科技债务是一个1992年由Ward Cunningham提出的比喻，指软件开发中因质量妥协而产生的隐性成本，类似于财务债务。与财务债务不同，科技债务的偿还时间不确定，可能突然爆发，如同1986年挑战者号航天飞机因质量妥协导致的灾难性事故。文章主张使用“科技风险”一词来更准确地描述这种不可预测的代价，强调质量妥协会创造风险，而风险的大小取决于预期回报。

🔍 科技债务源于1992年Ward Cunningham提出的比喻，指软件开发中因质量妥协而产生的隐性成本，类似于财务债务，但其偿还时间不确定，可能突然爆发。

⚠️ 与财务债务不同，科技债务的偿还时间不确定，可能突然爆发，如同1986年挑战者号航天飞机因质量妥协导致的灾难性事故，强调质量妥协会创造风险。

📈 文章主张使用“科技风险”一词来更准确地描述这种不可预测的代价，因为质量妥协会创造风险，而风险的大小取决于预期回报，而非按固定时间偿还。

📊 财务债务是可预测的，如按固定时间偿还的贷款，而科技债务则像“引爆式贷款”，其偿还时间不确定，可能随时爆发，造成巨大损失。

🧐 文章通过对比财务债务和科技债务，指出科技债务的不可预测性使其更像是一种风险而非债务，需要更谨慎地管理和评估。

Tech Debt becomes due swiftly and chaotically. Tech Risk is a better metaphor.

What is Tech Debt?

Tech debt is a metaphor coined in 1992 by Ward Cunningham¹.

In software engineering and adjacent disciplines like DevOps, it relates shortcuts on quality to financial debt.

We all know how financial debt works: you pay to use someone else's money. The amount you pay is proportional to the amount you borrow, how long you borrow it, and the interest rate. Borrowing can be a good idea since having money right now enables valuable events like buying a house or starting a business.

Wardites hold that tech debt, like financial debt, incurs a penalty which over time compounds the amount of labor required to continue delivering. Etiologies vary: Some folks hold that tech debt originates when stakeholders intentionally trade quality for delivery speed. Others say it happens inadvertently. Some distinguish between reckless and prudent tech debt².

Financial Debt is Predictable

The tech debt metaphor appeals because it is easy to explain. Unfortunately, the metaphor is strained. Tech debt is not like financial debt. Let's see why not.

Mortgages and other loan products are predictable, for both the lender and the borrower. Borrowers repay on a steady schedule which has a specified maturity term, e.g. 15 or 30 years. The interest rate is fixed or mostly fixed. From day one, a spreadsheet predicts exactly how much principal and interest the borrower will still owe in month 137. For a bank, any loan is part of a diversified portfolio of many loans. While banks do fail³, stricter underwriting rules since 2008 have made it less common. Moreover, the societal harm of any individual bank failure is mitigated by distributing its cost across the entire banking system, facilitated by the full faith and assurance of the US Federal Reserve.

Tech Debt Detonates

Here's how financial debt could be like tech debt. Imagine a hypothetical dystopian financial system in which banks could demand full mortgage repayment on a whim. We'll call these detonating mortgages⁴ — If you can't pay a few hundred thousand dollars right now, you lose your home. No, you didn't do anything wrong, but the bank needs some liquidity today, so you and your children are now living in a hotel.

Tech debt is like a detonating mortgage. Its payment becomes due unpredictably. Its timing is not fair or uniformly distributed.

The people who investigated the Challenger disaster⁵ didn't have the term tech debt, but they listed quality shortcuts among the root causes. These quality shortcuts exposed the space program to risk. If NASA could have had at their disposal a spreadsheet predicting that the O-ring would fail on January 28, 1986, they would have decided not to fly.

Conclusion

Quality shortcuts create risk. Some risk is necessary to reach our goals; how much we can tolerate depends on the expected reward⁶.

Let's not delude ourselves when we're taking risks. Deferred tech quality is not like a loan. The cost of quality shortcuts doesn't manifest on a schedule.

Use the term tech risk.

Next...

Want to identify, prioritize, and mitigate tech risk? Subscribe to my email list below. I plan to write more about this topic.

References

Ward CunninghamTechnical Debt QuadrantBanking in Very Uncertain TimesPatrick McKenzie invented this thought experiment in Banking in Very Uncertain TimesRogers Commission ReportThe Most Important Thing