V8.
The ECMAScript, HTML, and CSS specifications today are (for the most part) stellar technical documents whose algorithms can be implemented with considerably less effort and guesswork than in the past.
I’ll be watching that project with interest. Not because I plan to use the brower. I’d just like to see some evidence against the complexity argument.
Meanwhile most other browser projects are building on the raised bar of a shared browser engine. Blisk, Brave, and Arc all use Chromium under the hood.
Arc is the most interesting one. Built by the wonderfully named Browser Company of New York, it’s attempting to inject some fresh thinking into everything outside of the rendering engine.
Experiments like Arc feel like they could have more in common with tools-for-thought software like Obsidian and Roam Research. Those tools build knowledge graphs of connected nodes. A kind of hypertext of ideas. But we’ve already got hypertext tools we use every day: web browsers. It’s just that they don’t do much with the accumulated knowledge of our web browsing. Our browsing history is a boring reverse chronological list instead of a cool-looking knowledge graph or timeline.
For inspiration we can go all the way back to Vannevar Bush’s genuinely seminal 1945 article, As We May Think. Bush imagined device, the Memex, was a direct inspiration on Douglas Engelbart, Ted Nelson, and Tim Berners-Lee.
The article describes a kind of hypertext machine that worked with microfilm. Thanks to Tim Berners-Lee’s World Wide Web, we now have a global digital hypertext system that we access every day through our browsers.
But the article also described the idea of “associative trails”:
Wholly new forms of encyclopedias will appear, ready made with a mesh of associative trails running through them, ready to be dropped into the memex and there amplified.
Our browsing histories are a kind of associative trail. They’re as unique as fingerprints. Even if everyone in the world started on the same URL, our browsing histories would quickly diverge.
Bush imagined that these associative trails could be shared:
The lawyer has at his touch the associated opinions and decisions of his whole experience, and of the experience of friends and authorities.
Heck, making a useful browsing history could be a real skill:
There is a new profession of trail blazers, those who find delight in the task of establishing useful trails through the enormous mass of the common record.
Taking something personal and making it public isn’t a new idea. It was what drove the wave of web 2.0 startups. Before Flickr, your photos were private. Before Delicous, your bookmarks were private. Before Last.fm, what music you were listening to was private.
I’m not saying that we should all make our browsing histories public. That would be a security nightmare. But I am saying there’s a lot of untapped potential in our browsing histories.
Let’s say we keep our browsing histories private, but make better use of them.
From what I’ve seen of large language model tools, the people getting most use of out of them are training them on a specific corpus. Like, “take this book and then answer my questions about the characters and plot” or “take this codebase and then answer my questions about the code.” If you treat these chatbots as calculators for words they can be useful for some tasks.
Large language model tools are getting smaller and more portable. It’s not hard to imagine one getting bundled into a web browser. It feeds on your browsing history. The bigger your browsing history, the more useful it can be.
When it comes to retrieving information from this online memex of mine, I use tags. I’ve got search forms on my site, but usually I’ll go to the address bar in my browser instead and think “now, what would past me have tagged that with…” as I type adactio.com/tags/... (or, if I want to be more specific, adactio.com/links/tags/... or adactio.com/journal/tags/...).
It’s very satisfying to use my website as a back-up brain like this. I can get stuff out of my head and squirreled away, but still have it available for quick recall when I want it. It’s especially satisfying when I’m talking to someone else and something they say reminds me of something relevant, and I can go “Oh, let me send you this link…” as I retrieve the tagged item in question.
But I don’t think about other people when I’m adding something to my website. My audience is myself.
I know there’s lots of advice out there about considering your audience when you write, but when it comes to my personal site, I’d find that crippling. It would be one more admonishment from the inner critic whispering “no one’s interested in that”, “you have nothing new to add to this topic”, and “you’re not quailified to write about this.” If I’m writing for myself, then it’s easier to have fewer inhibitions. By treating everything as a scrappy note-to-self, I can avoid agonising about quality control …although I still spend far too long trying to come up with titles for posts.
I’ve noticed—and other bloggers have corroborated this—there’s no correlation whatsover between the amount of time you put into something and how much it’s going to resonate with people. You might spend days putting together a thoroughly-researched article only to have it met with tumbleweeds when you finally publish it. Or you might bash something out late at night after a few beers only to find it on the front page of various aggregators the next morning.
If someone else gets some value from a quick blog post that I dash off here, that’s always a pleasant surprise. It’s a bonus. But it’s not my reason for writing. My website is primarily a tool and a library for myself. It just happens to also be public.
I’m pretty sure that nobody but me uses the tags I add to my links and blog posts, and that’s fine with me. It’s very much a folksonomy.
