Appendix
Glossary and Reference
Every term ATEND uses, defined in one place. Plus quick-reference tables for the daily and weekly rituals, inbox processing paths, fit score components, and the core design principles that shape every product decision.
Glossary
A
Ease of starting
A task field that classifies the felt resistance at the moment of beginning — not the task's objective difficulty. Three values: Easy (you can start immediately with no friction), Some resistance (requires a moment to engage or involves mild aversion), Really hard (strong resistance at the moment of starting). Used by the fit score to suppress high-resistance tasks when your energy is low, and to surface the Activation Interview earlier when a task keeps stalling.
Anchor mode
Controls how the next due date is calculated when a recurring task is completed. Calendar schedules the next instance from the original anchor date regardless of when you finished. Completion schedules the next instance from the date you completed this one. Use Calendar for fixed obligations; use Completion for maintenance tasks where the interval matters more than the date.
Architecture Session
See Weekly Architecture Session.
B
Big Three
The three most important outcomes you're committed to completing in the current week, set during the Weekly Architecture Session. Not tasks — outcomes. The Big Three provide direction for the Launchpad and grant a +25 fit score boost to all tasks in any linked project. Only one or two Big Three outcomes are set during re-engagement after a break.
Buffer zone
A protected window of time before and after a calendar event — for travel, transition, and recovery. Buffer zones appear on the time strip as muted fills. Tasks are not scheduled into buffer zones by the fit score algorithm.
C
Cadence
A connection frequency set for a family member or close friend — weekly, monthly, or quarterly. When the cadence interval passes without a logged connection, a low-priority prompt appears: "You haven't connected with [Name] in a while. Want to reach out?"
Capture
The act of getting something out of your head and into ATEND's inbox. Capture should be as fast and frictionless as possible — voice, text, or document. Nothing is decided at capture time. Deciding happens during processing.
ATEND
The acronym that defines the five-phase system: Capture everything. Limit ruthlessly. Execute on context. Account for time. Review and reset. Also the name of the app.
Context filtering
The algorithm that decides which tasks appear in your Now list at any given moment. It scores every active task against your current energy level, time available, location, and other factors, then surfaces the highest-scoring tasks. Formally called the fit score algorithm.
Context notes
A freeform living field on each contact record — a few lines of what's currently true and relevant about this person. Surfaces automatically when the contact appears in a task, project, or appointment view. Updated by tapping "Anything worth remembering?" after completing a task involving the contact.
D
Daily Close
The evening ritual that closes out the workday. Runs for approximately ten minutes. Shows completed and deferred tasks, sets tomorrow's MIT, and ends with a capture prompt to offload any remaining open loops. Tracked as a streak.
Deferred count
The number of times a task has been moved out of today's active list without being completed. At three deferrals, the system surfaces the "Who Does This Serve?" prompt before allowing another. At repeated deferrals, the system may surface an activation interview.
Dependency
A hard relationship between two tasks where one cannot begin until the other is complete. A task with a dependency is hidden from all views — including the Launchpad — until the blocking task is marked done. When the blocker completes, the dependent task activates automatically.
E
Energy level (morning)
Your self-reported cognitive state at the start of the day, set during Morning Setup. Four options: High (ready for deep work), Medium (can focus but not for long), Low (need easy wins), Fragmented (scattered day of small slots). Shapes the fit score for the entire day by matching task energy requirements to your actual capacity.
Energy required
A task field classifying the cognitive demand of the work. Three values: Deep (sustained concentration, creative or analytical work), Medium (engaged but not depleted), Admin (low-cognitive routine work). Matched against your morning energy level by the fit score.
F
Fit score
A number computed for every active task that reflects how well it matches your current context — energy, time available, stakes, and other factors. The Now list on the Launchpad shows the tasks with the highest fit scores right now. The score is never shown to you; only its result (what surfaces) is visible.
Focus Mode
A distraction-free environment that launches when you tap a task to begin working. Shows only the task, a countdown ring, session progress, and optionally a relevant file. All other views are hidden. Sessions default to 25 minutes. Completed sessions generate a difficulty check and time confirmation before returning to the Launchpad.
H
Role
A distinct life area — a context with its own relationships, obligations, and definition of "on top of things." Every task, project, appointment, and contact belongs to a role. Maximum seven active roles. The role system drives the color coding throughout the app and feeds the neglect detection algorithm.
Role neglect detection
A background algorithm that tracks engagement across your roles using a recency-weighted activity score. Surfaces observations — not instructions — in the Weekly Architecture Session role pulse screen, the progress dashboard, and (for critically neglected roles only) a single weekly notification. Never interrupts the Launchpad or a focus session.
Horizon
A project's time frame: This week, This month, This quarter, or Someday. Shapes how frequently the project surfaces during planning. A This Week project is reviewed every Weekly Architecture Session. A Someday project is reviewed monthly.
I
Inbox
The single collection point for everything you capture. Nothing in the inbox is prioritized or categorized — it's a raw holding area. Items move out of the inbox only through processing, which routes each item to its correct destination: task, project, appointment, reminder, Someday, or trash.
