Phonograph and blockchain - in search of the perfect record

Profile picture for user Peter Coffee By Peter Coffee June 24, 2019
From phonograph to blockchain - there's a path there, says Salesforce's Peter Coffee.


In 1878, Thomas Edison’s early writings concerning his then-new invention—the phonograph—focused largely on the capture of spoken language – not so much on music, which the early technology was not well suited to record. “The phonograph will perfect the telephone,” Edison wrote—with that emphasis in the original—adding that the telephone was, at that time, “restricted in that it leaves no record of its transactions.

Were our telephone-conversation automatically recorded, it would be expressly resorted to as a means of perfect record…doing the work of a conscientious and infallible scribe.”

Edison could scarcely have imagined the disruptions, including the resignation of a U.S. president, that would someday pivot on the existence of such recordings.

Edison did not conjure up from scratch that desire for a “perfect record” – automatically produced in the course of our activities, and entirely trusted by all concerned. It’s clear, in his 1878 language, that this was something whose absence was already keenly felt, more than 140 years ago – “reinforced by the demands of an economy of time and money at every mile of increase of distance” between the parties to a transaction. Again, Edison could scarcely have imagined how much our demands for timely and economic action would intensify, even as the distances in question stretched from mere tens or hundreds to many thousands of miles.

We have before us today an alternative notion of Edison’s “perfect record”: the distributed ledger, often named by what used to be the special-case term of “blockchain.” These are commonly thought to be recent inventions, with Google Trends showing almost no search traffic for “distributed ledger” until 2014. The “blockchain” label emerged into popular use at about the same time, eclipsing almost immediately the more general term.

In actuality, “The earliest identified occurrences of the concept of a ‘blockchain’ can be traced back to Haber & Stornetta (1991) and Bayer et al. (1992), who introduced the notion of a chain of cryptographically-linked data blocks to efficiently and securely timestamp digital data in distributed systems using cryptographic hashing functions and Merkle trees,” notes a report produced last year by the Centre for Alternative Finance at the University of Cambridge. What? No bitcoin? Are we in the right movie theatre?

As the U. Cambridge report (sadly?) narrates, the data-management advantages of a cryptographically protected chain “attracted little attention, in contrast with recent enthusiasm around cryptocurrencies.” To say the least: if we replot the Google Trends graph for “blockchain” search activity on the same scale as “bitcoin,” the planet’s interest in the former looks like mere noise along the horizontal axis. This has been, for the past several years, a dismaying mis-allocation of attention.

Letting cryptocurrencies continue to suck the oxygen out of the room would be, today, as much an error as continued attention to early primitive games for the Apple II after 1979 – when the pioneering Visicalc spreadsheet became the principal reason for people to buy a desktop computer. “Thousands of businessmen walked into computer stores and said, ‘Sell me a copy of Visicalc and something to run it on,’” noted a 1984 retrospective in PC Magazine – and in much the same vein, a blockchain conference in May of this year drew comments on a pivot of emphasis toward far more serious, albeit unexciting, applications. “The focus seemed to be on continuing to build new, user-friendly decentralized applications and services that might be compelling for people who want an alternatives to the traditional financial system, and businesses hoping to take advantage of shared, cryptographically secure databases. There was much less talk of token sales. ‘The tourists are gone,’ I heard someone say as a way to explain why only half as many people were there.”

What's it for? 

On the cusp of readiness for everyday application, the whole family of blockchain data-management solutions is in a Cambrian explosion of rapidly evolving implementations and confederations. One does not wish to be a laggard, still cobbling together a pilot project while competitors and new-entrant challengers are already refining their production offerings. Neither, though, does one care to commit to early-stage technology at high risk of early-onset antiquity. “By 2021, 90% of current enterprise blockchain platform implementations will require replacement within 18 months to remain competitive, secure and avoid obsolescence,” warned Gartner Inc. at the beginning of this month.

When something is complex, rapidly evolving, and in need of spreading costs across a large number of emerging opportunities, that seems like the ideal breeding ground for cloud services. Blockchain-as-a-service—announced last month by Salesforce, and with other players like IBM already in the game—seems like a natural way to let organizations treat this as merely another model for data storage. The trade-off is one involving various parameters, including volume, volatility, diversity of data-sharing partners, and value of an intermediary-free solution, to decide between a blockchain-family model and other more familiar options.

As is often the case with innovations, in any century, the choice has at least as much to do with the business model as with the enabling technology. To return to the example of Thomas Edison’s first phonograph, another decade would pass before he would develop a more rugged recording medium—the wax cylinder, supplanting his initial use of fragile metal foil—and thus pave the way for mass production and distribution of recorded music. Further, it would be more than another decade still before the emergence of recognizable disc-format phonograph records.

Perversely, what was thought to be the distinctive innovation of Edison’s phonograph—its ability to record, and then play back, with a single device—would quickly become the province of magnetic tape, and then of digital equipment. The phonograph, as Edison first conceived it, would evolve instead into almost solely a playback device – for a medium whose economics would change the entire nature of music as an industry. Until subsequent technologies, with other economics, changed that industry again.

We see a comparable pivot, from what the technology can do to why a customer should care, in the news this month of a pilot project for pharmaceutical product tracking using a blockchain platform. As noted by one of the project leaders in an interview elsewhere, “one of the main items in the pilot is a mobile app, which is designed to scan a QR code at the time of prescription pick up. Individuals with the app will be able to scan a product’s QR code to see the provenance, such as the manufacturing site, how long it has been on shelves at participating stores and other useful information.” It truly is more about the customer benefit, than about the operational convenience.

Blockchain today, like sound-recording technology over a hundred years ago, is just getting started. Bitcoin probably has as little to do with blockchain’s future as Thomas Edison’s first recording of “Mary had a little lamb” was a predictor of Apple’s iTunes (may it rest in peace). What Edison (should have) taught us in the 1800s—and what blockchain services teach us today—is that until there is adoption, invention does not become innovation (to paraphrase Peter Drucker). Cloud services, in general, accelerate opportunities to discover the value cases that tell an inventor “See? This is what it’s for.”