Authors: Dr. R. Declined, Prof. M. OnMute, Dr. S. CouldHaveBeenAnEmail
Affiliation: Department of Applied Workplace Suffering, Grumpy Coworker Institute of Technology
Published in: Proceedings of the 14th International Conference on Unnecessary Meetings (ICUM 2026)
DOI: 10.1337/gcit.qcrrt.2026.pleaseno

---

Abstract

We present Quick Call Request Response Theory (QCRRT), a formal game-theoretic framework for modeling the strategic interactions that occur when one knowledge worker sends another the message “Hey, do you have a minute for a quick call?” We prove that no equilibrium exists in which both parties are satisfied, that the word “quick” is semantically unbounded, and that the optimal strategy for the receiver is, in all cases, to pretend their headset is broken. We further introduce the Call Inevitability Conjecture, which states that the probability of a “quick call” occurring approaches 1.0 regardless of the receiver’s response, including no response. Novel contributions include the Emoji Reaction Signal Model, which demonstrates that a 👍 reaction is statistically indistinguishable from verbal consent, that ignoring a message is paradoxically less effective than reacting to it, and that the typing indicator is a catastrophic information leak with no known countermeasure. Our empirical results, gathered from 10,000 Slack channels over 18 months, confirm what every engineer already knows in their bones.

Keywords: game theory, quick calls, Slack terrorism, calendar violence, could-have-been-an-email, meeting-industrial complex, emoji semiotics, typing indicator threat model, thumbs-up paradox

---

1. Introduction

The phrase “do you have a minute for a quick call?” represents one of the most consequential unsolved problems in distributed systems. Unlike the Byzantine Generals Problem, which concerns faulty nodes in a network, the Quick Call Problem concerns a fully functional node (the Caller) deliberately disrupting another fully functional node (the Callee) under the pretense of efficiency.

Despite decades of advances in asynchronous communication — email, Slack, Teams, carrier pigeon — the Quick Call remains endemic. Previous work by NoMeetingTuesday et al. (2019) established that “quick” is a non-deterministic qualifier with observed durations ranging from 47 seconds to 3 hours and 11 minutes. Our contribution extends this work by formalizing the interaction as a two-player extensive-form game and proving several discouraging theorems.

We structure the remainder of this paper as follows. Section 2 establishes our formal model. Section 3 presents the core theorems. Section 4 introduces three thought experiments. Section 5 presents empirical results. Section 6 offers conclusions that will change nothing.

---

2. Formal Model

2.1 Players

We define a two-player game G = (C, R, A, U) where:

• C (Caller): The agent initiating the quick call request. Motivated by an unwillingness to organize their thoughts into written form.
• R (Receiver): The agent whose flow state is about to be annihilated. Currently mass-assigning variables or, God forbid, enjoying their work.

2.2 Action Space

The Caller’s action space is a singleton: A_C = {send_quick_call_request}. While theoretically the Caller could write their question in the message itself, empirical observation confirms this action is never selected. We refer to this as the Writeout Aversion Axiom.

The Receiver’s action space is:

• Accept — Join the call immediately. Expected cost: 45 minutes of context-switching damage plus the call itself.
• Delay — Respond “give me 10 minutes.” Initiates a recursive subgame in which the Caller messages again in 4 minutes.
• Decline — Respond “I’m heads-down, can you Slack me the question?” Triggers the Passive-Aggressive Escalation Protocol (Section 3.3).
• Emoji React — Apply a reaction emoji to the message without producing a textual response. A strategically ambiguous acknowledgment. See Section 2.5.
• Ignore — Do not respond. The Caller will find you. The Caller always finds you.

2.3 Utility Functions

The Caller’s utility function is straightforward:

U_C(outcome) = {
    call_happens:       +10
    call_doesn't_happen: -∞  (existential void)
}

The Receiver’s utility is modeled as:

U_R(outcome) = {
    left_alone:          +10
    call_is_actually_quick (<5 min): -2
    call_is_not_quick:   -8
    call_becomes_screen_share: -15
    "let me add someone else": -∞
}

2.4 Information Asymmetry

A critical feature of QCRRT is radical information asymmetry. The Caller knows the topic of the call. The Receiver does not. The Caller could state the topic in their initial message but — and we cannot stress this enough — they never do. We formalize this as the Topic Opacity Principle:

For all quick call requests q, the probability that q contains the actual subject of the call is ε, where ε → 0 as organizational size increases.