Likewise, there’s a feature I added to my blog posts recently that is probably only of interest to me. Under each blog post, there’s a heading saying “Previously on this day” followed by links to any blog posts published on the same date in previous years. I find it absolutely fascinating to spelunk down those hyperlink potholes, but I’m sure for anyone else it’s about as interesting as a slideshow of holiday photos.
Matt took this further by adding an “on this day” URL to his site. What a great idea! I’ve now done the same here:
In 1990, the science fiction writer Douglas Adams produced a “fantasy documentary” for the BBC called Hyperland. It’s a magnificent paleo-futuristic artifact, rich in sideways predictions about the technologies of tomorrow.
I remember coming across a repeating loop of this documentary playing in a dusty corner of a Smithsonian museum in Washington DC. Douglas Adams wasn’t credited but I recognised his voice.
Hyperland aired on the BBC a full year before the World Wide Web. It is a prophecy waylaid in time: the technology it predicts is not the Web. It’s what William Gibson might call a “stub,” evidence of a dead node in the timeline, a three-point turn where history took a pause and backed out before heading elsewhere.
Here, Claire L. Evans uses Adams’s documentary as an opening to dive into the history of hypertext starting with Bush’s Memex, Nelson’s Xanadu and Engelbart’s oNLine System. But then she describes some lesser-known hypertext systems…
In 1985, the students at Brown who encountered Intermedia had never seen anything like it before in their lives. The system laid a world of information at their fingertips, saved them hours at the library, and helped them work through tangles of thought.
What I love about the web is that it’s a hypertext. (Though in recent years it has mostly been used as a janky app delivery platform.)
I am very much enjoying Matt’s thoughts on linking, quoting, transclusion, and associative trails.
My blog is my laboratory workbench where I go through the ideas and paragraphs I’ve picked up along my way, and I twist them and turn them and I see if they fit together. I do that by narrating my way between them. And if they do fit, I try to add another piece, and then another. Writing a post is a process of experimental construction.
And then I follow the trail, and see where it takes me.
On the 50th anniversary of Vannevar Bush’s As We May Think, Tim Berners-Lee delivered this address in 1995.
To a large part we have MEMEXes on our desks today. We have not yet seen the wide scale deployment of easy human interfaces for editing hypertext and making links. (I find this constantly frustrating, but always assume will be cured by cheap commercial products within the year.)
Craig writes about reading and publishing, from the memex and the dynabook to the Kindle, the iPhone, and the iPad, all the way back around to plain ol’ email and good old-fashioned physical books.
We were looking for the Future Book in the wrong place. It’s not the form, necessarily, that needed to evolve—I think we can agree that, in an age of infinite distraction, one of the strongest assets of a “book” as a book is its singular, sustained, distraction-free, blissfully immutable voice. Instead, technology changed everything that enables a book, fomenting a quiet revolution. Funding, printing, fulfillment, community-building—everything leading up to and supporting a book has shifted meaningfully, even if the containers haven’t. Perhaps the form and interactivity of what we consider a “standard book” will change in the future, as screens become as cheap and durable as paper. But the books made today, held in our hands, digital or print, are Future Books, unfuturistic and inert may they seem.
The text of a fascinating talk given by Tim Berners-Lee back in 1995, at a gathering to mark the 50th anniversary of Vannevar Bush’s amazing article As We May Think. The event also drew together Ted Nelson, Alan Kay, Douglas Engelbart, and Bob Kahn!
Every now and then, I like to revisit Vannevar Bush’s classic article from the July 1945 edition of the Atlantic Monthly called As We May Think in which he describes a theoretical machine called the memex.
A memex is a device in which an individual stores all his books, records, and communications, and which is mechanized so that it may be consulted with exceeding speed and flexibility. It is an enlarged intimate supplement to his memory.
It consists of a desk, and while it can presumably be operated from a distance, it is primarily the piece of furniture at which he works. On the top are slanting translucent screens, on which material can be projected for convenient reading. There is a keyboard, and sets of buttons and levers. Otherwise it looks like an ordinary desk.
1945! Apart from its analogue rather than digital nature, it’s a remarkably prescient vision. In particular, there’s the idea of “associative trails”:
Wholly new forms of encyclopedias will appear, ready made with a mesh of associative trails running through them, ready to be dropped into the memex and there amplified. The lawyer has at his touch the associated opinions and decisions of his whole experience, and of the experience of friends and authorities.
My legal Memex builds a network of the people and laws available in the public records of politicians and organizations. The infrastructure for this vision relies on open data, free access to law, and instantaneously availability.
