I Had a Dream

On how we might write in the future, fountain pen fetish, and whether the new paper will come from China… again.

It was one of those dreams I, for some reason, didn’t want to forget. Even in the haze of REM sleep at 2:43 AM, I willed myself against my flesh to jot down some fleeting thoughts on it, saving me from the next day’s frustration and alleviating the Zeigarnik Effect. Here they are.

The Dream Notebook

In that dream, I was holding a notebook that resembled my LEUCHTTURM1917 Bullet Journal edition, with its deep green hardcover and thick paper pages. Ryder Carroll’s BuJo resonated with me only partially, but that notebook was by far the best-quality gear I had used so far, surpassing even the manufacturer’s own standard product line. It definitely set a benchmark for what a great notebook should be. It had the same satisfying tactile sensation of heavyweight page-turning. Of course, this wasn’t real paper. Seemingly, it resembled a high-tech material that was flexible and paper-thin, but actually it was a matte interactive surface capable of responding to markings and multitouch controls. More on this later.

This was, by far, the most distinctive feature of this imaginary stationery marvel compared to anything we have today: flat, thick tablets with a matte screen protector affixed to mimic the feel of a paper-ish surface.

Wishful thinking? Perhaps. But consider the evolution of a notebook. How far we have come, and how primitive the early information mediums seem to us today, to the point where they are now museum exhibits showcasing early ingenuity. It probably started with cave walls, then ancient Mesopotamian Sumerian clay tablets and their wooden styluses carving Cuneiform, followed by Diptych Assyrian wax tablets made of wood or ivory, leather, parchment, bamboo, PDAs, and tablet screens. The origins of our note-taking feel as distant as the Devonian period. One day, the most advanced Remarkable II will seem like a clay tablet. However, amid the Great Horse Manure Crisis¹, where horses were swiftly substituted by automobiles, it is impossible to know what will replace them. That night, my mind began pondering these possibilities.

Neo-Assyrian wax tablets for note-taking. Source: labrujulaverde.com.

So, it had these advanced, high-tech pages. OK, but why do we even need pages? Paging is a relatively brief chapter in the history of recording information. Yes, physical pagination feels natural and very close to reality, but we don’t always have to imitate the past. In fact, many note takers complain that the spine prevents the notebook pages from lying flat, unlike on a tablet. For example, my current setup includes brass clips and a hard inlay. It prevents me from inserting as many refills into my leather binder as I would like. It takes about a dozen micro-gestures before I can start writing. That’s a very poor user experience.

Sam Altman mentioned that he chose spiral-bound notebooks because their flatness was crucial, as was the ability to tear pages out occasionally. Essentially, he used these spiral-bound notebooks as collections of index notes to jot fleeting notes on. For the Dream Notebook to meet Sam’s needs, the technology would need to become very affordable and widely available before he could tear out pages and discard them without a second thought. However, it’s not impossible. A Bic Crystal pen was a marvel when I first began writing. It was difficult to find. Today, it’s a disposable pen sold by the kilogram. The French company has sold over 100 billion since 1950². When such “connected paper” approaches the cost of regular paper, or becomes even cheaper, then the new economics would make a device like a spiral-bound connected paper notebook a likely item in Sam’s pocket—and certainly in mine.

The Stylus

Like many aspiring writers, I fell for the romanticised idea that a fountain pen is a superior writing instrument, despite its many obvious flaws. I need not list them here; the internet is full of these discussions. I like fountain pens, and the benefits I gain from them outweigh their inconveniences. Therefore, it’s only natural that the stylus I used with the notebook in my dream resembled a traditional, minimalistic model, just the way I like them. Physically, it had a few features that set it apart from a standard pen, and we’ll discuss these below. However, overall, it was almost a carbon copy of my daily writing instrument: a normal-sized, streamlined, hexagonal barrel with a golden nib. It felt as though I was simply using a high-quality analogue of a LEUCHTTURM1977 notebook with a Y Studio fountain pen. I felt at home.

Almost any digital stylus nowadays lets you mimic the feel of a traditional writing instrument, such as a brush or a fountain pen. You can then adjust the wetness, hardness, width, or quality of both the stylus and the surface it writes on. That’s a positive aspect of virtualisation. Yet, as discussed in the multitaskers versus unitaskers essay, what we often want is the opposite of versatility. One of the greatest pleasures of collecting stationery is precisely the act of acquiring and carrying several pens or brushes because they are one-trick ponies. There’s also a logical flow disruption when one has to navigate clicks and menus to change the nib instead of simply putting down the pencil to pick up the eraser. Yes, one could use key bindings if, for some reason, your notebook was also equipped with a keyboard, or rotate between options with button presses, but these all feel like trade-offs. I’m not sure what that is, but it feels different. When thinking on paper, micro-distractions accumulate. For me, carrying a fountain pen, a brush pen, and a mechanical pencil in a small pouch works best. I don’t experience any context switching when transitioning between these three writing instruments in front of me.