2.5 The Emoji Reaction Signal Model

The introduction of emoji reactions to messaging platforms created a novel action in the Receiver’s strategy space: the ability to acknowledge without responding. This fundamentally altered the game by introducing semantic ambiguity as a defense mechanism.

We define the Reaction Interpretation Function I(e, c) → [0, 1], which maps an emoji e applied in context c to the probability that a call will still occur. Our analysis reveals that emoji reactions form a spectrum from capitulation to soft deterrence, but critically, no emoji exists that maps to P(call) = 0.

Table 1: Emoji Reaction Taxonomy and Call Likelihood

Emoji	Caller Interpretation	Receiver Intent	P(call)
👍	“Great, I’ll call you now”	“I have acknowledged your message. That is all.”	0.92
👀	“They’re looking at it, I’ll follow up in 2 min”	“I see this and am choosing not to engage further.”	0.87
✅	“They’re ready, let me send the Zoom link”	“I am marking this as resolved without resolving it.”	0.78
⏰	“They need a minute, I’ll wait”	“I am busy. Please interpret this as a boundary.”	0.71
🫡	“They’re on board”	“I am saluting you ironically.”	0.83
😅	“Ha, they know it’ll be fun”	“I am in visible distress.”	0.76
🤔	“They’re thinking about it”	“I am performing contemplation to buy time.”	0.69
➡️	“They want me to send details”	“Please redirect this into text.”	0.54
📝	“They’re taking notes already”	“Write. It. Down. I am begging you.”	0.48
🙃	“They’re being playful”	“I am experiencing an emotion that HR would not approve of.”	0.81

Several critical observations emerge:

Observation 1: The Thumbs-Up Trap. The 👍 reaction is the most commonly deployed emoji response to a quick call request (41% of all emoji reactions in our dataset). Receivers select it because it requires zero cognitive effort — a single tap that says “I have seen this.” However, the Caller universally interprets 👍 as enthusiastic consent. This interpretation gap produces a P(call) of 0.92, making 👍 less effective at preventing a call than ignoring the message entirely (P(call | ignore) = 0.89; see Theorem 1). We term this the Thumbs-Up Paradox: the lowest-effort acknowledgment produces the highest-effort outcome.

Observation 2: The Stationery Deterrent. The only emoji reactions that reduce P(call) below 0.55 are those that explicitly encode a written medium preference: 📝 (P = 0.48) and ➡️ (P = 0.54). These succeed because they reframe the interaction as “I am requesting text,” forcing the Caller to confront the Writeout Aversion Axiom (Section 2.2). However, even these deterrents are overcome in 48–54% of cases, confirming the robustness of the Inevitability Theorem.

Observation 3: Emoji Reactions vs. Direct Textual Responses. A direct textual response of any kind — including a decline — carries a paradoxical cost. By responding with text, the Receiver proves they are (a) present, (b) at their keyboard, and (c) capable of synchronous communication. We formalize this as the Availability Proof Lemma:

Any textual response r to a quick call request constitutes a proof-of-availability that the Caller will enter as evidence in their case for the call.

Emoji reactions, by contrast, exist in a liminal authentication state. A reaction might have been applied from a phone while the Receiver was in a grocery store, on a treadmill, or in a dentist’s chair. The Caller cannot distinguish a reaction sent from a fully-equipped workstation from one sent while the Receiver is physically incapable of joining a call. This plausible deniability is the emoji reaction’s only strategic advantage.

However, this advantage is fragile. If the Receiver’s Slack status shows them as Active (green dot), the plausible deniability collapses entirely, and the emoji reaction becomes functionally equivalent to Accept.

