Hm, there's an argument I haven't seen before- what costs are those? At first I thought you meant transaction costs- but it seems like you also mean the cost of accountants? Why aren't accountants required if a company uses crypto instead of, say, USD? I've tried to think of examples but can't figure it out- mind explaining a bit more about what you mean? Sorry if I'm missing something obvious.

My last job was on the billing team at a payments company, so my perspective is colored almost entirely by that (somewhat traumatic) experience. To be fair, it may not be as representative of business in general as I think.

We had an app store, so maybe a merchant would pay $5 monthly for an app that made their cash register double as a time clock. It was my team's job to slice that pie.

We keep $1, the bank that sold them the cash register gets $1, and the app developer gets $3. Or something.

But the merchants and the developers all live in various countries, so the "billing solutions" that we had to partner with in order to ensure that everybody gets their slice of the pie couldn't just accept "Give Joe $1" they had to know what the money was for so that they could justify the fees they were extracting from this pipe of money.

The smart way to do this would be to get all stakeholders to agree on a few hundred lines of code that would handle whatever regulatory needs they were on the hook for. That code would go in a smart contract and then we could use a blockchain as a single source of truth for the data.

Instead we had four companies all with their own separate, private databases, all attempting to reach consensus about what was owed to whom and why--usually based on out of date information from various channels.

What we needed were distributed systems engineers, and what we got was four stacks of middlemen and a directive to multiplex money between them. If they add a fifth stack of middlemen, the codebase will explode.

At least for me, once I saw one egregious inefficiency that would be solved by financial consensus algorithms I started seeing them everywhere.

Take wireless service for example. Why do we have several separate overlapping cell networks? Wouldn't it be better to have one big network and just pay the nearest network operator per-kilobyte? The only reason that doesn't work is that such a volume of USD payments would constitute more in fees than it did in payments.

----

Sorry for the rant. The point is that the way we do accounting in USD is so saturated by middlemen that it takes weeks for them to agree and allow money to move anywhere. During those weeks, the accounting complexity continues to increase.

Since our money moves so slowly, each company has to build at least two billing systems--one to handle the debts as they occur and another to move the money (usually monthly) to resolve the debts. But if the money can move as fast as the debts accrue, then you only need to build one billing system, and you don't have to pay an army of middlemen to badly approximate consensus.

It's not about the token itself, it's about the practices in place around that token. The people who occupy privileged positions in the ebb and flow of USD haven't had reason to change their ways for a very long time. The alternatives, on the other hand, are evolving rapidly.

Thanks for the response- and no need to apologise for the rant, I liked reading it. Now I get what you mean.

I sort of feel like this is just a nitpick, since it was one big rant, but re: the cell network thing- we already have a mechanism where you can use data (and be billed for usage) on networks you don't subscribe to: roaming. The problem is the network operators don't really have much reason to allow other phones to roam on their network for cheap- they're selling data plans, and competing against others selling data plans. If a Network A customer can roam on Network B towers for not much more than the cost of data on Network A, Network B can't say "hey, customer, come switch to us, we have better coverage". So roaming costs are rather nasty.

This doesn't have to be the case, of course! They could certainly provide roaming at cost instead, so the user pays one flat price, almost like mutual peering in internet exchanges. The point is, even if there was a zero-fee, zero-latency micropayments system, this wouldn't really affect the dynamics of the market for mobile data- it's neither necessary or sufficient for the ideal world where a cell tower is a dumb RF-to-packets gateway that any device can use for a reasonable per-kilobyte fee.

But again, that's just one paragraph. I really appreciate the insight into some of the problems you had to deal with. And I can definitely see how - even regardless of the merits of a token itself! - the disruption it brings could cut out a lot of middle-men who are there just because they've sort of always been there. I just hope the practices that come with switching to a token don't have even worse downsides.

I.e. "Hooray! Now my code can control the flow of money, authoritatively, and nobody can override it!.... Oh, shit. My code controls the flow of money, authoritatively, and nobody can override it..."

Reminded of those AWS billing horror stories- when you can do expensive things with just a few lines of code, you can do expensive things with just a few lines of code.

Good points about wireless. You're definitely right in principle, the problem is "solved" . I guess we'll just have to wait and see if the micropayments approach ends up being more efficient in practice.

The reason I think it will is that in a world with _n_ wireless providers that are set up to provide handsets, sim cards, marketing, account management, and billing, and _m_ network operators that only handle things like towers and cables, you need n*m roaming relationships and many many employees to keep them all straight.

I used to work for one of the first sort, and would occasionally get calls from customers in Minnesota who lived so close to the border that they accidentally connected to Canadian towers from time to time. We'd go through their bill and provide a credit to offset the charge since it wasn't fair to force them to pay for international roaming unless they truly traveled to that country--even if their service did.

It's that sort of retrospective debt juggling (and the overhead of maintaining custom infrastructure for it) that goes away if you just use an open source protocol to pay at the time of services rendered.

One thing I can't quite put my finger on is coming up with a quantitative answer to how high the cost of doing business in USD is.

Even if you're not a fan of these examples, we can probably agree that somewhere out there is a company that has inserted itself between supply and demand (maybe it did something helpful once), and by doing so slowed things down, and now it only justifies it's existence by mitigating the problems caused by it's inclusion in the system in the first place.

Sure, this usually involves fees of some sort, but it seems to me that the real cost isn't in dollars lost to transaction fees but instead in an overall increase in... Financial friction? Dollars and cents are a fiction, but there are real problems out there that need to be solved and we've accumulated so many reasons to resign ourselves to solving half of them and at half pace. Not sure how to put a number on that.

> I just hope the practices that come with switching to a token don't have even worse downsides.

I like the sentiment, but I'm gonna nitpick this statement for a moment.

We're already using a token, it's the USD. It has a fragmented ecosystem of competing API's (debit, credit, ach, wire, cash) most of which require you to pay for access to and none of which rely on modern cryptographic techniques to protect privacy and prevent fraud.

There certainly will be downsides to switching to any of the cryptocurrency tokens available today, but the thing that they all have that the dollar doesn't is an open API. The friction of switching from USD to BTC is high, but the friction of switching to from BTC to ADA or IOTA (or whatever token provides the right set of features for whatever the application is) is much lower.

It's like going from cable TV, where you have to deal with what they give you, to web-based content, where netflix and amazon and hulu and disney compete to be the one that makes you happiest.

I'm not arguing for one token over another, I'm arguing for competition among them. Who knows? Maybe if the USD has to compete for it's position it'll shape up.

And you're definitely right about the terrifying responsibility that a world without middleman-arbiters is going to face. We've got a lot of learning to do re: how to responsibly handle those sorts of problems, but it really looks like the alternative to accepting that challenge is stagnation and parasitism.