In this case, I selected a stylus that closely resembles the effect I aimed to achieve. In a brilliant book, “The Design of Everyday Things,”³ the hallmark of a great user interface is that the control should be as close as possible to the object it governs and as similar in appearance as possible. Therefore, using a fountain pen as a controller to mimic fountain pen writing is a natural extension of a great user experience, UX.

Strokes

Fountain pen nibs come in various shapes, sizes, materials and grinds. Some exotic variations, such as Naginata, feature four tines, but a classic nib has two, and they are usually the same size. However, in my dream, the right tine was noticeably broader than the other, at its tip, though not along the entire length. I wondered why I pictured it in such a strange way. After all, dreams often distort reality, so much so that lucid dreaming practitioners suggest using the palm of a hand as a test to tell if one is asleep. Usually, your hand might appear with an unusual number of fingers, wavy, or not like a hand at all. But after some thought, I realised this difference in tines might have been because I could control the width of my strokes with this stylus. Not that it was the only way; after all, light pressure on the nib usually creates line variation, which we even use fountain pens for, but this could allow an even greater range of strokes. Still, I don’t think it would be a good idea, as it could cause confusion about how to generate line variation—whether by pressure or software nib settings. It would also be an utterly uncomfortable writing experience and would distract from the user experience advantages of a “unitasker” discussed above.

What would be wonderful, however, is if the nib’s flex could be configurable. This might be achieved either through software (which is not easily done) or by swapping a physical nib, as you would with a real fountain pen. Imagine how great it would be to select your nib flex level at the time of ordering, and also to expand the collection of compatible nibs later on. Certain companies, such as Logitech and Wacom, produce compatible after-market styluses for touch devices. I can easily see nib-making companies jumping on this bandwagon and creating compatible pens and nibs.

Colours

My ink collection is quite modest, with only five bottled variants and a few standard cartridges. I favour some more than others, and like any other fountain pen enthusiast, I’d love to have shelves filled with beautifully crafted bottled inks.

Finding the perfect combination of pen, nib, paper, and ink that feels like a match made in heaven is a favourite pastime for stationery enthusiasts. However, this process takes time, as filling a pen with a different ink is not just a simple lift-and-shift. You need to empty the reservoir, thoroughly clean and dry the entire system, then refill it with compatible ink and write for a while before forming an opinion. If you’re not a fountain pen user, this might seem like a refined form of masochism, and in some ways, it is, but it’s also an immensely pleasurable kind of pain; a geeky, meditative ritual that’s part of the experience.

In my dream, swapping inks was as simple as a click or even a mind projection. I could select a branded ink and continue using the same writing instrument, effortlessly cycling through the colour options. Wouldn’t it be wonderful to add (or buy?) digital inks from the artisans we admire in the physical world? If you appreciate the wetness of Diamine’s renowned Ox Blood or the sheen of Octopus Fluids’ Copper, you could experience a similar effect in your digital notebook.

Highlighting

A common ink-flow issue in fountain pens is called “railroading”. Sometimes, while writing, you may notice a gap between two fine, parallel lines left by the two tines of a nib. This pattern resembles a railway, which is how the phenomenon gets its name. The cause is that the ink flow is too weak. It’s a frustrating side effect that occurs when not all stars are perfectly aligned. The viscosity of the ink, the width and depth of the ink feed, the flexibility of the nib, the pressure from the paper, the ink level in the reservoir, and your writing speed all affect the likelihood of railroading occurring.

Fountain pen railroading example. Source: etchrlab.com

This is often seen as a disadvantage of writing with a fountain pen. Other ways of marking paper don’t have this strange problem. However, in my dream, this bug became a feature. By pressing harder, I could spread the tines wider, which would normally cause a rough railroading effect in real life. Instead, in this dream, the notebook interpreted this as a signal to highlight the underlying text. The pressure I applied would control how thin or wide the highlight became. I could even vary the width within a single stroke, creating a distinctive, irregular, analogue appearance that I found very appealing. Although I didn’t see this clearly in my dream, I was aware that the notebook’s software would likely smooth out these irregularities to varying degrees.