Definition: The Reaction-Response Gradient. We define the engagement gradient Ω as:

Ω = f(response_type, response_length, response_latency)

where:
  Ω(emoji, -, t)         = 0.3 + 0.4·(status == green)
  Ω(text, L < 5 words, t) = 0.7
  Ω(text, L ≥ 5 words, t) = 0.85
  Ω(text_with_question, -, t) = 0.95
  Ω(typing_indicator_then_nothing, -, t) = 1.0

The final case — displaying a typing indicator and then sending nothing — is the worst possible outcome for the Receiver. The Caller has now seen evidence of active composition followed by deliberate withdrawal, which is interpreted as either extreme interest (the Receiver is carefully composing a response) or evasion (the Receiver started to decline and lost their nerve). Both interpretations increase Caller persistence to maximum.

Corollary (The Typing Indicator Catastrophe): If the Receiver begins typing a response and then deletes it, the Caller’s resolve increases by a factor of 2.3× relative to no response at all. The typing indicator is a one-way ratchet. You cannot un-type.

---

3. Core Theorems

Theorem 1: The Inevitability Theorem

Statement: For any Receiver strategy s ∈ A_R, the probability that the call eventually occurs is:

P(call | s) ≥ 1 - δ,  where δ is negligibly small

Proof sketch: We proceed by exhaustion (both mathematical and emotional).

• If R accepts: call occurs trivially.
• If R delays: C re-initiates within T < stated_delay. By induction on messages, R eventually accepts.
• If R declines: C responds “it’s easier to just talk through it.” R’s written medium preference has been overridden. Call occurs.
• If R ignores: C escalates to (a) multiple messages, (b) email, (c) walking to R’s desk, or (d) messaging R’s manager with “having trouble reaching R.” Call occurs with additional emotional damage. ∎

Theorem 2: The Quicklessness Theorem

Statement: No call described as “quick” will conclude in under T_min = 12 minutes.

Proof: By contradiction. Assume a quick call concludes in t < 12 minutes. Then the content of the call could have been communicated asynchronously in a message of length L ≤ 280 characters. But then the Caller would have written the message instead. The Caller did not write the message. Therefore, the content exceeds what can be communicated briefly. Contradiction. ∎

Theorem 3: The Scope Creep Lemma

Statement: The number of topics discussed in a quick call is a monotonically increasing function of elapsed time.

Proof: Let T(n) be the set of topics at time n. We observe that for any topic t_i discussed, the Caller will generate a transition edge to topic t_{i+1} via one of the following connectives:

• “Oh, while I have you…”
• “Actually, one more thing…”
• “This is semi-related, but…”
• “Real quick—” (recursive invocation of the quick call within the quick call)

Since each connective is uttered with probability p > 0 at every time step, and no termination condition is ever met, lim(n→∞) |T(n)| = ∞. ∎

3.3 The Passive-Aggressive Escalation Protocol (PAEP)

When the Receiver selects Decline, the following state machine is activated:

State 0: "No worries! Can you look at [link] when you get a chance?"
State 1: (4 hours later) "Hey, did you get a chance to look at that?"
State 2: "I think it's just easier to hop on a call."
State 3: Calendar invite appears. No description. No agenda. 30 minutes blocked.
State 4: "I added [your manager] since this touches their area too."
State 5: You are now in a recurring weekly sync.

Theorem: Once PAEP reaches State 3, the system is irrecoverable.

---

4. Thought Experiments

4.1 Schrödinger’s Agenda

Consider a Receiver who has received a quick call request. Until the call begins, the topic exists in a superposition of all possible states: a genuine five-second question, a full architecture review, an HR concern, a vent session about another coworker, or a screen-share where the Caller slowly scrolls through a spreadsheet saying “do you see what I mean?”

The Receiver cannot observe the topic without joining the call, thereby collapsing the wave function. However, collapsing the wave function requires being on the call, which is the outcome the Receiver is trying to evaluate. This creates a measurement paradox: the only way to know if the call is worth taking is to take the call, at which point the knowledge is worthless.

Corollary: All quick call requests must be evaluated under maximum entropy assumptions. Assume the worst.

4.2 The Prisoner’s Desk Dilemma

Two engineers, Alice and Bob, sit in adjacent cubicles. Their manager has asked each of them, independently, “can we do a quick call about the deployment?” Each must decide whether to Accept or Deflect.

