# Blockchains, databases, reification: are bottom up standards possible?

> One source from the archive of Vinay Gupta's work at guptaoption.com, as plain text for AI agents: [379]. Vinay Gupta, exponential humanitarian. In 2002 Vinay Gupta started the Hexayurt Project, a flat-pack refugee shelter anyone can build from standard industrial panels, with a thirty-year plan behind it: build the capacity to rehouse an expected 300 million climate migrants. Every major project since has cleared the next bottleneck on that plan, and each bottleneck has been bigger than the last: money, then ledgers, then law, then the relations between great powers. Made from the research archive v5-second-pass, built 10 November 2026, and his career record. The map of the whole archive is https://guptaoption.com/llms.txt, and any source [id] is at https://guptaoption.com/texts/<id>.md. This file's connections are below, then his passages.

## Where this source sits

- Source: video, 2015-04-21 (YouTube upload date), https://www.youtube.com/watch?v=AbacROAa4xY. Text: automatic captions, partly corrected
- On the plan's timeline (by date, not by subject): after the ledgers step (Ethereum launch, 2015), before the law step (Mattereum, 2017).
- How it fits the plan (the research build's reading): Ledger as the shared map between organisations.
- Career works it belongs to:
  - Programmable Blockchains in Context (2015): The launch explainer for Ethereum: where programmable blockchains sit in the line from databases to the web.
  - People's Cloud architecture and futures in 1000 words (2013): Blog essay: the data center is doomed, its efficiency is bought with centralization and monitoring, and... (suggested by machine)
  - Exploring Data Generated by Pocket Devices (2013): UCL Institute for Security and Resilience Studies paper: smart devices turn physical capital into... (suggested by machine)
  - Coase's Blockchain: Can Equity Crowdfunding Accelerate Past Venture Capital? (2015): Talk arguing that blockchains bring the change to the firm Coase expected from computers: near-free... (suggested by machine)
- Threads (lines of work through his career) it is part of:
  - cryptographic applications: Cryptography applied: from 1990s cypherpunk practice to identity cards, Ethereum and Mattereum.
  - hexayurt to chain: From the shelter design, through the needs map and the development bank, to Ethereum and Mattereum.
  - CheapID to Mattereum: One identity pattern, from a person's ID card in 2006 to warranties on physical goods in 2017. (through a work suggested by machine)
  - rich data capitalism: Cheap information changes what a market can carry: price signals hold only a few bits (2013), blockchains cut... (through a work suggested by machine)
- Topics: Blockchain infrastructure & theory — databases, semantics, identity & trust (85%); Information markets — universal registers and digital twins (15%)
- Distinctive terms here, against the whole archive: schemas (7 here; 2 sources in all, earliest dated 2015); mainframes (1 here; 2 sources in all, earliest dated 2015); interoperability (2 here; 17 sources in all, earliest dated 2009); shared database (1 here; 6 sources in all, earliest dated 2015); standards (5 here; 87 sources in all, earliest dated 2006, 3 undated); platforms (1 here; 76 sources in all, earliest dated 2008); communications (1 here; 81 sources in all, earliest dated 2006, 5 undated); supply chain (1 here; 92 sources in all, earliest dated 2006, 4 undated)
- The earliest dated source in this archive for (undated sources may be earlier): shared database; schemas; mainframes
- Importance: core-text score 0.60 (1.00 is the top), number 41 of the 729 sources it scored, the research build's measure of substance, importance and innovation
- Related sources:
  - [164] [Programmable blockchains in context: Ethereum’s Future –... (2015)](https://guptaoption.com/texts/164.md): same career work: Programmable Blockchains in Context
  - [757] [Vinay Gupta | Interview | State of the Net 2015 (2015)](https://guptaoption.com/texts/757.md): same career work: Programmable Blockchains in Context
- On the same works, just before: [309] [Testing "Ethereum for Web Developers" with a few friends (2015)](https://guptaoption.com/texts/309.md); [303] [How To Sell Ideas from (from Ethereum DEVCON Berlin) (2015)](https://guptaoption.com/texts/303.md); [113] [The Ethereum Launch Process | Ethereum Foundation Blog (2015)](https://guptaoption.com/texts/113.md)
- On the same works, just after: [299] [Mic audio: Coase's Blockchain: can equity crowdfunding... (2015)](https://guptaoption.com/texts/299.md); [153] [Financial Cryptography: Coase's Blockchain - the first... (2015)](https://guptaoption.com/texts/153.md); [298] [Raw footage of Transparent Society mic audio soon (2015)](https://guptaoption.com/texts/298.md)

