(Bloomberg Opinion) -- The market for cryptocurrencies has gotten off to a lackluster start this year, with the Bloomberg Galaxy Crypto Index up just 3.50 percent through Monday, versus 11.5 percent for the S&P 500 Index. Four scheduled events in the rest of 2019 should help determine what comes next for digital currencies.

There are no shortage of naysayers doubting the future of Bitcoin and its peers, but transformational technologies never have smooths path to adoption. Mass acceptance of automobiles required investment in roads, service stations, assembly lines and oil drilling; and millions of individuals to pay for cars and learn to drive them; and legal and social changes. Ditto for telephones, personal computers, the Internet, smart phones and other major innovations. So adoption is characterized by bubbles, frenzies and crashes.

The first decade of crypto saw investor bubbles and developer frenzies aplenty, but no sustained consumer craze.(1) This caused some people to work on integrated solutions from core blockchain functionality all the way to direct user interface, in conflict with the crypto ethos of “release self-contained tools, with no copy protection or controlling organization.” Others build niche applications with value-add so great even small user bases can pay for development.

Middleware, or blockchains that connect other crypto applications, is an intriguing third way. The idea is to leverage network effects. If you have 1,000 networks with 1,000 users each, and if the value of a network is the square of its users, connecting those networks multiplies their value by 1,000. (Full disclosure: I own Bitcoin and other cryptocurrencies.)

In less abstract terms, before the internet there were lots of sites and bulletin boards you could dial up and log into. But they weren't connected. The internet made them appear to be one giant connected site to users, with links, search engines and integrated security. Once a user bought the equipment and learned one site—which she might do for a compelling reason—accessing other sites was free.

Integrated solutions could make news in 2020, and niche applications are important but won't underlie explosive growth. Middleware has four major releases in 2019, any one of which could be dramatic or, if all four fizzle, could indicate that connecting blockchains is not the key innovation cyrpto is waiting for.

Cosmos, Nervos, NuCypher and PolkaDot are my big four.(2) All have 2019 mainnet releases. In traditional software, releases are minor events. You release alpha versions, then beta versions, then an official version everyone expects to be buggy and maybe around version 4.2 or so you have a stable product. In crypto, everything before the mainnet release is just testing.(3)

After mainnet, modifications are difficult. You can't afford to release a buggy mainnet and fix it in production. Moreover, you need a full suite of tools for users and add-on developers built and thoroughly tested before mainnet release. So mainnet releases come later than traditional software releases and are more consequential.

The four applications connect other crypto ideas in different ways, like a search engine differs from a password keeper. Cosmos is a hub other blockchains plug into. Once a blockchain is taught to interact with Cosmos, it can interact with any other blockchain connected to the hub.

Nervos is like the digital security certificates and secure transmission protocols for Internet safety. It offers the security features of a blockchain without overhead that makes Bitcoin and other global blockchains slow as usage grows. Users keep most data on fast local blockchains or on private shared databases, Nervos only collects aspects essential for security and contract enforcement.

NuCypher is like a password keeper. Today you sign away privacy in every website visit, and cannot verify the site complies with its terms or is secure from hacking. NuCypher promises enforceable secure storage of all your personal data, which you can release selectively in ways that naturally integrate with crypto.(4) PolkaDot is the most expensive and ambitious of the four. Like Cosmos, it connects blockchains, but through bridges rather than a hub. Like Nervos it provides common security for all connected chains.(5)

By the end of 2019, we should be able to judge the success of these ideas, and many other versions that didn't make my top four. Success would point to much greater network effects in crypto development, creating more user incentive to make the effort to plug into that network. More users means more developer effort means more applications means more users. And eventually perhaps more investment.

Failure, either technical or just lack of interest in adoption, suggests either crypto growth will be slow and driven by niche applications, or will lose some of its decentralized public features with future growth coming from centralized organizations protecting their investments with proprietary features—like Microsoft selling operating systems, developer tools and end-user products; or Apple bundling hardware and software. I don't expect major excitement in crypto prices in 2019, nor in rapid user adoption. Keep your eye on the middle layer.

(1) Silk Road was developing into one, but it was shut down. Non-bank electronic value transfers, especially across borders or in financially repressed countries, remain a steady business but not a fast-growing one,and crypto has induced competition from services like Venmo and ApplePay. Lots of big companies jumped on the crypto bandwagon, but they stripped away the revolutionary features of crypto in favor of technical enhancements to existing processes. The biggest public craze was trading crypto, but that can only survive if real uses are found for the technology.

(2) Only PolkaDot raised big money, $140 million in an October 2017 ICO. Cosmos had a $17 million ICO in April 2017, Nervos took $28 million of institutional money in July 2018 and NuCypher got $5 million of seed and venture funding in August 2018 and December 2017.

(3) If you create a document in an alpha or beta version of a word processor, you still have the document to edit when later versions are released. But if you post a transaction to a testnet release of a blockchain, it will be erased when the mainnet is released.

(4) You could offer your shopping history to the highest bidder for bitcoin, or let your insurance company see only those portions of your medical records relevant to claims, or make your resume available only to companies you might like to work at that aren't too closely connected to your current employer. These applications mainly practical with automated smart contracts running on blockchains for both the sources and uses of your data.

(5) The first three tools are like bitcoin—a protocal released for anyone to use as she likes—PolkaDot will be more like ethereum—an ecosystem in which developers build functionality on top of base functionality.

To contact the author of this story: Aaron Brown at aaron.brown@privateeram.com

To contact the editor responsible for this story: Robert Burgess at bburgess@bloomberg.net

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Aaron Brown is a former Managing Director and Head of Financial Market Research at AQR Capital Management. He is the author of "The Poker Face of Wall Street." He may have a stake in the areas he writes about.

©2019 Bloomberg L.P.