As John Sheridan from the UK’s National Archives points out, hypertext is the perfect medium for laws:
Despite the drafter’s best efforts to create a narrative structure that tells a story through the flow of provisions, legislation is intrinsically non-linear content. It positively lends itself to a hypertext based approach. The need for legislation to escape the confines of the printed form predates the all major innovators and innovations in hypertext, from Vannevar Bush’s vision in ” As We May Think“, to Ted Nelson’s coining of the term “hypertext”, through to and Berners-Lee’s breakthrough world wide web. I like to think that Nelson’s concept of transclusion was foreshadowed several decades earlier by the textual amendment (where one Act explicitly alters – inserts, omits or amends – the text of another Act, an approach introduced to UK legislation at the beginning of the 20th century).
That’s from a piece called Deeply Intertwingled Laws. The verb “to intertwingle” was another one of Ted Nelson’s neologisms.
There’s an associative trail from Vannevar Bush to Ted Nelson that takes some other interesting turns…
Picture a new American naval recruit in 1945, getting ready to ship out to the pacific to fight against the Japanese. Just as the ship as leaving the harbour, word comes through that the war is over. And so instead of fighting across the islands of the pacific, this young man finds himself in a hut on the Philippines, reading whatever is to hand. There’s a copy of The Atlantic Monthly, the one with an article called As We May Think. The sailor was Douglas Engelbart, and a few years later when he was deciding how he wanted to spend the rest of his life, that article led him to pursue the goal of augmenting human intellect. He gave the mother of all demos, featuring NLS, a working hypermedia system.
Later, thanks to Bill Atkinson, we’d get another system called Hypercard. It was advertised with the motto Freedom to Associate, in an advertising campaign that directly referenced Vannevar Bush.
And now I’m using the World Wide Web, a hypermedia system that takes in the whole planet, to create an associative trail. In this post, I’m linking (without asking anyone for permission) to six different sources, and in doing so, I’m creating a unique associative trail. And because this post has a URL (that won’t change), you are free to take it and make it part of your own associative trail on your digital memex.
The world exploded into a whirling network of kinships, where everything pointed to everything else, everything explained everything else.
— Umberto Eco, Foucault’s Pendulum
A is for Apophenia
Apophenia is the name for that tendency in humans to see patterns where none exist, to draw connections, to make links.
Every conspiracy theory is an example of apophenia. But you don’t have to be a conspiracy theorist to experience it. We do it all the time. We see shapes in the constellations in the night sky. We see faces in just about everything.
Today, I would like to engage the apopheniac in you.
A is for Anchor
An anchor is an odd device to represent a link. I can even remember seeing anchor symbols used in the interfaces of rich text editors. If it wasn’t an anchor, it was a chain. I suppose that was meant to represent a link …because chains have links.
This looks pretty familiar to us today. There’s an A Element with opening and closing tags, some text in between, and an HREF attribute for the destination. But there’s also a NAME attribute. This has since been deprecated—now we can just use an ID attribute on any element. The idea was that A elements could be used to create destinations for inbound links. They were, if you like, anchors within a page that other pages could tether themselves to. Each anchor is given a unique identifier (unique within the page, that is). Here, the identifier is simply the number zero, because this page was created by a programmer and in the mind of a programmer, counting begins with zero.
This use of the A element—using NAME attributes to create in-page anchors—never really took off. But the other attribute, the HREF attribute, that spread like wildfire. It’s short for hypertext reference, and in this particular instance, the reference is to another page in the same directory on the same server. It’s a page about hypertext.
Hypertext is text which is not constrained to be linear.
The term was first coined by Ted Nelson. He didn’t just talk about text either. He also coined the term hypermedia. He coined lots of interesting words. He talked about things being deeply intertwingled. He also coined the term teledildonics, but that is not directly relevant to hypertext or hypermedia.
If hypertext is text which is not constrained to be linear, how did we ever manage with good old-fashioned non-hyper text and non-hyper media? We used archives.
A is for Archive
In 1965, JCR Licklider was commissioned to write a report on Libraries Of The Future. He—and his company Bolt, Beranek and Newman—would prove to be instrumental in creating our modern interconnected hypermedia world. More on that later. For now, let’s look at this report. It has two parts:
Concepts and Problems of Man’s Interaction with the Body of Recorded Knowledge
Explorations in the Use of Computers in Information Storage, Organization, and Retrieval.
I love the scope of that first part, looking at the body of recorded knowledge.
It’s interesting that he talks about knowledge, not information, not data, but knowledge. How does data become information? How does information become knowledge?
The Library Of Babel is a short story by Jorge Luis Borges, who I think of as the poet laureate of hypertext . He imagines a vast library that is filled with data, but frustratingly short on knowledge …because this library contains not only all the books ever written, but all the books that could ever possibly be written, with every possible permutation.
