The nomenclature and procedures that define decentralization
I enjoyed the ideas presented in this paper by Professor Angela Walch. I was impressed by the professional manner in which the paper was written and the thorough examples provided by the author to support her case for the claim that certain blockchain networks, suggested by the SEC to be decentralized, are, in fact, not yet decentralized.
The author suggests that the things we can see, such as the geographic location of permissionless blockchain nodes, appear to be dispersed enough to call the Bitcoin blockchain and the Ethereum blockchain decentralized but that, in fact, there are many other unseen, unquantifiable and rarely discussed levers and controls operating behind the scenes of these permissionless blockchains that make them centralized power machines.
Professor Walch’s article provides the beginning of a staircase which enables the reader to rise above the issue of the word “decentralized.” Her paper helps pave the way for a better perspective and more complete understanding of the seperation between the words “centralized” and “decentralized”. It is of value for an author to lead readers up a staircase for a better view. But I would like to add another step in this Decentralized Language Story.
The author ends the paper by proposing we entirely eliminate the use of the world decentralized, for now, without providing an alternative.
To abandon the reader there, giving them no other alternatives, leaves them to fall off the idea cliff. It drops them back down where they came from, only more confused than they were before.
In my experience, the most helpful way to point out a problem is to also provide at least one alternative solution, ideally two or more, of some slightly improved direction we might use to move forward, paving the way for progress. I would like to contribute to the conversation in an effort to improve our language, I submit this post to create more clarity in our community. I hope it will contribute to improved language and help us more accurately articulate the meaning of the words centralized and decentralized as they apply to blockchain technology.
Allow me to point out that this post is in response to extremist language that exists in our community, centralized vs. decentralized, and proposes to invite more nuance into our language. This post is not a final solution, does not propose to be “the answer” nor do I want it to be. I’d like this post to simply be a signpost toward a more careful and observant path, one which nurtures the evolution of nuance in our language as we move forward into our fourth industrial revolution.
I have two proposals for a better path forward. The first is a suggestion to improve our terminology. The second is a suggestion to improve our procedures.
- Terminology suggestion. Our culture is changing. Our language needs to evolve with it. I suggest we create a more nuanced vocabulary to define centralized and decentralized. It’s time.
When one steps out of the darkness into the light it is understandable that one’s eyes only see light. When one steps back into the dark again one can only see dark. These are two easily defined terms; light and dark. But as one spends more time in either the light or the dark ones eyes adjust and begin to see edges, shapes, shadows and the nuances that exist but not observed before. This is where we stand now; in the light and the dark without the language to identify what we see.
Take snow for example. Even though there is some debate about whether or not the indigenous people of Alaska have fifty different words for snow (it was established in 2010 they do) this idea of giving specific names to different kinds of snow is what I’m aiming for here. In the same way that snow can be broken down into words such as aput, expressing SNOW ON THE GROUND; qana, meaning FALLING SNOW and piqsirpoq, signifying DRIFTING SNOW, decentralization should share a similar privilege of being identified with a wider variety of more accurate, descriptive words.
My proposal is to begin by breaking the two terms, centralized and decentralized, into three terms: centralized, semi-centralized and decentralized. From there I propose we break those three terms in nine identifiers. We can keep sub-dividing them for as long as you like. But here is how I see it looking to begin with.
Natively Centralized — A centralized company with no intention or practice of decentralization of any kind.
Median Centralized — A centralized company open to thought and discussion about the advantages of becoming less centralized. Importantly, this company has at least one experiment in place to test the notion of decentralization within their organization.
Shifting Centralized — A centralized company actively in the process of implementing some structure and procedures to become less centralized.
Early Semi-Centralized — An entity in the early stages of semi-centralization. This involves running more than one part of their structure or procedures in a decentralized way with plans, clearly articulated, for moving into the next phase; Median Semi-Centralization.
Median Semi-Centralized — An organization that is equally weighted centralized and de-centralized
Late Semi-Centralized — an entity that has more de-centralized practices than centralized practices.
Early Decentralized — A network that can be legally defined as centralized or does not have a central structure defined by a global government as a company structure.
Median Decentralized — A network that is mostly decentralized but at least one point of centralized power or practices can be identified.
Natively Decentralized — Started in a decentralized manner and remains decentralized to this moment. This is philosophical categorization. Is there any such thing as natively decentralized? Doesn’t everything begin somewhere?
2. Procedural solutions. Just because there is no perceived central authority does not mean we cannot incorporate agreed upon, human procedures. I propose we make moves within the community toward discussing acceptable procedures to safely implement bug fixes in permissionless blockchains with an aim to establishing a set of acceptable and expected procedures. Let’s simply bring the arguing and debating forward in time, before a bug is found, instead of waiting for disagreement and calls of foul play after a bug fix is complete. It is unjust to cry foul when we don’t even have an agreed upon procedure for those intelligent individuals who find the bugs to follow when exterminating the bug. We need emergency procedures in blockchain technology protocol development. If we can establish them in permissionless blockchains then perhaps there won’t be such an apoplectic response by the community when bugs are dealt with and it is perceived that the “right thing” has not been done even if the person or group making the discovery does their best to implement a fix while still maintaining the security of the network. Let’s talk about Acceptable Bug Fix Procedures.
I think there is more to explore here but those are two ways forward that I can identify this morning after having just read the brilliant paper by Angela Walch and the engaging outlining piece by MIT introducing Dr. Walch’s article.
I’d like to make it clear, these thoughts I have laid out would not have been possible without the profound intellect of Professor Angela Walch upon who’s shoulders I stand. I commend her for her keen observational powers and the discipline, fortitude and courage it must have taken for her to write, constructed and complete her brilliant paper. Brava Professor Walch!
Now I must re-focus my efforts back onto my little corner of blockchain technology; Proof of Location. Ciao for now.
