You are reading content from Scuttlebutt
@dan %8FQ3NkuF6QS358f4vVQxJ8uKPeE2EJUnsOOPSOkC670=.sha256

THE WORLD OF DECENTRALIZED MONEY: A STORY IN THREE PARTS

by: JAYA KLARA BREKKE

from: Moneylab Reader: Overcoming the Hype

Part 2: A tribute to those who will never be peers

Bitcoin github repo, August 1st, 2017

This day in Bitcoin-landia had been marked as the culmination and possible final resolution of a conflict that had been raging for some three to four years. The conflict was about an existing hard limit on the data-size of blocks on the Bitcoin blockchain, set to 1MB — whether and how it should be increased. And today was the day the protocol was going to be hard-and-soft forked into several versions for how to move forward on this question and the development of the project overall.

Other authors have written excellent informative pieces about the politicized nature of what, to some, might seem an obscure technical question and the ways in which it in fact reflects very different understandings of decentralization, power and governance. The story I want to address here is slightly more meta . Here is what I think is really at stake in the Bitcoin scaling conflict: The possible shaping of new types of subjectivity.

Bitcoin was first introduced in 2008 as a proposal for a peer-to-peer electronic money system. In this enticing combination of network technology, cryptography and a systems-architecture-that-pays-for-its-own-security-through-some-basic-market-logics the idea was that we could get rid of the need for authorities, like banks, to guarantee relations and trust between people who don’t know each other, replacing such trust with cryptographic proof, and thereby run global money systems through networks of peers.

The experiment has since grown immensely — in numbers of people, places, organizations and companies involved, and value moved through the system. It is now facing a problem of scaling. If you read through crypto-currency news sites, Twitter, threads on Reddit and commentary on Medium, the conflict over scaling was brought on by a concern that the increasing number of transactions were clogging the network because of the hard limit on size of blocks of transactions that were being verified. Some therefore argue the hard limit needs to be increased, while others are wary of such a development as it might centralize aspects of the network and have therefore developed other solutions.

The point is that there are different versions of what decentralization and scaling means and what is important in terms of use-cases and features and the future of the system. While some want to out-compete existing payment systems (Pay-pal, Visa, Mastercard), possibly risking centralization of aspects of the network in the process, others want to hold out on such ambitions in order to keep moving towards something entirely different, a vision, perhaps, of Bitcoin more as a vast and still open-ended scientific experiment than (or in addition to) a new global payment system.

Let’s look closer at the word ‘scale’. Scale has many different meanings: increase in quantity, size, volume and geographical spread. I would like to introduce another vector to the concept of scale — differentiation. As the Bitcoin network grows, the fact that you have increasingly differentiated users, uses, and roles means that the nature of the network changes, as do questions of power.

@dan %WPa73v8fl0lHa39+KK4cb78iGT0WfX31Su8iO7F7jZ0=.sha256

Here’s where I am going with this: for a system to be peer-to-peer you need to have a network of, well, peers. The assumptions that are usually associated with peer-to-peer, such as horizontality, decentralized power, neutrality, equality and openness can only be considered true if those involved indeed are peers. As more and more people get invested, and are interacting with and developing different aspects of the system, these applications, interactions and people will be different, have different capabilities, experiences, abilities and needs (and hey, that do not necessarily run a full client, contribute to or fork the code base.)

This is a tribute to all those who are never going to (be able to) be peers. They might be busy doing other useful things. ‘Peer’ is one of those curious words that (potentially) means both a person and a node in a technical network. In Bitcoin, a peer would be miners and full nodes that verify and secure the network, but it also alludes to these people who contribute, are developers and researchers and so on. There is indeed an incredibly active and lively network of actual peers that maintain, experiment with and develop Bitcoin. But I would dare say that most interactions with the infrastructure are no longer by miners nor full nodes, but by people transacting using exchanges and wallets that they probably don’t even hold the keys to.

The scaling conflict is not only about increasing block sizes on the blockchain, it is also about Bitcoin having reached such a scale that it can no longer be thought of as composed by peers.

The conflict and now quite high stakes of this infrastructure raises questions of accountability in the decisions being made — by developers, by miners and full nodes (see #UASF). Some aspects of the infrastructure provide an outstanding basis for this: ‘commits’ and contributions to the code are transparent, the developer’s mailing list is all open and there is an impressive amount of glossaries and guides for those who want to look into it. But to benefit from such openness requires not an insignificant level of literacy, time and interest, and the big rifts and murky rumors that have characterized the scaling conflict are not easy to comprehend. Many people are left having to trust the word of some over the word of another (an ironic condition for what is supposed to be a trustless system).

