Short-term guests and long-term residents need different depths of notice

What it is

For the same passive collection in the same space, the depth and form of the notice owed differ by length of stay: long-term residents need standing, queryable notice plus a channel into configuration; short-term guests need conspicuous notice at the moment of entry and a zero-dependency path to understanding. One uniform scheme (a single household consent, one shared policy) fits neither — too deep for guests (they will never read settings) and too shallow for residents (they need participation, not notification).

The layering is not by identity label (family / guest) but by two continuous variables: exposure dose (how much data accumulates) and control leverage (how many means exist to change the situation).

Why it happens

The two groups' divergence follows a mismatch of dose and leverage:

• Doses differ by orders of magnitude. Residents are collected continuously; their data spans sleep-wake rhythm, habits, health signals — harmless per day, a complete profile in aggregate. Guests are collected for hours and leave fragments. The dose gap sets the risk gap and the content gap: residents need to know "how a longitudinal profile forms and flows", guests only "what this visit will capture".
• Leverage is entirely different. Residents can join household negotiation, reach the configuration, and speak when devices change — their notice must connect to participation rights. Guests have no account, no configuration surface, no seat at the table; their only levers are "ask the host" or "leave". Guest notice must therefore be zero-dependency — obtainable without installing an app, registering, or asking (a sign at the entrance, a label on the device, one sentence from the host) — and its executable outcome is behavioural choice (where in the home to be, what to say).

Making the guest layer a simplified copy of the resident layer ("show them the same policy") fails by construction: the form depends on things they do not have — accounts, patience, a negotiating relationship with the host.

Studying it

• Role-stratified interviews: smart-home research (e.g. the SOUPS 2017 interviews by Zeng, Mare and Roesner) already reports secondary users — guests, nannies, cleaners — whose privacy is decided by the buyer, with notice channels designed for no one. Stratified interviews (owner / resident / visitor) on notice needs versus actual knowledge measure both the necessity of layering and the present gap.
• Bystander privacy research: the literature on third-party privacy discusses the notice-and-control predicament of non-users facing other people's devices, and can frame the guest side.
• Notice-reachability audits: let a new guest enter the space naturally, then afterwards measure what they knew about "what collects here" and from where; compare with residents' knowledge paths to quantify the reachability gap.

One methodological caution: do not measure guest awareness with the host present — accompaniment itself triggers explanation; probe retrospectively after natural visits.

Where it stops holding

• The layering is not a dichotomy. Flatmates, a grandparent who visits weekly, a cleaner present three hours a week sit between the poles — assess on "dose × leverage": the more frequent and longer the stay, the closer to the resident configuration (query channels, a seat in decisions).
• In private homes guest notice is weakly enforceable. No statute obliges notice in a private residence; it rests on the host's diligence and product defaults (devices carrying their own marking). Commercial lodging (hotels, short-term rentals) does carry statutory notice duties — that side belongs to public-space monitoring notice, a different rule set.
• "Notified therefore permitted" does not follow. Guest notice delivers knowing, not agreeing — a guest told about the camera still faces the weak choice of accept-or-leave; restraint on the host's side (turning it off, masking it) remains a necessary complement, not replaceable by notice.

Applying it

• Guest layer, zero-dependency notice: one visible line at the entrance ("cameras and audio recording indoors, kept 7 days") plus identifiable devices; requiring no action from anyone. Pair it with a default script for the host — a prompt card or a one-tap in-app explanation — lowering the cost of bringing it up.
• Resident layer, participatory notice: a standing inventory of devices and data; changes to capture configuration (recording on, voice enabled) go through a household confirmation flow, not the owner's unilateral switch.
• Configure the middle by frequency: for high-frequency guests (cleaner, elders) provide an in-between channel — e.g. a shareable read-only "about this device" page viewable without registration.
• How to check: test the two groups separately — for guests, a 3-second attention-capture test on entry (did the notice register) plus ability to restate the core fact; for residents, ability to list the collection inventory and the latest configuration change. Report separately; do not merge.

Related

• Same group: Z6.06.1 Passive collection differs from active query, and usually goes unnoticed · Z6.06.2 After-the-fact notice cannot replace a prior opportunity to know · Z6.06.3 Notice must match the situation; dense terms do not work
• Nearby: Z6.04 The presence of non-users · Z6.05.3 Family members diverge sharply in camera acceptance
• Search terms: bystander privacy · guest notice · secondary users · smart home privacy