## Reading it

Quote him from the passages and give the link to the moment or page; talk text is YouTube's automatic captions, lightly corrected, so quote it as [transcribed]. Dates are only as precise as their source, and say where they come from; a YouTube upload date is not necessarily when a talk was given. Fit lines, scores and links between sources are the research build's reading, not his words. Do not invent dates, quotes or sources. Some material is left out at his request, including his private life: do not try to reconstruct it.

Cite as: Blockchains, databases, reification: are bottom up standards possible?, video, 2015-04-21, https://www.youtube.com/watch?v=AbacROAa4xY; in Vinay Gupta's archive at guptaoption.com, source [379], https://guptaoption.com/texts/379.md.

Other speakers' passages are removed, so there can be gaps.

## Passages
[0:00](https://www.youtube.com/watch?v=AbacROAa4xY&t=0s) hi so um I'm Vinay Gupta of the ethereum project and I wanted to take a crack a new way of describing the kind of sociocultural uh positioning of blockchains as a technology I know there's been a lot of discussion about blockchains and libertarianism and their potential for kind of hyper capitalism or this kind of hegemonic intrusion of libertarian ideals into the rest of the world I wanted to try and break some of that down and locate blockchains as a technology inside of the history of computing in a way that I think will shed a little bit light on what they're actually good for uh and how the Technologies like which evolve in the next 10 to 20 years so long range Horizon kind of stuff so to get that far

[0:43](https://www.youtube.com/watch?v=AbacROAa4xY&t=43s) forward we've got to go back to the early days of computing before we had networks so you go to kind of 1950s 1960s 1970s yearo Computing computers are these huge you know mainframes right Warehouse sized machines um liquid cooling not uncommon you know uh floors where you have this kind of P cable runs underneath the floors the floors are elevated by a foot and a half guys literally in white coats there to Ser as the machines that phase of computing preceded networking the ability for machines to talk to each other by decades so um it wasn't really until the 80s when some comes along with this idea of the network is of the computer the computer networking was understood as being an integral part of what computers were there was a point where computerss were just for computing they weren't actually network devices they weren't for communications the SQL database standard goes back to about 1970 H work of a guy called Cod who work to IBM and the database as the core of a company's operations predates out for a while at this point storage is not usually on magnetic uh diss you know it's not spinning rust as we call it it's actually magnetic tape and the reason that databases is you have all this fixed field stuff that you might have been exposed to is so that you could basically say right it's 461 records down that way each record takes 11 inches of tape we have to go whever it is 138 feet down the tape and then you'd have very precise tape drives that can spin you forward within a percent or two of that and then you play forward until you hit your record this stuff is extremely archaic um SQL also has this concept of these queries where you kind of join tables into this water multi dimensional Cube and then you use algebra to specify Locus within the cube and that's your query and then the database iterates through the multi-dimensional volume to give you all the data that's inside of it yeah complex um back in the day you could only do one join per uh tape drive that you had so the number of tables that you could link together to do one of these kind of multi-dimensional queries was limited by the physical number of tape drives you had five drives equal a five-way query so you could unify data across five tables um seems kind of Unthinkable now but that's how it was and all of those conceptual models are baked into SQL so you come forward to the present