Inbox processing
The act of reviewing captured items one at a time and routing each to its correct destination. Four paths: Task, Reminder, Appointment, Park it (Someday), Project, Trash it. Processing is where decisions are made. Capture is intentionally decision-free.
L
Launchpad
The main workday screen. Shows your MIT at the top, the time strip, and the Now list — the three to five tasks best matched to your current context. Everything else is hidden by design. The Launchpad is not a task list; it is a decision surface that answers the question: what do I do next?
Lead time
For recurring tasks: the number of days before a due date that the instance enters your active task pool and becomes visible in the Launchpad. Default is three days. A task with a lead time of seven days will surface a week before it's due, giving you time to complete it without urgency.
M
MIT
Most Important Task. The one task that, if completed, would make the day a success regardless of what else happens. Set during Morning Setup. Pinned to the top of the Launchpad. MIT completion rate — the percentage of days you complete your MIT — is tracked in the dashboard.
Morning Setup
The brief three-screen ritual that starts each workday. Step 1: energy check — sets the context filter for the entire day. Step 2: MIT confirmation — confirm or swap the suggested Most Important Task. Step 3: first task selection — pick what you'll start with from three options filtered to your energy. Takes two to three minutes.
N
Next commitment arc
The circular countdown indicator on the Launchpad showing time until your next calendar commitment. Green when you have plenty of time, shifting to red as the commitment approaches. Hidden when the next commitment is more than 12 hours away.
Now list
The three to five tasks shown below the MIT on the Launchpad — the highest-scoring tasks for your current context. Not a complete task list. Not a priority ranking. A curated set of the right tasks for right now. The "Got 5 min? 3 easy wins" banner at the bottom offers quick Admin tasks when you have a short gap.
P
Park it
Sends a captured item to Someday — acknowledged as real, but not timely. Available from the type selection screen during processing. No setup required. The item stays in Someday until you act on it or it returns to your inbox on a deferred date.
Processing queue
A holding state for items that have been acknowledged but not yet fully processed — for example, inbox items deferred during a re-engagement flow, or imported tasks that received only a role assignment during migration. Items in the processing queue are visible but have minimal metadata and will eventually need full processing.
R
Recovery mode
A streak mechanic that allows a broken streak to be partially restored. After a break in a Daily Close or MIT streak, completing three consecutive days activates Recovery, which marks the streak with a gap indicator rather than resetting it to zero. More honest than pretending the absence didn't happen; more forgiving than loss.
Re-engagement flow
The structured return sequence that fires when you open ATEND after an extended absence. Five steps: silent cleanup of expired data, inbox quick pass, triage of what still matters, fresh week setup (one or two outcomes), and today's MIT. Designed to make coming back feel like the right decision — never to confront you with accumulated failure.
S
Significant date
A date linked to a contact — birthday, anniversary, work anniversary, or other. Automatically generates a recurring annual reminder that fires seven days before the date with an action prompt: "Sarah's birthday is in 7 days. Want to add anything for it?"
Someday
A low-pressure holding area for ideas and commitments that are real but not yet timely. Completely segregated from the active task pool — Someday items never appear in the Launchpad. When you're ready to act on an item, a single swipe offers five options on a timeline axis: Now, Next week, Next month, Next quarter, or Never. Items deferred to a future date return to your inbox on that date and run through standard processing. There is no scheduled monthly review — Someday is quiet until you open it.
Sovereign time
Discretionary time that belongs to you — windows in the day not consumed by external commitments, appointments, or maintenance. Shown as the second bar on the time strip. The Launchpad summary "3 sovereign windows · 7.2 hrs yours" tells you how much of the day is genuinely available for self-directed work.
Stakes
A task field classifying what breaks if this task doesn't get done. Four values: Trivial (no real consequence), Functional (life gets harder gradually — things degrade without it), Relational (another person is directly affected; social consequences are immediate), Critical (something significant fails if this slips). Used by the fit score to increase urgency weighting as time runs short.
T
Task Iceberg
The principle that every task or appointment has a visible portion — the event itself — and a hidden portion that includes preparation time, travel there, travel back, and follow-up. Schedules collapse when planning accounts only for the visible tip. ATEND's buffer zones address the transition layers; the travel and follow-up layers require honest time accounting when creating the appointment. See Chapters 10 and 13 for the full treatment.
Who does this serve
A task field classifying who the task is ultimately for. Three values: Just me (self-directed work that primarily benefits you), Someone else (a commitment or obligation where another person is counting on you), Both (serves you and someone else). Used by the fit score to weight urgency — tasks serving someone else surface with additional priority — and by the coaching prompt when a task keeps getting deferred.
Time sovereignty
The concept that your sovereign time — the hours that genuinely belong to you — is finite and worth protecting deliberately. The Time Sovereignty screen in the Weekly Architecture Session shows next week's calendar exactly as it is, before you plan it, so you can see how much time is actually available for self-directed work.
Time strip
The two-bar visual at the top of the Launchpad that makes your day physically visible. The top bar shows your full day — committed events and elapsed time. The bottom bar shows only your sovereign windows. A single vertical line marks the current moment across both bars. Time labels run along the bottom from your workday start to end.