Here’s how it works:
The universe (which others call the Library) is composed of an indefinite, perhaps infinite number of hexagonal galleries… The arrangement of the galleries is always the same: Twenty bookshelves, five to each side, line four of the hexagon’s six sides… each bookshelf holds thirty-two books identical in format; each book contains four hundred ten pages; each page, forty lines; each line, approximately eighty black letters.
Let’s figure out how many books are in the Library of Babel. First, we need to know how much data each book holds.
There are eighty symbols (or letters) per line, and
40 lines per page.
There are 410 pages in each book.
Multiplying 80 by 40 gives us 3200, the number of symbols on each page. Multiply that by 410 and we get a total number of 1,312,000 symbols in each book.
We have two other pieces of information to work with. Borges tells us:
The orthographic symbols are twenty-five in number.
That’s 22 letters, the comma, the period, and the space.
Here’s the crucial bit of information that ensures that the library has boundaries:
In the vast library there are no two identical books.
Knowing that, we can calculate the number of books in the library. It’s the number of symbols (25) raised to the power of the number of symbols in each book (1,312,000).
25 to the power of 1,312,000 expressed in base ten is 10 to the power of 1,834,097. Remember that’s just the number of books: a figure that’s over 1,834,097 digits long. That number wouldn’t fit inside one book in the library (which, if you remember only holds 1,312,000 symbols).
So the number of books in the Library of Babel is not infinite …but it is really, really, really big. To give you some idea of just how big 10 to the power of 1,834,097 is, it has been calculated that the observable universe contains approximately 10 to the power of 80 atoms. There are more books in the Library of Babel than there are atoms in the universe.
And yet, thanks to the World Wide Web, you can theoretically peruse every single one of them.
Jonathan Basile has created libraryofbabel.info—an online representation of Borges’s creation. It contains all possible text. You can browse by hexagon, then shelf, then row, then book, then page. Or you can search for a specific piece of text, because—of course—that text must be in there somewhere.
The very words I am speaking now must be somewhere in the library.
There’s also a plug-in for Chrome so you can highlight any piece of text on the web, and then find its corresponding page in the Library of Babel.
This library has every possible piece of data …but it’s sorely lacking in information (although the online version helps).
A is for All
Having all the data isn’t enough. It needs to be organised—turned into information—for us to make use of it (and hopefully further transform that information into knowledge).
There have been many attempts to organise information. When those attempts are limited to a subset of data—instead of trying to create a Library of Babel—then they can be quite successful.
Carl Linnaeus organised the natural world using a naming convention for describing species—binomial nomenclature.
Melvil Dewey created the Dewey Decimal System to help librarians organise their collections. But this was a proprietary system, not an open standard. So the Belgian librarian Paul Otlet devised a Universal Decimal Classification system. I think it’s fair to think of him as the father of information architecture.
The crazy old uncle of information architecture would be bishop John Wilkins, who in 1668 wrote An Essay towards a Real Character and a Philosophical Language. Centuries later, this would pique the interest of Borges who described Wilkins’s madcap idea in an essay called The Analytical Language of John Wilkins. The idea was that the world could be classified into sounds.
He divided the universe in forty categories or classes, these being further subdivided into differences, which was then subdivided into species. He assigned to each class a monosyllable of two letters; to each difference, a consonant; to each species, a vowel. For example: de, which means an element; deb, the first of the elements, fire; deba, a part of the element fire, a flame.
It didn’t scale well.
Gottfried Wilhelm Leibniz—inventor of calculus and nemesis to Newton—had a similar idea to Wilkins, but whereas Wilkins was trying to classify information using sounds, Leibniz wanted to use symbols: characteristica universalis. His leap of genius was to realise that if you could do this—represent the world with symbols—then you could perform calculations on those symbols. He described the conceptual framework for performing such operations as a Calculus ratiocinator. Centuries later, Norbert Wiener, the creator of cybernetics would say:
The general idea of a computing machine is nothing but a mechanization of Leibniz’s calculus ratiocinator.
A is for Ada
Leibniz’s calculus ratiocinator was an idea, rather than a real machine. Charles Babbage was a Victorian inventor and entrepreneur who was given seed funding by the British government to create his Difference Engine: a machine for computing logarithmic tables …a computer, if you will.
The Wozniak to Babbage’s Jobs was Ada Lovelace, the daughter of the notorious Lord Byron. Ada’s mother did everything in her power to steer her daughter away from following in her father’s footsteps of becoming a poet. Instead Ada immersed herself in the world of mathematics. It was through mathematics that Ada hit upon the same insight as Leibniz—if we can perform calculations upon symbols, and those symbols don’t have to just represent numbers, then we can perform calculations on anything …maybe even poetry.