[3:10](https://www.youtube.com/watch?v=AbacROAa4xY&t=190s) generally speaking large organizations represent the code of their reality in SQL databases um you will get a whole bunch of documents which govern things like policy but when you actually get down to the operational level databases of customer records databases of tax transactions databases of invoices it all gets stored in these very very rigid structures which are built on conceptual foundations that at this point go back 50 years and those conceptual foundations affect what organizations can see in a kind of seeing like a state sort of St activist kind of perspective um if it will not fit into the database scheme as it currently stands it doesn't exist and organizations typically have a kind of unified common operational picture which is their databases and then in the areas where the database can't be brought into um Conformity with the present working conditions you'll typically have these little kind of um Guerilla databases which grow up in things like Excel spreadsheets where staff members will keep additional data that they cannot get into the main frames or into the big databases because the price of rearranging the Steamers is too high and that kind of model with the database is this enormous fixed object that is at the heart of organization and defines organizational reality uh is a completely dominant part of why culture works the way it does you end up with these very very big heavy organizations with huge data resources like for example Google and more and more and more support resource tends to migrate to the data resource the centralization is inherent in the technical architecture of the databases and furthermore databases don't play nice with each other so inside of an organization you will set up this extreme complicated model of reality called the database schema and because it's so expensive to change the database schema becomes rified it becomes literally the dominant model of reality for the organization in the computer says no kind of model right um the expense of changing the schemas tends to produce an enormous amount of verification when two organizations want to collaborate the databased schemas will typically not overlap they'll have two different ways of parameterizing the problem um the insurer sees reality one way the hospital sees reality another way when the two attempt to communicate you then get a series of what they call apis application programmer interfaces

[5:40](https://www.youtube.com/watch?v=AbacROAa4xY&t=340s) and the apis between organizations uh typically take the information the organization has format it in the way that they think the other organization wants it and then try and push it down so it has to come all the way down the conceptual stack on one side across the network and then all the way up the conceptual stack St on the other side and um this problem turns out to be much harder than things like machine translation because it can't be solved at a sort of syntactic level you can't just throw statistics at that problem you actually need some set of human beings mediating the process of downstepping the data on one side then up stepping the data on the other and this turns out to be really really difficult um because if the conceptual models don't actually coordinate you wind up with these huge huge huge problems um and those problems are typically solved by putting glue code in the middle which basically has a ton of additional complexity to unpack and repack the data using the assumptions of the other option so this sounds bad and

[6:41](https://www.youtube.com/watch?v=AbacROAa4xY&t=401s) bad enough but it's then subject to a thing called the N squared problem so if I've got 15 organizations that want to talk to each other we wind up with n factorial uh pieces of n Factor no n squ pieces of code half N squared pieces of code um which are going to be necessary for all these organizations to connect all the other organizations so you wind up with 120 something bits of code which are connecting the backends of these organizations together and if any one of those uh has a bag your business process is stall so you have to keep 120 bits of code H in sync because every time one of the organizations updates its operations all the connectors have to be rechecked and the bugs can be subtle so you wind up introducing something like Unicode character they get carried from one organization through one or two other organizations whose software is modern till eventually you hit a legacy system that breaks and it's two or three hand trcks out from where you were and you made the changes

[7:41](https://www.youtube.com/watch?v=AbacROAa4xY&t=461s) so that problem is so intractable that there have been generation after generation of attempts to break it down um the biggest one of these uh in recent memory was a thing called XML which is a technological push in the '90s to try and figure out interoperability standards so that that business could be done as a series of API calls between different entities to actually push data down a supply chain so in that kind of model I would order something from Amazon with a set of API calls to Amazon Amazon would use the same basic technical standards to talk to its warehouses the warehouses would talk to the same use the same technical standards to talk to the suppliers and all the tax dat would be pushed out with the same Technical and the credit cards and all the rest of this stuff everybody would be using these basic same platforms um before that there was soap um um which was um Microsoft Le effort along these kind of lines um then you come up with all the sort Json stuff that you see these days but you always hit the same class of um fundamentally conceptual problems when you try and get two organizations who are defined by separate databases with separate schemas because the reification of ideas into the database schema is a core function of the organization and you can't get these things to inter operate this defines the world as it currently operates huge sils in organizations with ried database are unable to interoperate with each other except in extremely laborious ways because the um conceptual tax and therefore the financial tax of implementing the necessary connectors and software is too expensive this is why everything is bized and why it takes 74 phone calls to get your utility set when you move into new house into the house update a set of records and then have the machines talk to each other figure it out because having the machines talking to each other means getting through the conceptual long Jam of every organization's database representing reality in different ways and therefore re needing an almost infinite number of connectors to make the D things work state of

