Durable design of digital artefacts

With an emphasis on artefacts built for the web.

There are different kinds of durability, absolute material durability is contrasted with practical, sustainable, and evolving durability through time. In the former case the goal of design is to create something formidable and unassailable, something as unsinkable as the Titanic perhaps.

To give it a fairer shake, there are times when a bruteforce, more wood behind fewer arrows approach is probably a good fit, such as when working to a deadline. But when design can be open ended we have an opportunity to create living structure1, forms adapted to and integrated into their environment. These structures emerge in harmony with their environment and lean into that connection more, not less as time goes on, lending a deepening resilience to both.

I’m more interested in this second form of durability, where thought is given not only to creating a durable initial design but also to the full life-cycle of the design, and to its place in its surroundings.

Where the former seeks to stay the forces of decay through sheer strength, the latter acknowledges the inevitability of decay and plans for it.

This site should be durable like the Ship of Theseus, durable not by all-out resistance to the incessant tides of decay, but by adopting a posture of structural agility. Success here would be marked by qualities of anti-fragility. Anti-fragility in digital design probably hinges on maintaining optionality.

But what forms of decay are we guarding against?

Durability in the digital age is often viewed through the binary lens of the medium itself. Whether we are talking about bit rot, or its web borne counterpart Link Rot, digital decay is generally a boolean, either a thing has not decayed at all or has decayed and is lost.

But if the durability of this site is measured also by the durability of its ideas then all of a sudden we have more axes of decay to account for in its design. The decay of ideas is not boolean, it is one of gradual degeneration, introducing the complication of a possible Sorites paradox upon each idea entombed in this site. Do ideas decay? How can the decay of ideas be measured? And relevant to the paradox, can such a measure of decay determine when an idea reached a state of decay? Further, do all ideas decay? And should all ideas be protected from such degeneration?

To that last question, I think probably they need not be. Most of the ideas with which this site is stocked are of low quality, let time reveal their (lack of) merit, only a small fraction are likely to warrant special attention. This recognition of a low signal-to-noise ratio raises its own questions though: What should be done with those ideas that reveal themselves to be mere noise? What ought their lifecycle to be? If left to sink naturally beneath the surface of a growing corpus, do they pose any kind of threat to stronger ideas with which they cohabitate?

Christopher Alexander’s The Nature of Order would probably be a very helpful resource for figuring out workable patterns for this kind of design but at 2500 pages, I have not read it.

Primitives for durable digital design

Durable web design, the kind of design to which this site aspires should consider:

Durability of medium

  • Separation of prose and presentation
  • A conservative approach to adoption of new technologies to mitigate the dangers of inaccessibility and accrued technical debt.
  • Progressive enhancement. Non-essential features should be designed and implemented in such a way as to not calcify or degrade the basal feature set.

Relevant prior art here would include Jon Postel’s Robustness principle

Durability of ideas

  • Methods for distinguishing and developing ‘evergreen’ ideas from the medley of nonsense

  1. A phrase explored by Christopher Alexander↩︎