In enabling mechanism to combine together general symbols in successions of unlimited variety and extent, a uniting link is established between the operations of matter and the abstract mental processes of the most abstract branch of mathematical science.
The Difference Engine was abandoned in favour of an even more ambitious project: The Analytical Engine. Finally, the world would get a mechanical machine capable of performing calculations on symbols that could represent concepts or thoughts—a thinking machine!
As with so many start-ups, Lovelace and Babbage never quite managed to make it to market. But at least they did genuinely have an incredible journey before going broke.
A is for Alan
The Analytical Engine, like the calculus ratiocinator, remained an idea. It would be another century before we’d get a real nuts’n’bolts computing machine, thanks to the brilliant Alan Turing.
But as well as having a hand in creating the world’s first physical computer, he also presented us with an imaginary machine. This theoretical machine was described as having:
…an infinite tape marked out into squares, on each of which a symbol could be printed.
Like Leibniz’s calculus ratiocinator, a Turing machine would operate on symbols. How many symbols are we talking about here? Remember the Library Of Babel had twenty five symbols to work with.
Two. Two symbols. One and zero. On and off. True and false. With an infinitely long piece of tape and infinite amount of time, two symbols are enough to calculate literally anything.
This reduction of the world into its smallest pieces was the brainchild of Claude Shannon. He coined the term "bit" to describe this indivisible unit of information.
By the way, Turing himself referred to his theoretical machine as an automatic machine, or a-machine. A is for A-machine.
A is for Atlantic
There’s another imaginary machine that serves as a wonderful conceptual prototype for working with hypertext and hypermedia. In 1945 Vannevar Bush published an article in the Atlantic Monthly called As We May Think. In this article, Bush describes a machine called the Memex, a contraction of Memory Index.
The Memex is built into a desk. Screens and switches on the surface of the desk allow the user to interact with huge amounts of information stored within the desk on microfilm. This would provide an "enlarged intimate supplement to one’s memory."
Because everyone’s mind is different, no two people would use the Memex in quite the same way. Bush described these individualistic approaches to linking concepts together as associative trails:
Wholly new forms of encyclopedias will appear, ready made with a mesh of associative trails running through them, ready to be dropped into the memex and there amplified.
The information stored in the Memex is the same for everyone, but the associative trails created by the user in navigating this information are unique. Bush also proposed that these associative trails could be shared. Users of the Memex could follow the breadcrumbs left by others.
This is hypertext.
A is for Augmentation
So many ideas about hypertext confined to the imagination! When do we get a working demo?
Ladies and gentlemen, on December 9th, 1968, we get the mother of all demos, courtesy of Douglas Engelbart. Six years previously, he had set out his goals for human-computer interaction in a work entitled Augmenting Human Intellect: A Conceptual Framework. To achieve this aim of augmenting humanity, he created a working demo of his oNLine System, abbreviated to NLS.
Oh, and for this demo he just happened to invent video conferencing, the graphical user interface, and the mouse. He also implemented hypertext.
With Bush’s Memex and Engelbart’s NLS, concepts could be linked together, turning them from text storage devices to hypertext storage devices. But for hypertext to reveal its true power, we need a network greater than anything seen up to this point. Towards the end of the mother of all demos, we hear the first rumblings of just such a network.
A is for Arpa
ARPA stands for the Advanced Research Projects Agency.
This government agency turned to our old friend JCR Licklider, author of Libraries Of The Future. His company—Bolt, Beranek and Newman—set about designing a communications system that used this new-fangled packet switching that Leonard Kleinrock was so excited about. With packet switching, information was broken down into discrete chunks, routed around a network independent of each other, and then re-assembled at the destination.
At the same time, this idea of packet switching was independently discovered by Paul Baran at the Rand Corporation, who was trying to find a resilient network architecture capable of surviving nuclear attack.
The packet-switching idea was put to the test with the creation of a new network called the ARPANET. The very first message sent over the ARPANET was at 10:30pm on October, 29th, 1969. It was the command: LOG IN.
The message was sent and… the system crashed after the first two characters. Fittingly, then, the real first message sent over this proto-internet was "LO".
But they fixed the bugs and they kept working on making the system better as it grew bigger. It morphed from being a single network, the ARPANET, into being a network of networks, or an Inter-network, soon shortened to simply Internet.
For this internet to work, it was essential that all the individual networks connecting to it were using the same protocols to communicate. That’s what really makes the internet the internet—regardless of what kind of hardware is being used, there’s an agreement on how to switch those packets around. That’s really all a protocol is: an agreement. They’re more like treaties than code.
