The same event should not be fully repeated on every channel
Aliases: full duplication · verbosity · partial redundancy
What it is
One event does not need to be repeated in full on every channel. Showing, reading aloud, and vibrating the same message in full makes users process three equivalent copies, multiplying cost without adding information. It is more useful for channels to carry parts of the content, or for some to carry only a presence cue.
Why it happens
Full duplication's cost comes from processing that does not scale with information. Users must process everything to determine the copies are duplicates, and that de-duplication judgment itself costs time and attention. Partial redundancy—one channel carrying full content and another only signalling presence—preserves delivery while avoiding multiple processing passes. This is why "use multiple channels for critical information" is not the same as "put all the content on multiple channels": the former is settled by arbitration rules, the latter is indiscriminate copying.
Studying it
Compare full duplication with partial redundancy: measure processing time, perceived workload, and downstream task performance for the same event. Variables include how complete the duplication is, event frequency, and content familiarity. Include the time spent stripping redundant content, since it shows users were forced to re-judge duplication.
Where it stops holding
When the information is very short—a single status word—full duplication is cheap and can be acceptable or even safer. When users need complete content without looking at the screen, as while driving, hearing may need to carry it all, and duplication has a real justification. If content has several fields of differing importance, full duplication can be limited to the critical field.
Applying it
- Let the primary channel carry full content and the others carry only cues or critical fields.
- Avoid duplicating every field indiscriminately across channels.
- Where full content must come through one channel, designate it the primary one and keep the rest as cues.
- Verification: measure processing time and workload for full duplication versus partial redundancy and confirm the level of redundancy matches the benefit.
Related
- Within the group: D5.06.1 Simultaneous multichannel output needs arbitration rules · D5.06.3 Priority follows consequence level, not channel capability
- Adjacent: D4.08.2 In low-load settings, redundancy has limited marginal value and may overload · D5.01.1 Redundancy improves reliability but adds load
- Search terms:
full duplication·partial redundancy·processing cost
Cards in the same group
- D5.06.1Simultaneous multichannel output needs arbitration rules
- D5.06.3Priority follows consequence level, not channel capability
- D5.06.4Arbitration must handle queueing and dropping among simultaneous events
- D5.06.5Interruptive arbitration cuts off feedback; queued arbitration lets it finish first
- D5.06.6Cross-application arbitration usually belongs to the system layer
- D5.06.7Users should be able to override the default arbitration for a situation