Plenty of people I have spoken to have assumed that I am interested in them and/or invested in them, the former of which I'm not sure of, and the latter of which is not (currently) true.

Saying that I don't find NFTs interesting isn't quite right - it's just that the technology behind them is boring in comparison with the social phenomenon and questions that they raise about ownership. Let's get into some of those.

What is an NFT?

Put shortly and without going into technical detail (as I said - I find that bit boring) an NFT, or non fungible token, is a representation of ownership on an immutable ledger (i.e. a blockchain). Think of it as an entry in a public database that says X owns Y, that a huge number of people have to agree is true and correct before it's stored in such a way that it can't be changed.

If I draw a picture on my computer of a horse, or something, then create an NFT on a blockchain and announce to the world that ownership of this horse is conferred by that particular NFT, I can then sell it to you. Once it has been transferred, you now "own" the picture of the horse (and can then sell it to someone else).

This is the kind of NFT that has been making waves in the press, as artists sell their work for staggering sums (reportedly $69m for one artwork, though there are lingering doubts about the authenticity of the method, payment, whether it qualifies, etc).

Scarcity from abundance

There's a famous phrase that Marc Andressen attributes to Jim Barksdale: “there are only two ways to make money in business: One is to bundle; the other is unbundle.”

It's always fun when you can put down a line in the sand like this - for any business, are you bundling or are you unbundling? I wonder if you can say something similar about technology making things either more abundant or more scarce.

It's so obvious that it might seem banal - technology has made a bunch of stuff that was scarce abundant. Video content used to be manufactured by a handful of large corporations, but YouTube took over TV in terms of video hours watched in the USA in 2017. A certified course of education could previously only be gained from a school or university, but now you have a myriad of free and paid options to choose from online.

Images were one of the first things that the internet made abundant. An image of a digital artwork, once distributed over the internet, can no longer be "owned" in a way that makes the ability to view it scarce like with a physical artwork. If you have the set of instructions to remake the image (i.e. a file of it), you have the image in the exact same way that the artist has the image. We consider the artist to "own" the image in the sense that they control the copyright, but any attempt to constrict the spread of the image would be futile.

One thing I find interesting about blockchains (and the stuff that you can do with them) is that they are using technology to create artificial scarcity. Only a certain number of bitcoins can ever be mined, and there can only be one NFT per artwork (in theory, anyway).

While it's difficult to envisage what a currency might look like with no scarcity, I generally see the creation of artificial scarcity as a negative force in society. It makes me think of things like designer clothing brands destroying $38.8m of overstock to preserve brand value, or diamond miners investing millions in marketing to convince customers lab grown isn't as good as natural (with prices moving from parity to 50% below in around a decade).

I suppose it's quite difficult to see a loser in the NFT system in the same way (putting to one side concerns about energy use from validating the transactions on the blockchain). An artist can get paid for selling the "ownership" of their digital artwork, but in a way that does not actually restrict access to the end product for the wider consumer.

Which raises the question - what exactly are NFT owners buying?

Selling a tweet

One of the most widely circulated stories in the past weeks has been Jack Dorsey selling his first tweet as an NFT for more than $2.5m.

So what exactly was he selling? According to the FAQs on the seller site, a "digital certificate of the tweet, unique because it has been signed and verified by the creator".

A few thoughts on this:

1. Presumably, if Twitter collapses, and whoever buys it out does not choose to preserve Jack Dorsey's tweet in any meaningful way (i.e. that web page does not exist, the entry in the twitter database for the tweet does not exist, etc), then the "owner" has not lost the ownership - they own the "idea/concept" of Jack Dorsey's first tweet as encapsulated in said "digital certificate". Something like how Disney owns the concept of Mickey Mouse, and keeps (successfully) lobbying for extra years of copyright protection to keep it that way, but without any of the protections on that ownership.
2. Presumably, the "owner" can't transfer "the tweet" into his or her own private twitter - they only own a certificate, and anyway, the nature of what they have purchased only exists because it is on the twitter network and it is posted on Jack Dorsey's feed.
3. Presumably, the "owner" can't make Jack Dorsey take the tweet down, like you could take a real piece of art down from a gallery and store it away from public view. Where would it be being taken down to?
4. Presumably, the "owner" can't sue me if I take a screen capture of Jack Dorsey's tweet and put it on a t-shirt, because they don't own the reproduction rights or the rights to income from an image (like artists do even for digital art, or those that buy those rights from an artist).
5. Presumably, the "owner" can't stop a gallery from printing an image of the tweet, or the underlying code that the tweet is formed of or some such, and displaying it in their gallery and charging entry to see it.