Bob Kahn and Vint Cerf were the statesmen crafting the internet’s protocols, and Jon Postel was the diplomat ensuring adoption went smoothly. They wanted a robust network, resilient not to nuclear attack, but to any kind of top-down control. They set out to create a protocol that would work for a network with no centre.
Together they crafted TCP/IP: the Transmission Control Protocol and Internet Protocol. TCP/IP is a deliberately dumb set of protocols. The protocols care not a whit for the contents of the packets being switched around the network. It’s a simple low-level agreement. They used to joke that you should be able to implement TCP/IP using two tin cans and a piece of string.
You can then create more complex protocols on top of this simple, low-level, dumb foundation. You can create protocols for sending and receiving email, protocols for telnet, gopher, file transfer protocols that sit atop TCP/IP. Best of all, you don’t need to ask anyone for permission. If you want to create a new protocol today, you can just go ahead and do it. All you need is rough consensus and running code. It turns out that running code isn’t the hard part. The hard part is convincing people to use your protocol. Otherwise you don’t benefit from Metcalfe’s Law:
The value of a network is proportional to the square of the number of connected users of the system.
Think about it. The first person to own a telephone had a completely useless object. As soon as one other person had a telephone, it suddenly become exponentially more useful. That was the challenge facing the creator of a new protocol at the start of the 90s: that protocol was HTTP—the HyperText Transfer Protocol.
HTTP is one part of a three-part stack: HTTP is the protocol, URLs are the identifiers, and HTML is the format. Together they form the World Wide Web project.
The project was the brainchild of a young computer scientist named Tim Berners-Lee. This wasn’t his first attempt at creating a hypertext system.
In the 1980s, he created a system called ENQUIRE. It was named after a Victorian book of manners called Enquire Within Upon Everything, which I always thought would be a great name for the web.
Enquire didn’t work out in the end, but it would influence the design of the World Wide Web project.
Another influence on the design of the system was the place where Tim Berners-Lee was working. CERN—the European Centre for Nuclear Research. It’s an amazing place. The greatest experiment in the history of our species is being conducted beneath the border between Switzerland and France. In the 16-mile wide ring of the Large Hadron Collider at CERN, human beings are recreating the conditions from the start of our universe. Protons are smashed together at velocities approaching the speed of light. It’s a truly awe-inspiring endeavour.
When I visited CERN, I expected to be blown away by the science, and I was. I also expected to be blown away by simply being at the birthplace of the web, and I was. But what I wasn’t expecting was to be blown away by how things get done at CERN. There is very little hierarchy. People from all kinds of backgrounds—from Nobel prize winning physicists to students on a Summer internship—collaborate on experiments for pure scientific research.
Trying to manage the flow of information in this collaborative but chaotic place was the challenge that Tim Berners-Lee was trying to solve. You can’t just mandate a particular operating system or piece of software—people at CERN can and will use whatever they want. In the same way that the internet is a network of networks, what CERN needed was some way of allowing all these different computers with different operating systems to share information with each other.
Tim Berners-Lee submitted a paper to his supervisor, Mike Sendall. It had the uninspiring title Information Management: A Proposal. Well, Mike Sendall must have seen some potential, because he scrawled across the top:
Vague, but exciting.
The proposal described:
…a solution based on a distributed hypertext system.
Tim Berners-Lee was very familiar with previous hypertext systems. All of these fed into his project:
Vannevar Bush’s Memex,
Douglas Engelbart’s oNline System,
Ted Nelson’s Xanadu, although it was still vapourware at this point,
Apple’s Hypercard system for the Mac,
and his own Enquire project.
But just creating the code wasn’t enough. He—and his colleague and collaborator, Robert Cailliau—needed to convince the scientists at CERN to use this technology. To start with, they needed a catchy name.
For a while, they floated the idea of calling it the Mesh.
Then they kick around the idea of calling it The Information Mine. But Tim Berners-Lee wasn’t keen on this one. He knew that, whatever name they chose, it would end up getting abbreviated, and he was worried it would look a bit egotistical.
And so they settled on World Wide Web. You have to admire the chutzpah of calling it World Wide Web when, at that point, it only existed on one person’s computer.
Sure enough, it did end up getting abbreviated. Except in this case, there are actually more syllables in the so-called abbreviation—WWW—than there are in the full name.
They even made a logo. Graphic designers they are not. But there is a reason for the green colour of the Ws. Robert Cailliau is a synesthesiac—he “hears” the W sound as the colour green.
And so the web was born. Good job!
Learning from the lessons of TCP/IP, Tim Berners-Lee made sure to keep the individual parts of the system as simple as possible (but no simpler). The World Wide Web didn’t succeed because the technology was the best; far from it. It succeeded because the technology was just simple enough—but also powerful enough—for people to get started with straight away.
