Lindy Effect

A theory that states, the future of an idea or technology is proportional to how old it currently is so that every additional period results in longer life expectency.

For example, the IRS use of Cobol–each year the systems written in Cobol exist, the greater the likelihood Cobol will stick around for many more years.

In technology, this happens regularly. In the case of programming languages, once it has reached a certain age and has been used in some critical systems, it’s much harder to justify rewriting it. As it gets even older and the programming language ‘falls out of style’ the pool of workers shrinks, knowledge of the system is lost (e.g. the IRS lost the source code!) and the cost of replacing it with a new language means writing it again from first principles–including the now multitude of years of exception handling.

Lindy’s Law: the article in the New Republic that first coined the theory.

See also:

  • Don’t Combine Domain Name Hacks

    I see a lot of startups with domain names like getbarai.co and trymspledword.io which combine multiple domain name hacks. These look untrustworthy, are hard to search, difficult to spell, and long. It’s hard enough to get anyone interested in what you are doing that introducing any friction can cause people to forget about it.

  • GDP Declined by $181 Billion Due to COBOL

    A recent paper analyzed the causal effects of delays in updating unemployment insurance systems written in COBOL. States that used COBOL to process UI claims had a 4.4% decline in total card consumption compared to states that didn’t use COBOL. The author’s Fermi estimate is a $181B decline in real GDP due to COBOL.

  • Smart Contracts Are a Computing Model That Requires Maintenance Forever

    Smart contracts that run on a distributed ledger effectively need to work forever to be useful. Working forever requires maintenance and security—the longer a computer program runs, the more exposure it has to change and the opportunity to be exploited.

  • Data Is Centrifugal

    The predominant system for managing data today pushes it far away from the people it represents. This is bad for the world because data is the primary way identity is represented (census, social profile, certification) and creates the grounds for exploitation (theft, ads, terrorism).