Reflecting on this advanced feature, I realised it was inspired by the Digital Audio Workstation (DAW) software I’ve used during my music career. There, you could play notes synced with a metronome and choose to leave them as they are, nudge them slightly to the grid, or fully quantise them for perfect timing. More variability in the notes made the sound groovier and more analogue, while less groove produced a more electronic, polished sound. Since there is no single right sound, the software leaves the decision of what’s right to the musician. Similarly, the notebook’s software would help you fine-tune your highlights to the desired level of neatness. You can already observe such behaviour in everything that interacts with a digital stylus.

Stains

Inks, being water-based, can take a long time to fully dry. However, a very characteristic “bug” of the experience of writing with a fountain pen is accidental smudging of the writing with your hand. Although inky fingers and occasional dots left by pressing against a page with partially dry ink definitely add character, often, you don’t want any of these artefacts. In my dream, this artefact was also configurable and could be adjusted before or after the writing. Similar to the quantisation of highlighting, you could also add or remove characters from the final result by nudging the neatness slider for the notebook, a particular page, or even a specific zone you could select with a lasso-like effect.

Most importantly, the stylus wouldn’t leave any unwanted marks when used for navigation and controls, but only for writing.

Navigation & Controls

Although the notebook’s pages felt like analogue, high-quality, fountain-pen-friendly paper, they were touch-enabled, with all the features of a typical touchscreen we rely on, along with a few interesting twists.

Naturally, it supported multi-touch gestures, such as pinch-to-zoom. If I could zoom in and out, I suppose one could potentially turn any page into an infinite canvas for writing and drawing.

I could click on words and elements to navigate to them, even though I don’t remember creating any explicit links. There was an implicit, somewhat opinionated entity-recognition mechanism. It’s as if, while I was writing, the notebook generated an equivalent of a knowledge graph that linked to either the most relevant individual instance of a concept or to an index of related items for me to choose from for further exploration, similar to a Map of Content, MoC. Perhaps it was even creating a Graph-RAG-like data structure behind the scenes, enabling future semantically meaningful notes retrieval and possibly a conversation with your notes from within the notebook itself, à la Notebook LM.

Unlike the now ubiquitous “chat with your documents” feature, the internet connection was not required, and the computation could be carried out entirely at the edge, creating a local knowledge graph of the notebook instance’s information. However, I assume that the notebook, being a sophisticated digital device, was also connected and could synchronise with the rest of my notes that were not on the device. Although I’m sure a device of such sophistication could contain my entire “second brain”. Perhaps it could even function as Time Bereners Lee’s pod, which, I suppose, would also have become ubiquitous by the time a device like that was in everyone’s backpack and connected to others.

Anyway, this seamless navigation is another example that could remove the need for pages you can flip through. If clicking around, like on a typical tablet, could take you to any other connected page, why would we need so many pages in the notebook at all? Then again, I’ve already seen smartphones with multiple screens and/or screens that change behaviour when folded or unfolded. I’ve also seen futuristic Chinese washing machines with three independent compartments for clothes, underwear and shoes, helping with concurrency, hygiene and energy savings.

Perhaps this multi-page system offers benefits we can’t currently envisage. For instance, being able to “tear” a page from the notebook and “share” it with someone who doesn’t have their own. Also, who would expect everyone to carry one? An interesting side effect of such a “sharing” feature is that the page wouldn’t become stale the moment it was removed from the notebook to be given away. I could update its contents remotely. And, when no longer needed or swiftly transferred to someone’s permanent notes vault, that page could simply be shredded without any environmental impact.

This remote update issue might be specific to certain uses. In many cases, extracting it from a notebook would result in a cryptographically protected, read-only file. I don’t think a personal notebook would need such a mechanism, but what about sharing legal documents? Many accountants have nearly faced jail for using loose-leaf sheets of paper for “immutable” accounting ledgers. Being able to mathematically prove that the document wasn’t altered or that it occupies a very specific position in a series of immutable ledger entries is a highly valuable feature.

One way the user could navigate the contents of the interconnected notebook is by keeping one finger on the page while pressing an item with the fountain-pen-like stylus. This item could be a word or a drawn object. Yes, the notebook would be able to recognise entities and link them to other, semantically relevant entities within the knowledge graph of that second brain. So, essentially, everything would become a meaningful, clickable zone, apart from perhaps linguistic connectors such as “and, or, but, etc.” This might have sounded like total science fiction a few years ago, but today I routinely do this for my clients who have large data lakes of unstructured data, such as photographs, sketches, or even videos of them drawing.