Take HTML, for example. There was no official Version 1 specification for this hypertext markup language.
Instead there was a document called simply HTML Tags, presumably written by Tim Berners-Lee. This document listed the entirety of HTML, which was a grand total of 21 elements.
Most of those elements weren’t even invented by Tim Berners-Lee. Instead, he borrowed the vocabulary already being used by scientists at CERN. They were used to writing documents in GML, which is supposed to stand for Generalised Markup Language, but was coincidentally created by three people whose last names were Goldfarb, Mosher, and Lorie: G, M, and L.
There was one element that was completely new to HTML:
A
This one single element is what enables the HT part of HTML. With this element, and its href attribute, anyone could link to anything on the web. It is brilliant in its simplicity.
Pleased with the way their project was progressing, Tim Berners-Lee and Robert Cailliau submitted a proposal to present their World Wide Web at a hypertext conference.
They were rejected. Hypertext experts thought the World Wide Web was stupid.
As Ted Nelson put it:
Today’s one-way hypertext—the World Wide Web—is far too shallow. The Xanadu project foresaw world-wide hypertext decades ago, and endeavored to create a much deeper system. The Web, however, took over with a very shallow structure.
He’s not wrong. Hypertext on the web is shallow. It is stupid. Like TCP/IP, it is not smart.
The thing is, if you’re hoping to get mass adoption, being smart is a bug. Being stupid is a feature.
Just about every other hypertext system embodied the idea of two-way linking. There was an awareness at both ends of the link. If the resource being linked to were to move or change, the link could be updated. It’s robust, but it’s complicated.
On the World Wide Web, by contrast, links only work in one direction. If the resource being linked to ends up moving or changing, well, tough luck. The result is link rot. That’s the price we pay for a very simple hypertext system.
But, now that the web has been around a couple of decades, there is a sort-of, kind-of implementation of two-way linking.
It uses the humble rel attribute.
Rel is short for relationship. The value inside the rel attribute describes the relationship of the linked resource to the current document.
<a href="…" rel="…">
Some rel values were officially canonised in HTML.
rel="prev" means that the linked resource has the relationship of being the previous document to the current document.
<a href="…" rel="prev">
rel="next" means that the linked resource has the relationship of being the next document after the current document.
<a href="…" rel="next">
rel="author" means that the linked resource has the relationship of representing the author of the current document.
<a href="…" rel="author">
You get the idea.
Incidentally, there was once a corresponding rev attribute that described the reverse relationship. In other words, the relationship of the current document to the linked resource.
<a href="…" rev="…">
So you could simultaneously say the current document has a relationship of being the previous document to the linked resource and that the linked resource has a relationship of being the next document to the current document.
<a href="…" rev="prev" rel="next">
Confused? Well, that’s why the rev attribute was eventually dropped from HTML. It was just too complicated.
There was an attempt to use the rev attribute in an early microformat called vote-links. This was proposed way back in 2004. Blogs were hot. Political blogs were very hot indeed: Bush Jr.—no relation to Vannevar—was running for re-election in the States, and John Kerry was the challenger. There was much blogging and gnashing of teeth.
The problem was the rise of PageRank, the algorithm that drove Google’s search engine. The fundamental premise of PageRank was that linking to something counts as an endorsement. But there were many bloggers linking to articles that they disagreed with very strongly.
By using a rev value of “vote-for”, authors could explicitly say that this document is a vote for the resource being linked to.
Or by using a rev value of “vote-against” they could make it clear that this document is a vote against the linked resource.
It never really took off because, as I said, the rev attribute was just too hard to grok.
And that’s okay. The whole point of microformats is that they are the very embodiment of the motto of the Internet Engineering Task Force: rough consensus and running code. There just wasn’t enough take-up of vote-links for it to thrive.
Another early idea—that actually preceded the official creation of microformats—was XFN, which stood for XHTML Friends Network: the most Web 1.0 format name ever.
XFN built on existing behaviour. Bloggers would often have a list of links in their sidebar pointing to other bloggers they had some kind of relationship with. If you assume that a URL can represent a person, then the rel attribute is perfect for encoding that relationship information.
I can link to a friend’s website and say that the person represented by the linked resource has a relationship of being a friend to the person represented by the current document: me!
<a href="…" rel="friend">
Or I can link to a colleague’s website and say that they have a relationship of being a colleague to me.
<a href="…" rel="colleague">
And because rel values—like class values—can be space-separated, I can combine rel values into one attribute. I can link to someone and say that they are both a friend and a colleague.
<a href="…" rel="friend colleague">
I still have XFN values in the sidebar of my blog, but again, it never really took off.