If each of these presumptions is true, then in what sense is the owner of the NFT the owner of Jack Dorsey's tweet? Well, you might reply, he's the owner because we all believe he is the owner - just like how a Bitcoin is worth a dollar amount because we all agree it's worth a dollar amount.

By the transitive property, if we all agree that owning the "idea" of Jack Dorsey's first tweet via a digital certificate (even if you're not entitled to any of the rights) is worth a dollar amount, and that ownership is conferred by this NFT, then this NFT is worth a dollar amount.

Hang on. What about in this situation?

Jack Dorsey goes back on his word, and says that the owner of the NFT is no longer the owner of the tweet. Maybe he says it's now another NFT - or simply that it returns to him.

What recourse does the owner of the NFT have with Jack Dorsey? Well, you might say - they probably have legal recourse to sue him for fraud in selling the NFT. But that probably wouldn't return them their ownership (in any meaningful way), just their money (plus some more, one might assume).

If they're looking to retain their ownership, has Jack Dorsey signed a legal contract which states that the owner of the token is the owner of the tweet? I would assume so - so who has that contract been signed with - the current owner? Does the contract transfer over if the current owner sells the NFT to someone else?

If there is said contract (or certificate), then is the NFT the thing that really backs the claim of ownership? Or could you remove the NFT entirely from the loop, and in fact it is the name that is on the contract that defines the ownership?

Where this leads me to is three thoughts on NFTs:

1. It is ill defined what you are purchasing when you purchase an NFT.
2. It isn't clear that you need a blockchain to confer ownership in this way. You may as well just have a legal contract, or a piece of paper, that declares you as the owner that can be bought and sold.
3. I don't find the concept of ownership that NFTs provide especially satisfying, and if I had a couple of million dollars to invest in art I would buy a painting.

But I don't imagine the new owner will care very much about the kind of conceptual ownership they may or may not have if the value of their "digital certificate" skyrockets.

Banksy and NFTs

Here is an idea. As far as I know, Banksy has no plans to make an NFT - but if I were him, I might do that right now, because it would earn him lots of money and probably expose some of the weird contradictions around ownership and NFTs.

Here's what Banksy could do. He could find a wall, like he always does, and spray paint one of his stencils onto that wall. It doesn't even really have to be a good one (in fact, if it's pretty shit, that's probably better). Hey presto - you now have a valuable work of art!

Who "owns" that Banksy as a piece of art - the person who owns the wall, or Banksy himself? Well, sometimes they are on buildings owned by the council, or the national rail network, or a private individual. I might suggest that Banksy picks the last of that list, to really amp up the confusion.

When a new Banksy is confirmed by the artist, typically the person who's wall it has been sprayed on moves to preserve the now-valuable piece of wall. I imagine their eyes turn into dollar signs the moment "do you think that might be a Banksy?" leaves someone's lips. The art collectors come calling pretty soon after.

Once the wall has been properly protected, and purchased by a rich collector, Banksy should announce that he's selling the NFT of the artwork to one lucky buyer for several millions of pounds. I assume there will be a massive incentive for said collector to pay up, because now his "ownership" is at stake. He can own the physical bricks, and he can own the spray paint on those bricks, but now he needs the NFT too - or does he really own it?

Else, someone else will buy it. Then there's an NFT owner and a physical owner out there, with their competing claims on the artwork. Would that make it the first piece of art where one person owns the "idea" and the other person owns the actual artwork?

Flash in the pan?

I'm not sure I have very strong feelings about whether NFTs are here to stay, or whether this is the zenith of a huge bubble about to burst. They are sustained by collective belief that trading ownership in this way is valid, and so long as that's the case, I can't see why they wouldn't continue to be traded and sold (albeit perhaps with slightly less speculation on future value of that ownership).

Or maybe an artist will do something that drives a wedge between real ownership and NFT ownership, and we will find out where the cracks are. Either way, I will be following - but probably not buying.

Feedback on this article? Something I've missed? Comments to elliot@pkld.io.

As such, it seems almost too obvious a subject to break down into a framework - but I've made mistakes here in the past, and seen others make them too. Here's a stab at a model of a type of pitfall when communicating ideas to others.