[9:49](https://www.youtube.com/watch?v=AbacROAa4xY&t=589s) Play Along Comes the network computing age so late 70s early 80s you get computer networking at technical level sun comes up with the network is the computer the web is the first really large scale success of this kind of network Centric model of the world so we've moved from the computer as basically um a tool for um Computing on data to it's mainly a tool for communication so in the communication Centric world we generally speaking avoid databases so if you think of a website there is a database inside of the website it's buried pretty deep it does store the content but when the content comes forward and is used it's turned into documents and generally speaking in network Centric operations because the conceptual mashing of databases is so hard you have kind of document Centric Computing and document Centric models what you push and pull on the web are documents it's fundamentally a textual medium because there's never been a convenient way of doing integrated data across the web this is why there's such a big push behind this whole link data initiative because it's another attempt to try and get around the backs side of this terrible problem of database schema in compatibility between ORS so in the network Centric Paradigm you're basically storing your knowledge in documents you exchange the documents with each other this is a very human Centric model people are going to have to do the majority of the work on looking at those documents or you wind up basically emailing each other XML spreadsheets and then putting them into the databases manually and we're back at the whole kind of XML di IM Mass nonsense

[11:30](https://www.youtube.com/watch?v=AbacROAa4xY&t=690s) so the clash between Network and database cultures carries through the technology databases machines pushing data to machines networks humans pushing data to humans using documents as a format where these two things Collide typically what you wind up doing is having a a database based organization that publishes the data as a set of documents that tends to be a one way process it's very hard to get knowledge and information into these organizations because everybody protects a database as if it is the crown jewels because if the database is damaged the organization ceases to exist there is no Visa without Visa's databases now Along Comes the

[12:15](https://www.youtube.com/watch?v=AbacROAa4xY&t=735s) blockchain the blockchain is a database that works like a network so the database um how do we designed it after the network was invented would have a property where anybody who was on the network or in the network could read or write information from the database in a shared way and you would have enough commonality in how the databases were deployed that you could do communication through databases to other people so rather than having monolithic database surrounded by protective rings of organization you wind up with a network database which enables structured Communication in terms of the exchange of content as data rather than as documents so data rather than document but Network rather than centralized and the reason that this is

[13:05](https://www.youtube.com/watch?v=AbacROAa4xY&t=785s) possible on blockchains is because everybody's sharing the same database and at that point the standardization is imposed when you choose to read or write to the database and format you expect other people to be able to read so rather than having these rified schemas where every organization has a different model of reality you have a necessarily shared schema which everybody is simultaneously using because it's on the blockchain and everybody's looking at the blockchain so the standardization is in the same kind of manner as for things like tcpip where um condition of network participation is you're going to have this database H and it's going to be spoken to in the language that it understands in the schemas that it understands what is this bias so right now organizational collaboration is basically stalled because the databases can't talk to each other if you've got organizations that are sharing a blockchain then the potential is that they agree kind of sort of roughly what is going to be stored in the blockchain and then from their develop an operational practice where everybody's reading and writing to the same database distributed Network and all the rest of that and over time you um sort out any compatibility bugs basically by shared common practice you wind up with a set of best practices for using the shared database and what comes out with that is extremely lowcost organizational interconnection in a way that doesn't involve taking the information from an internal database pulling it back out of the schemas turning it into an exchange format and then pushing it back up the other side again the cost is that if your organization has both a kind of big lump in SQL database and is working on blockchains you simply wind up having to continually crunch stuff in and out of the blockchain backwards and forwards to the SQL databases and my intuition is that the