	Bob Accepts	Bob Deflects
Alice Accepts	Both on a 40-minute call. Manager happy. Engineers sad. (-5, -5)	Alice alone on call. Alice becomes “point person.” (-8, -1)
Alice Deflects	Bob alone on call. Bob becomes “point person.” (-1, -8)	Manager schedules a meeting for all three. With a shared doc. And action items. (-12, -12)

The Nash Equilibrium is (Accept, Accept), confirming that both engineers will sacrifice their afternoon because the alternative is worse. This is also known as the Meeting Arms Race.

4.3 The Trolley Problem (Calendar Edition)

You are standing at a calendar. A runaway meeting invite is hurtling toward five of your coworkers. You can divert it onto a side track where it will hit only you. However:

• The meeting has no agenda.
• The organizer listed it as “optional” but will absolutely notice if you don’t join.
• The description says “Sync.”
• It recurs weekly.

Do you pull the lever? Trick question. The meeting was already accepted automatically by your Outlook rules, and you won’t discover it until the reminder fires in 3 minutes while you’re debugging a segfault.

---

5. Empirical Results

5.1 Dataset

We analyzed 10,000 instances of “quick call” requests across 47 organizations, collected via instrumented Slack workspaces (IRB approval pending; our IRB is in a quick call).

5.2 Key Findings

Actual duration of calls described as “quick”:

Claimed Duration	Actual Mean	Actual Median	Max Observed
“one minute”	14 min	11 min	2 hr 47 min
“real quick”	22 min	18 min	1 hr 55 min
“five minutes tops”	31 min	27 min	3 hr 11 min
“super quick I promise”	38 min	34 min	the call is still going

Topic disclosure rate in initial message: 3.2% (n = 10,000). Of the 3.2% that included a topic, 61% were inaccurate.

Screen share probability by call minute:

• Minute 0–5: 12%
• Minute 5–10: 34%
• Minute 10–15: 67%
• Minute 15+: 89%

Correlation between “this will be quick” and actual quickness: r = -0.23 (p < 0.001). The assertion of quickness is a negative predictor of quickness.

5.3 The “While I Have You” Phenomenon

We observed that 73% of quick calls contain at least one instance of the phrase “while I have you.” Each utterance adds an average of 8.4 minutes to the call duration. The record in our dataset is seven consecutive “while I have you” transitions, converting a question about a typo in a README into a full roadmap planning session.

5.4 Response Modality and Call Materialization Rates

We classified all 10,000 quick call requests by the Receiver’s initial response modality and tracked whether the call ultimately occurred. Results are presented below.

Call materialization rate by response type:

Response Modality	n	Call Occurred	P(call)	Mean Time to Call
Direct accept (“sure!”)	2,847	2,847	1.000	0.4 min
Textual response with question about topic	1,203	1,144	0.951	3.2 min
Textual response (any other)	1,891	1,610	0.851	8.7 min
Emoji reaction (👍 👀 🫡 ✅)	2,104	1,820	0.865	6.1 min
Emoji reaction (📝 ➡️ 🤔)	612	341	0.557	22.4 min
Emoji reaction (😅 🙃 🫠)	394	319	0.810	9.8 min
No response (ignored)	949	844	0.889	14.3 min

Several findings merit discussion.

First, the affirmative emoji cluster (👍 👀 🫡 ✅) produces a call materialization rate of 0.865, which is statistically indistinguishable from a direct textual acceptance at the α = 0.05 level. The Receiver believes they are buying time. They are not.

Second, the deflective emoji cluster (📝 ➡️ 🤔) is the only response modality that reduces P(call) below 0.60. These emojis succeed because they impose a cognitive burden on the Caller: to proceed, the Caller must explicitly override the Receiver’s implicit request for written communication. Approximately 44% of Callers lack the activation energy to do so, particularly after lunch.

Third, and most disturbingly, ignoring the message entirely produces a higher call materialization rate (0.889) than using an affirmative emoji reaction (0.865). This appears paradoxical until one considers the Silence Escalation Effect: ignoring a message triggers the Caller’s loss aversion, producing escalatory behaviors (repeated messages, channel-switching, physical approaches) that carry more social pressure than an organic conversation flow. In short, saying nothing is worse than saying 👍. The game is rigged.