A framework

When you are explaining a subject to someone, you are balancing the power of your description with the ease to communicate what you want to explain.

The descriptive power is how far the explanation goes into the detail of the idea you're communicating. A model with low descriptive power doesn't have as much nuance as one with high descriptive power.

The communicative ease describes how well information is passed on. Length is a key factor, but also things like prior knowledge needed and whether specialist terms are used.

The crux of this is: when a model isn't descriptive enough, it has too little power to explain its subject matter - but it is very easy to communicate. When a model is too descriptive, it has lots of power, but it's very hard to communicate.

Here's how I see it in practice:

This might seem a little glib, but I'm not claiming this model is comprehensive - it's just one way of looking at this. In that vein, there are three other key factors at play here:

• What the base complexity of the subject matter is;
• Who the audience is;
• How comprehensive the explanation claims to be.

The goldilocks zone is a moveable feast; submitting a study to an academic journal and giving driving directions are starting from two very different points. And when giving driving directions, if they are prefaced by "I'm not quite sure as I've never been, but I think it's near..." then it's much less of a problem when they have low descriptive power.

I think how comprehensive the explanation claims to be is the key point here. When something claims to be comprehensive, and is very easy to communicate, it's often mislabelled as "common sense".

In this kind of case, it's pretty easy to hide that the model given doesn't have the descriptive power to back up the claims being made. There are plenty of examples from the political world here (see: "our country is in too much debt, so we need to tighten our belts and stop spending").

Getting it wrong

Here is an article titled "Traction vs Growth" that I believe sits in the low descriptive power, high communicative ease space - but that also falls foul of the "comprehensiveness" test (sorry to pick on it, it's just the most recent example I've seen).

Although it opens with "Broadly speaking", it goes on to be pretty proscriptive about the "three phases of growing"  that a startup goes through. The first stage it talks about is "Traction", and the article gives a suggested metric to focus on while in this stage:

Plain and simple your eye should be on retention.  If your product does not retain users, there is no point in growing the top of the funnel.

Hang on - what if my startup, for instance, sells mattresses? My understanding is that you are supposed to replace a mattress every 6 to 8 years. That's a long time to wait for my customer retention metric to play out.

In fact, I can think of plenty of other examples of business models where the retention metric isn't as important as others in the early stages. But explaining that would make the article much less snappy - and so we are in the muddy waters of the "potentially misleading" part of the diagram above.

Here's another part of the article from the "Transition" phase, regarding the advice to focus on just a single marketing channel:

The fastest way to increase growth rate is to expand on something that is already working (until you’ve saturated the channel) rather than trying new channels that you know nothing about.  

My response to that would be: Well, maybe. For some businesses that might be the case, but I think it ignores a lot of the nuances around growth in marketing channels.

In my experience, sometimes a channel looks like it's not working for you - then something in your product and/or marketing changes that unlocks it. On the other hand, sometimes a channel bottoms out much faster than its initial promise would suggest. There's also a pretty big range of what "working" means.

I don't think these are the only assertions in the article that fall at this hurdle. I think the entire framing of startup growth into three phases is an attempt to squash complexity down for communicative ease purposes, and I'm not convinced it's a model that would have had enough power to explain any of the early-stage startups I've worked for especially well.

Layering

My one piece of advice on this subject is that when you are conveying complex information and you want to say something proscriptive, the best way to hit the goldilocks zone is to layer information.

By layering, I mean breaking down your message into chunk-able pieces. Start with the key statement, then move down the chain of propositions that underlie it until you get to the base premises that you believe your audience will understand (can you tell that I studied philosophy?)

Here's an example:

Statement:

For most startups, the number one metric to focus on at the earliest stages is retention.

L2 propositions

1. "Most startups" means startups that have a subscription model of some variety, that relies on collecting recurring revenue from the customer every month (often in a B2B context). Other startup models might vary.
2. By "earliest stages", we mean where a startup has launched to market but is pre product-market-fit.
3. The reason that a focus on retention makes the most sense is that...

The more complex the statement you want to make, the more levels of propositions you need to drop down while still keeping the overall top-level statement clear. For instance, I'd probably want to explain what I meant by "product market fit" to some audiences, but not to others.

This isn't necessarily a good way to actually write an article, paper, or have a conversation about a topic, but it is I feel a relatively neat way to think about whether you have hit the right level of detail while keeping things well-communicated.

Think I've failed to communicate this subject well enough? Feedback welcome to elliot@pkld.io