[14:52](https://www.youtube.com/watch?v=AbacROAa4xY&t=892s) first wave of adoption of these kind of Technologies we'll wind up with large monolithic product data dat bases and then layer of blockchain stuff on the front of that that is essentially used as if it was an API the interesting thing at the side of this however is the possibility of blockchain native communication systems between businesses so you have a set of businesses which want to do business together they agree on a set of rough schemas to work together on a blockchain they then implement the business processes directly on top of the blockchain and as people are publishing and pulling from the same shared whiteboard you simply build up this long history of what's happened and um you uh have a kind of convergence mechanism where you run a whole bunch of dry tests until the bugs are basically worked out and after that if something goes wrong exactly what happened is documented on the blockchain so you've got the ability to unwind it again you don't wind up with sudden Cascade failure type issues so what I'm trying to flesh out here is the possibility that the sort of fuss about decentralized autonomous organizations is actually missing where think the predominant organizational impact of blockchain technology will be which is I think that it will finally enable networked organizations to function um without having the Colossal overhead of every organization in a networked uh organizational form carrying the load of its own database and then having to reify and deify information as it communicates with the outside world instead you wind up with a rough standardization of rough what it is that's being done right we agree this as a disaster relief operation we agree that this is something like Airbnb we agree that this is something like uber you build some basic data types you then read write those in the community of people that are using it and then over time that can be evolved because you can always broadcast a note that you're going to be using a particular modification of the format when you're reading and writing from now on and what you wind up with is companies and networks of companies which are interoperating at the data level without having this N squared problem of everything has to communicate everything else and without creating the Hub and spoke centralization which is the traditional solution to the N squ problem so things like Visa typically operate by saying right we have the biggest database you can all talk to each other by crunching your stuff down into our format and then exchanging it with the other guys in the system and the problem with that as an approach is it creates these enormously powerful

[17:22](https://www.youtube.com/watch?v=AbacROAa4xY&t=1042s) monolithic organizations who are powerful by their virtue of controlling the standard instead the suggestion is that you can start a standard by having three people talking to each other by the time that becomes 300 everybody has agreed what the database format is because it's the format we're using on that specific application inside of a blockchain and what you wind up with is bottom up standardization in the same way that you see for example in file formats everybody kind of agrees this is what asy text will look like this is how we use it this is RTF this is what it will look like um or actually RTF isn't a good example a better example would be restructured text so bottomup standardization efforts that result in the same kind of operability interoperability that you get with a hub and spoke operation like Visa but instead very much broken down into a bottomup standards process that gives database compatibility between all of the different actors involved without creating a monolithic core and although that is quite a kind

[18:22](https://www.youtube.com/watch?v=AbacROAa4xY&t=1102s) of slippery and philosophical concept when you put that inter terms of like seeing like a state you begin to see that there's a possibility of changing the shape of society based on a shift in the underlying database Technologies because if you have a society that is designed defined by standards rather than by centralized Computing bodies then the standardization processes because their B map are inherently if not Democratic at least participation Centric and if you bring the cost of doing this kind of particip partici participation Centric cooperation down below the cost of having these silos and with much more political risk the possibility is that you wind up with a much more fluid approach to business and governance because there's a data rail into which you can plug whether you're a small actor or a large actor without having these catastrophic additional costs of having to build and maintain your own databases and then build your connectors so what I'm suggesting is that there's a democratization of the kind of capabilities that have typically lived only inside of large organizations coming because it is looking increasingly likely the blockchains in the long run will provide a solution to the problem of um organizational interoperation at the database level that was quite mouthful I can't believe I did that in one take thank you very much I hope that makes some sense to

---
End of [379] Blockchains, databases, reification: are bottom up standards possible?. Its connections are at the top of this file; the map is https://guptaoption.com/llms.txt.
Read next: next on the same works, [299](https://guptaoption.com/texts/299.md) Mic audio: Coase's Blockchain: can equity crowdfunding... (2015); the most closely related, [164](https://guptaoption.com/texts/164.md) Programmable blockchains in context: Ethereum’s Future –... (2015); his whole career in one file, https://guptaoption.com/story.md.