It might be fair to say that a lot of non-peer interactions are by users that we might want to call ‘DIY financial speculators’, who should take full responsibility for any risk they get themselves involved in. The project started in a sense as a giddy experiment in which the scientists, cryptographers and hackers involved were only really accountable to themselves and their own curiosity and tinkering. And certainly there is an element of this still, the project is as of yet a vast collaborative effort of enthusiasts. Why should they bother about such people?

If the infrastructure is indeed to grow and have use beyond speculation while still staying true to non-state, non-corporate sentiments, then it might be the time to recognize and consider new subjectivities that would form in the process, and the fact that not all of them are going to be peers. So who do we consider to be part of this new constituency and how do we take care of them? Are they ‘users’ or something else? If a state is supposedly accountable to ‘citizens’, a corporation is accountable to ‘customers’ or ‘clients’ and platforms are accountable to ‘users’, who are peer-to-peer networks accountable to? When peer-to-peer networks scale how do they deal with
differentiation and the emergence of non-peers who are dependent on the system? From a less conspiratorial perspective, regulation might be considered a response to this condition in order to ensure accountability. But surely it must be possible to work out some more interesting ways to go about accountability and taking care of differences and differential capacities that feel better than the-long-arm-of-the-law, or that do not entail black holes of tedious reporting, nor automated village granny reputation systems...

Unless this is worked out, and until then, this project of ‘disintermediation’ for those who are not peers is actually a project of reintermediation — simply swapping one set of intermediaries (the banks, politicians and legal system) for another (developers, computer scientists and network technology), or, even worse, adding another layer of intermediation and complexity.

@dan %0j+6EYEK/cl5g0V+PAy4ezLgTk4Ik6nDcOxo53SGIbs=.sha256

Check the original version of the postcard for references and links.

This is the paragraph which has really been animating my explorations in the space and in turn work with #mmt

This is a tribute to all those who are never going to (be able to) be peers. They might be busy doing other useful things. ‘Peer’ is one of those curious words that (potentially) means both a person and a node in a technical network. In Bitcoin, a peer would be miners and full nodes that verify and secure the network, but it also alludes to these people who contribute, are developers and researchers and so on. There is indeed an incredibly active and lively network of actual peers that maintain, experiment with and develop Bitcoin. But I would dare say that most interactions with the infrastructure are no longer by miners nor full nodes, but by people transacting using exchanges and wallets that they probably don’t even hold the keys to.

@dan %WHFsccCxaCqktpFRKQk1EaOtUoXHp2tZV3MkmrNfQFU=.sha256

Really this is one of the key kernels which informed the genesis of the proto-project (which I expanded on in much more detail 6 months ago) which has lead into #mmt . Very simply the question was "what would it take to get my friends involved in crypto" or "how can my friends become peers" or "how can I teach my friends to manage secrets" or "how can I teach my friends how to use a password manager without making them fall asleep".

There was a longer form reflection here.

With the grace and perspective of retrospect I am of the opinion that it was a mistake to focus so much on speculation rather than the simple (or not so simple) affordances of groups being able to securely manage passwords, secrets and private keys. Indeed this a major stream of our research focus within #mmt which can be seen in Socially Robust #interdependent Backups and Social Backups.

My thesis is that this technologies are not socially deterministic. That it is possible to hack and queer technologies built from a libertarian/propertarian mindset, to wrap them in different social logics and to transmute their purposes with an "alien logic" - one more commons oriented.

@dan %5ucsTzEIu+1kqe2ealGyrlY1dkWkdZzw17gHiiISnnc=.sha256

cc: @jamie_k my last comment here is a follow on from the podcast we did together - there has been a pivot in my thinking around this stuff. Or - more of a punctuation mark rather than a pivot :smiley:

:rainbow: :full_moon:

@dan %G9R1fooHRxVSBS8dbJgHU1mQLZyvqInoZb7DUVlHlhI=.sha256

Referring to this podcast

@dan %VCOY76j6MNVgcWWsFBCEHTqQr5JQW5acW0YQCFkpEw0=.sha256

cc: @Jayapapaya (author of these postcards)

@dan %WrD/ip+kM7mfXSNZgnYwtwjf0E17Z4HlGVJobjgbjLY=.sha256

cc: @Leedscath reflection on the reflection you read (my comment :) )

Join Scuttlebutt now