Fourth, the distress emoji cluster (😅 🙃 🫠) was hypothesized to deter Callers by signaling discomfort. It does not. Callers interpret distress emojis as camaraderie (“haha yeah I know, calls are annoying, but let’s do it anyway”). The melting face emoji (🫠) in particular was found to increase Caller confidence that the Receiver would join, as the Caller reads it as good-natured resignation. P(call | 🫠) = 0.84.

Compound reaction analysis: In 7.3% of cases, the Receiver applied multiple emoji reactions to a single quick call request. The most common compound reaction was 👀 + 👍 (interpreted by the Caller as “I see it AND I’m in”). The rarest was 📝 + ⏰ + 🤔 (interpreted as “I need to think about writing something later”), which achieved the lowest call materialization rate in our entire dataset: P(call) = 0.31. However, applying three emoji reactions to a single message also caused 100% of Callers to message the Receiver asking “…you good?” — initiating an entirely new conversational thread that itself had a 62% chance of culminating in a call.

---

6. Proposed Mitigations

6.1 The Mandatory Topic Protocol (MTP)

We propose that all quick call requests be rejected by automated middleware unless they contain a topic description of ≥ 15 words. Early trials show a 94% reduction in quick call volume, as the act of typing the topic causes 94% of Callers to realize they could have just sent the message.

6.2 Quick Call Futures Market

We propose a prediction market where engineers can short-sell the claim that a call will be “quick.” Initial modeling suggests this market would outperform the S&P 500.

6.3 Status Engineering

Advanced practitioners report success with the Perpetual Red Dot Strategy: maintaining a permanent “In a Meeting” or “Focus Mode” status. While ethically questionable, our utilitarian analysis shows net positive welfare outcomes across all simulated organizations.

---

7. Conclusion

We have established that the Quick Call is a dominant strategy for the Caller and a dominated strategy for the Receiver, that no mechanism design can prevent scope creep once a call has begun, and that the word “quick” has been so thoroughly divorced from its temporal meaning that it now functions purely as a social lubricant for calendar invasion.

Future work includes extending QCRRT to model the “let’s take this offline” → “let’s hop on a call to discuss” pipeline, formalizing the “I’m going to share my screen” threat model, and investigating whether the phrase “I’ll let you go” actually lets anyone go (preliminary results: no).

---

References

1. NoMeetingTuesday, A., et al. “On the Semantic Drift of ‘Quick’: A Longitudinal Study.” Journal of Calendar Abuse, vol. 12, 2019.
2. Declining, R. “Could Have Been an Email: A Proof by Exhaustion.” SIGMEETING Proceedings, 2021.
3. OnMute, M., SorryGoAhead, N. “Concurrent Speech Collisions in Videoconference Protocols.” IEEE Transactions on Awkward Silence, 2023.
4. Status, A. Green. “Why Is Your Status Green If You’re ‘Heads Down’? An Adversarial Analysis.” Workshop on Presence Detection Attacks, 2024.
5. ThisCouldHaveBeenASlackMessage, P. “The Thermodynamics of Meeting Sprawl.” Physical Review of Organizational Entropy, vol. 7, no. 3, 2022.
6. LetMeAddSomeoneElse, J. “Scope Creep as a Contagion Model.” Epidemiological Studies in Project Management, 2020.
7. Grumpy Coworker Editorial Board. “We Told You So.” grumpycoworker.com, 2024–ongoing.
8. ThumbsUp, K., et al. “The Semiotics of 👍: How a Single Tap Destroyed an Afternoon.” ACM Conference on Human Factors in Messaging (CHI-LL), 2024.
9. ReactOnly, D. “Emoji as Illocutionary Act: When 👀 Means ‘No’ But Everyone Hears ‘Yes.’” Computational Pragmatics Review, vol. 4, 2025.
10. TypingIndicator, F. X. “The Dot-Dot-Dot Threat Model: Information Leakage via Composition Signals.” USENIX Security Symposium on Workplace Surveillance, 2023.

---

© 2026 Grumpy Coworker Institute of Technology. This paper is licensed under CC BY-NC-SA 4.0 (Creative Commons By No-Calls, Share-A-Sigh).

Corresponding author: declined@gcit.edu. Do not call. Send an email. Or better yet, a Slack message with the actual question in it.