W
Weekly Architecture Session
The once-weekly planning ritual — typically Sunday evening — that sets the Big Three outcomes, reviews role health, ensures no life area is being neglected, and prepares the week ahead. Seven screens in sequence, plus a conditional eighth (Screen 4a: File the Unfiled) when more than five tasks have no role assigned. Takes 30–45 minutes when done consistently. The single most important ritual in the ATEND system. Tracked as a streak.
Who does this serve?
A prompt that fires when a task has been deferred three or more times. Asks you to honestly identify whether the task serves just you, someone else, or nobody anymore. The goal is not guilt — it's surfacing the underlying reason for avoidance so it can be addressed directly. "Nobody anymore" leads to a release with no judgment. "Someone else" prompts a check on whether the commitment is still real.
Quick reference
The daily and weekly rituals
Morning Setup
Each workday · 2–3 min
Energy check → MIT confirmation → First task selection. Sets the parameters that shape your entire day.
Daily Close
Each workday evening · 10 min
Review completions and deferrals → Set tomorrow's MIT → Final capture. Offloads open loops before sleep.
Someday Review
As needed — no fixed schedule
When ready to revisit: Now, Next week, Next month, Next quarter, or Never. Someday is quiet by default — no scheduled review prompt. Items return to your inbox on the date you chose, or stay parked until you choose to act.
Weekly Architecture
Weekly · 30–45 min
Retrospective → Inbox clear → Project review → Role pulse → File unfiled (conditional) → Big Three → Time sovereignty → Session complete. The most important ritual in the system.
Inbox processing paths
| Path | What it creates | Notes |
|---|---|---|
| A — Task | Active task | Four-screen setup: time + energy, weight, recurrence (optional), role. Enters active task pool and begins scoring. |
| B — Reminder | Calendar event (no time block) | Date, optional role, reminder offset, action prompt toggle. Does not consume sovereign time. |
| C — Appointment | Calendar event (time block) | Date, time, duration, role. Syncs to connected calendar. Appears on time strip. |
| D — Project | Project + optional first action | Lightweight inbox creation: outcome statement and role. Option to add first action immediately or return later. |
Items that aren't real are discarded at Step 0 triage ("Does this still matter?") before reaching path selection. Items you're not ready to classify can be sent to Someday from the type selection screen.
Fit score — what it weighs
| Factor | What it checks |
|---|---|
| Energy match | Does the task's energy requirement match your morning energy level? |
| Time fit | Does the task's estimate fit within your current sovereign window? |
| Stakes | Higher-stakes tasks (Relational, Critical) score higher when time is short; Trivial tasks score lower. |
| Time origin | Tasks serving just you get a boost during protected sovereign windows. |
| Big Three link | Tasks in a Big Three-linked project receive +25. |
| Deadline proximity | Tasks approaching a hard deadline receive a progressive boost. |
| Deferred count | Repeatedly deferred tasks receive a boost to surface them for resolution. |
Role neglect statuses
| Status | Meaning | Where it surfaces |
|---|---|---|
| Active | Healthy recent engagement | No special treatment |
| Quiet | Light but present activity | Soft indicator in WAS role pulse |
| Fading | No activity in 7–14 days | Amber flag in WAS role pulse; dashboard bar |
| Neglected | No activity in 15–29 days | Red flag in WAS role pulse; dashboard bar |
| Critically neglected | No activity in 30+ days | Warning card in WAS role pulse; one weekly notification |
The five day states
| State | When | What you see |
|---|---|---|
| Pre-workday | Before your workday starts | Good morning greeting, voice capture, inbox badge. No task list. |
| Morning Setup | Workday start, if not yet done | Three-step setup flow: energy check, MIT confirmation, first task selection. |
| Launchpad | During workday | MIT banner, time strip, Now list, coaching pulse. |
| Daily Close | Evening, after close is done | Results summary, tomorrow's MIT preview, final capture. |
| Late night | After your subjective day boundary | Clock display, late capture mode only. Tomorrow is already set. |
Core design principles
P1
Minimize visible complexity at all times
Default views show 3–7 items. The system filters — users cannot be trusted to filter their own view.
P2
Decide for the user wherever possible
Pre-select, pre-filter, pre-prioritize. Present choices only as a last resort — and then only 2–3 options.
P3
Asymmetric friction
Capture is nearly zero friction. Accessing overwhelming views requires deliberate navigation. The default path always leads to "what do I do next."
P4
Make time physical and visible
Countdown rings, progress arcs, and visual time bars supplement or replace clock times. The system assumes the user cannot feel time passing.
P5
The system reaches out
Passive systems fail ADHD users. The system pushes relevant information to the user at the right moment rather than waiting to be checked.
P6
Never punish falling off
After an absence, handle the backlog gracefully, present a path forward, and never confront the user with accumulated failure. Streaks are recoverable.
P7
Speak like a coach
Warm, direct, encouraging — not corporate. Error states, empty states, and prompts all follow this voice.
P8
Context is everything
The same task list is not appropriate at all times. The system knows what context the user is in and filters accordingly.