Online does not mean available for interruption
Aliases: presence versus availability · interruptibility · response availability
What it is
The presence–availability distinction separates technical presence—connection, an open application, or recent activity—from willingness and capacity to be interrupted. Someone can be online while presenting, writing, or caregiving, and appear offline while accepting an emergency call. Availability depends on intent, task cost, channel, and sender relationship.
Why it happens
Presence is cheaply generated from device events and is timely but semantically narrow. Willingness is contextual and changes with task, contact, and consequence. Treating connection as permission makes senders attribute delayed response to the person rather than the signal's meaning, creating pressure. Separating them preserves presence awareness without converting it into a response commitment.
Studying it
Experience sampling at message arrival can record presence, self-reported interruptibility, contact, channel, and task stage, then compare response delay, resumption cost, and retrospective suitability. Model people separately because thresholds vary. Fast response is not sufficient ground truth: pressured users may reply immediately at high recovery cost.
Where it stops holding
On-call and emergency roles have prior response agreements, but online still does not guarantee device access. Social availability does not imply work availability. Hiding presence protects boundaries but may harm tight coordination, requiring channel exceptions and emergency rules.
Applying it
- Label online as connection, not availability.
- Let availability vary by channel, contact, or task rule.
- Support quiet delivery and scheduling rather than alerting every online recipient.
- Validate with recipient appraisal and resumption cost, not response speed alone.