Except for one value, that seems at first glance to be completely pointless:
A rel value of "me": the linked resource has a relationship of being …me?
<a href="…" rel="me">
A is for Adactio
My website is adactio.com. I love my website. Even though it isn’t a physical thing, I think it might be my most prized possession.
It’s a place for me to think and a place for me to link.
Nothing unusual there. These are regular one-way hyperlinks.
What’s interesting is that many of those profiles on other websites provide a URL field where I can enter my own website. These third-party profiles then link back to my website also using a rel value of "me".
<a rel="me nofollow" href="https://adactio.com">
(They also use a rel value of "nofollow" to discourage spammers. The phrase "nofollow" makes absolutely no sense as a rel value—you can’t have a relationship of "nofollow" to anything—but it was invented by Google. We don’t get to argue with the 900 pound Google gorilla.)
Anyway, the result of having these reciprocal links, both using rel="me" means that we’ve kinda, sorta got two-way linking on the World Wide Web.
But …so what?
A is for Authentication
Some of those third-party profiles I’m linking to—Twitter, Github, Flickr—have something in common. They allow third-party authentication using OAuth.
If I can log into my Twitter, or Github, or Flickr profile using OAuth, and those profiles have two-way links with my website, then I can “borrow” that authentication flow for my own site.
That’s the idea behind IndieAuth. I enter the URL of my own website. It finds the links from there to my other profiles using rel="me". Then I can choose which one of those profiles I want to authenticate against. Once I’ve authenticated with that service, I’ve also authenticated my own website.
Just by adding a short rel value to some links, I can use my website as a log-in.
My website also has a small write API. The API is called micropub. By combining IndieAuth and micropub, I can log into somebody else’s posting interface using my website, and then use that interface to post to my own website.
These building blocks—microformats, IndieAuth, micropub—have emerged from a small but dedicated group of people.
We gather together at fun little events called Indie Web Camps.
An Indie Web Camp takes place over two days. The first day is split into discussions of technology and design. The second is all about implementing what we’ve discussed on the first day. I’m always amazed by how much gets done when you’re in the same space as like-minded people.
Still, the technologies being discussed and implemented aren’t the real focus of Indie Web Camp. The core tenet of Indie Web Camp is an idea. It’s a very simple idea, that at one point would have been uncontroversial. That idea is that you should have your own website.
But who’s got time for that? Especially when it’s so much easier to write and share and link using platforms like Twitter, Facebook, and Medium. That’s a good point. A lot of the time at an Indie Web Camp is spent reverse-engineering what those services are doing so well, and applying them to personal websites.
Also, we don’t necessarily want to stop using those services. After all, that’s where the people are (for now). Instead we want to figure out ways of making use of these services, while still keeping the canonical versions of what we create under our own control.
There’s a fantastic little Indie Web service called Brid.gy that allows you to not only post from your own site out to third party networks, but also receive replies, and likes, and retweets back at your own site, using another Indie Web building block called webmention.
In a way, webmentions allow a kind of two-way linking. I can cross-post something from adactio.com to Twitter, or Instagram, or Facebook. Then when someone replies on Twitter, or Instagram, or Facebook, I get notified with a ping back to my own site.
All of this is possible because I’ve verified the identity of those third-party profiles using nothing more than a simple rel attribute on a hyperlink.
The Indie Web uses a grab-bag of deliberately simple technologies that combine to create something so much more powerful than the sum of its parts …just like the web itself.
We often talk about places like Facebook, or Instagram, or Medium as walled gardens. Walled gardens are as old as the web itself. AOL was a well-cultivated walled garden back in the day.
There’s nothing wrong with walled gardens. They’re safe spaces. They take care of your enjoyment and entertainment, so you don’t have to.
But there also a bit boring. I certainly don’t relish the idea of spending my days within the boundaries of someone else’s vision.
The Garden of Forking Paths. It is uncontrolled. It is full of possibilities. It’s a bit scary. It takes more dedication to explore. You might get lost. But is that so bad? When was the last time you were truly lost on the World Wide Web, when you clicked through link after link—no cheating by opening new tabs, now—until you ended up somewhere, blinking and asking yourself “what I was looking for?”
I would like us all to spend more time in the garden of forking paths. I would like us all to continue to grow this garden of forking paths. Add your own website to this garden of forking paths. Use it to make more links.
On your website, you can link to this thing over here and that thing over there, and in doing so create an entirely new forking path.
Remember, the web, like the internet, has no centre. In theory I could start from any single A element, and by following all the forking paths, traverse the entire World Wide Web.
That opening hyperlink could be on your own website. One single A element can be the portal to an entire universe of knowledge.