May/June 2018 issue of acmqueue The May/June issue of acmqueue is out now



Development

  Download PDF version of this article PDF

ITEM not available

acmqueue

Originally published in Queue vol. 10, no. 3
see this item in the ACM Digital Library


Tweet



Related:

Jez Humble - Continuous Delivery Sounds Great, but Will It Work Here?
It's not magic, it just requires continuous, daily improvement at all levels.


Nicole Forsgren, Mik Kersten - DevOps Metrics
Your biggest mistake might be collecting the wrong data.


Alvaro Videla - Metaphors We Compute By
Code is a story that explains how to solve a particular problem.


Ivar Jacobson, Ian Spence, Pan-Wei Ng - Is There a Single Method for the Internet of Things?
Essence can keep software development for the IoT from becoming unwieldy.



Comments

(newest first)

Bob Binder | Fri, 14 Sep 2012 15:35:40 UTC

Although the technical debt metaphor is very catchy, it is not quite apt. I've argued that "technical equity" is an equally important notion for understanding the consequences of development choices. http://www.robertvbinder.com/technical-equity/

Recasting technical debt as a call option is interesting, but isn't quite right either - options are much more complex than an IOU. Buying a "naked" call on an underlying stock means that you spend a small amount in the present so that you may realize a gain in the future, but only if the value of the underlying stock increases by more than the cost of the option. The buyer risks the cost of the option to gain the possible increase. Selling a naked call has a mirror effect. Typically, you sell a call only if you expect the underlying value to drop, so that the counterparty will not require you to give them the optioned shares of the underlying. The seller's profit is the price of the option. If the underlying value rises and the buyer exercises your call, you have to buy the underlying at the higher price and give it to call buyer -- your net cost is the (higher) price of the underlying less the money received from selling the call. So, the call seller is at risk for that cost. All of this is predicated on a fixed and certain date for exercising the call, usually less than three months.

Is deferring or skimping on development like buying a naked call? Yes, in that choosing to pay a small cost now may lead to a larger benefit later (for example, designing an API, but only implementing the interface code.) But who is selling the call? Who believes that the value of your codebase will decline? Options don't make sense (and would not trade) without a counterparty that is more or less pessimistic/optimistic. The metaphor places the codebase sponsor/owner in the role of selling a naked call, which only works if we assume the sponsor believes the developer's work will be worth less later.

Similarly, making an implicit commitment to future payback for deferred or inadequate development (technical debt) is rarely based on a quantified analysis of cost/benefit or the sustainability of that commitment. The codebase sponsor/owner is the "lender" and rarely understands the economic implications of the arcane and often arbitrary decisions of developers, or of shortcuts taken to cut time and cost, or just plain incompetence. http://www.robertvbinder.com/how-technical-debt-turns-into-technical-bankruptcy/ http://www.robertvbinder.com/how-a-big-ball-of-mud-becomes-a-black-hole-why-software-architecture-and-process-matters/

I like these metaphors to the extent that they remind us of the consequences of software development strategies and trade-off choices, but they are very thin conceptual ice.


George Samaras | Fri, 25 May 2012 21:24:50 UTC

Excellent! Very well presented; now, if only folks will pay attention... BTW, your "3 variables" + debt correspond to the four fundamental attributes of new product development: Budget (resources), Schedule (time), Scope (functionality), and Quality (technical debt). So, what you are describing when you discuss different perspectives of managing technical debt is how specific stakeholders prioritize their own needs, wants, and desires and thus determine quality of the product, process, or service. GM Samaras Pueblo, CO


Leave this field empty

Post a Comment:







© 2018 ACM, Inc. All Rights Reserved.