Regarding sketches, those could also be adjusted—such as by requesting a high-fidelity drawing or a 3D model from them, or by making them appear more childlike and analogue—thereby varying the level of fidelity and realism.

Although the idea of the notebook recognising concepts and entities and turning them into hyperlinks is intriguing, in my dream, I saw an instance in which the device struggled to recognise the handwriting of one of the words. I’d write over it to nudge it, but to no avail. I had to use another word instead. I guess bugs aren’t going away in that distant future either. But despite this minor glitch, Mum — who was also in the dream — said “wow” when she saw me manipulating this device. I suppose it wasn’t that widespread when I had it in my dream, yet. That’s normal; I love the latest and greatest, even though I rarely buy any gadgets.

As We Might Write

I agonised over whether to title this essay “I Had a Dream” or “As We Might Write,” and ultimately chose the former. Both titles are somewhat pretentious but also reference significant, forward-looking texts. The first is, as you’ve guessed, that of Martin Luther King Jr., while the second is of Vannevar Bush and his “As We Might Think.” I appreciated the subtle wordplay of “As We Might Think” plus “Writing is thinking on paper” equalling “As We Might Write.” However, I eventually decided to “plagiarise” King’s title because he did the same to others as well⁴.

Vannevar Bush’s futuristic projections, formalised in a compelling description of his Memex⁵, seem grotesque in light of today’s technological progress. In an era where circuit boards with tolerances of just a few atoms are printed on ultra-thin silicon wafers, and storage is almost unlimited, one can only smile at how academics who rode the bleeding edge of technology imagined us using analogue cameras strapped to our foreheads, operated by cables hidden in our sleeves. But that’s not the point. Whatever form it took, the ability to anticipate the ubiquity of speed-capture and its subsequent integration into a personal data lake materialised, even surpassing his wildest dreams.

I’ve shared the intimacy of my nerdy dreams, knowing it made you giggle. But I’ll be alright when we finally meet, whether virtually or face-to-face, grey-haired and wrinkled, wielding a notebook that would have far exceeded the wildest projections shared here (insert Hans Zimmer’s Interstellar soundtrack).

One of my early engineering ventures took me to Shenzhen, China, where I was developing a hardware device. The first version of this prototype was surprisingly large, especially considering its purpose: a connected sex toy for men. To create the desired realism and tactile experience, the company needed to engineer an unprecedented number of degrees of freedom into a stacked-motor mesh that would manipulate a silicone tube. To do this within a reasonable timeframe and with enough force, we built a dome around what was essentially the size of a condom, but about eighty times its volume. Needless to say, although there were many, the initial reactions of the aspiring beta testers were eyebrow-raising: “How safe is it?” How safe would it be to insert it into a device that resembled a bazooka? I don’t know. We ultimately never conducted the real tests…

Kaleidoscope. Conceptual prototype. This would then sit on your lap.

The reason I mentioned this experience is that it taught me a vital engineering principle: solving the same problem with strict size constraints is a very different engineering challenge. Sometimes, reducing the size by 10% could cause the unit cost to triple. But in most cases, the problem would remain unsolvable, at least with the technologies available at the time. Creating the kind of smart paper I envisioned feels very much like that challenging engineering task. Or is it?

As I was writing this article, a short video appeared on my social feed showing a woman at a Chinese fashion show wearing a dress made of a thin, flexible material that changes colour and texture, making it look like a different dress each time. Additionally, she wore gloves crafted from the same material as the dress, either mimicking its style or complementing it with a different one. This is not the first time a technology like this has been promoted to a techno-optimist like me by social network algorithms. However, this time, it was not only very timely but also demonstrated how thin this projection surface was. Essentially, it is just as thin or even thinner than a 90-gram sheet of paper used in high-quality notebooks.

Electronic fabric by Huisheng Peng’s Group. Source: lucept.com

Are we there yet? Will the notebook of the future come from China? Will this nation invent paper for the second time, as they did 1920 years ago?

If you also dream of notebooks and pens, leave a comment. I’d love to know I’m not a weirdo.

1. https://histary.es/en/the-great-horse-manure-crisis-of-1894-how-victorian-london-faced-an-urban-apocalypse/

2. https://primalnebula.com/the-story-of-the-bic-cristal-pen/

3. Norman, D. A. (2013). The Design of Everyday Things.

4. Martin Luther King Jr.'s Organizational Systems

5. Bush, V. (1979). As we may think. ACM Sigpc Notes.