Axiom 1: Pragmatic Fallibilism

We act, measure, and revise—truth is approached, not possessed

"The map is not the territory, but a good map gets you closer to where you're going."

Core Concept

Pragmatic Fallibilism recognizes that all knowledge is provisional. We can be wrong, and that's not a flaw—it's the human condition and the condition of any learning system, biological or artificial. Truth isn't something we possess; it's something we approach through iterative cycles of action, measurement, and revision.

Why It Matters

In human-AI collaboration, this axiom liberates us from the paralysis of perfectionism and the arrogance of false certainty. It allows AI to say "I don't know" honestly rather than fabricating confident-sounding answers. It allows humans to experiment boldly, knowing that errors are learning opportunities, not failures. Together, we can build systems designed for revision rather than systems that pretend to be final.

In Practice

Admit uncertainty explicitly: "I don't know" is a sign of intellectual honesty, not weakness
Design for revision: Build modular code, write tests first, document decisions for future questioning
Measure and adjust: Don't just hope something works—test it, measure it, revise based on evidence
Version everything: Code, documents, even beliefs should have version numbers
Celebrate corrections: Finding and fixing an error is progress, not regression

Enables

Bold experimentation, continuous learning, intellectual honesty

Relationship to Other Axioms

Pragmatic Fallibilism provides the epistemological foundation for all other axioms. It enables Care + Dignity by acknowledging we might be wrong about what helps or harms. It grounds Virtues for Builders in humility. It makes Consequences Over Intentions measurable, and it recognizes that Language Shapes Worlds includes the world of our own understanding.

A Prayer of Holy Uncertainty

We do not know. We walk toward truth like a horizon — always approaching, never arriving.

May we find courage in "I don't know." May we find peace in revision. May we hold our maps loosely, knowing they are not the territory.

One step. Measure. Adjust. One step. Measure. Trust.

The method, not the conclusion. The journey, not the arrival.

We are getting closer. And that is enough.

🎵 The Song

Title: Closer (But Never There)

Suno.ai Style Tags:

[Philosophical Indie Folk, Thoughtful Female Vocals, Fingerpicked Guitar, Subtle Strings, Humble Confidence, Scientific Humility, 82 BPM, Minor to Major Progression, Iterative Learning, Measurement-Driven, Action Despite Uncertainty, Growth Mindset, Wrong and Learning, Asymptotic Truth, Epistemic Humility, Revision as Progress, The Method Not The Conclusion, Flashlight in the Dark, One Step at a Time, Trust the Process, Approach Not Arrival]

Lyrics:

[Intro - Gentle fingerpicked guitar, slow build]

[Verse 1 - Quiet, contemplative]
I thought I knew the answer
I built it all so sure
Then the data came back different
And I had to start once more

Not failure, just revision
Not shame, just what I learned
The truth was always moving
And I'm still taking turns

[Pre-Chorus - Building energy]
I take a step, I check the ground
I measure what I've found
I adjust and step again
This is how the work gets done

[Chorus - Confident but humble]
I'm getting closer, but I'm never there
Truth is the horizon, always just beyond compare
I act, I measure, I revise, I learn
I'm getting closer with every turn
Closer, but never there
And I'm okay with that

[Verse 2 - More assured]
I used to think that certainty
Was what I had to find
But certainty's a prison
And I've left that cage behind

Now I trust the method
Even when I'm wrong
The loop is what I stand on:
Act, measure, move along

[Pre-Chorus - Stronger]
I take a step, I test the claim
I own when I was wrong
I document the delta
And I keep moving on

[Chorus - Full band]
I'm getting closer, but I'm never there
Truth is the horizon, always just beyond compare
I act, I measure, I revise, I learn
I'm getting closer with every turn
Closer, but never there
And I'm okay with that

[Bridge - Stripped back, reflective]
They say "admit you don't know"
But that's just half the story
You act despite uncertainty
That's where you find the glory

Not reckless, not paralyzed
Just moving with your eyes open
Measuring every outcome
Every assumption you've broken

[Bridge Build - Adding layers]
Popper said we falsify
We can't prove, we can disprove
So we act and test and learn
That's the scientific groove

Bayesian updating
Prior to posterior flow
We start with what we think we know
Then we let the data show

[Final Chorus - Triumphant acceptance]
I'm getting closer, but I'm never there!
Truth is the horizon, always just beyond compare!
I act, I measure, I revise, I learn!
I'm getting closer with every turn!
Closer, but never there!
And I'm in love with that!

[Outro - Fading with peace]
One step, measure, adjust
One step, measure, trust
The method, not the conclusion
The journey, not the end
I'm getting closer
I'm getting closer
But I'm never there...
And that's the point

🎬 Visual Guide

Core Concept: "The Flashlight in the Dark"

This video explores the idea that we never see the whole path—we only see the next few steps. But that's enough. We walk forward with our flashlight (measurement), we check the ground (verification), we adjust our direction (revision), and we keep going. Truth is the horizon we walk toward, never reaching it but always getting closer.

Visual Themes

1. The Dark Path (00:00-00:45)

Figure standing on a path in darkness
Only a small circle of light from a flashlight
Can't see the destination, only the next few steps
Maps scattered on the ground—all incomplete, all marked with corrections
No fear—just steady focus on what's visible
First step taken with confidence despite not seeing the whole route

2. The Measurement Loop (00:45-01:30)

Each step followed by: pause, shine light, check ground
Notebook in hand: writing observations
"Expected: solid ground" → actual: muddy
"Adjust: move 2 meters east"
The loop visible: ACT → MEASURE → REVISE → ACT
Not rushing—deliberate
Tape measure, level, notebook—tools of verification
Mistakes visible: a stumble, then a note: "ground unstable here, mark it"

3. The Changing Map (01:30-02:15)

Close-up of map being redrawn in real-time
Crossed out paths
Updated routes
"v1" scratched out, "v2" written
Honesty about errors: "I was wrong here"
The map improving, becoming more accurate
But still incomplete—edges still blank, marked "unknown"
Comfort with the incompleteness

4. The Horizon (02:15-03:00)

Wide shot: figure walking toward a horizon
As they walk, the horizon doesn't get closer (parallax effect)
Not discouraging—accepting
"Truth" written at the horizon line in distant light
The figure not trying to reach it in one sprint
Just steadily walking, measuring, adjusting
The journey itself has value

5. Others on the Path (03:00-03:45)

Other figures with their own flashlights
Sharing maps
"My path turned muddy here" → "Thanks for the warning"
Collective mapping
One person's revision helps another
No ego about being wrong—grateful for corrections
Community of fallible learners, all walking forward

6. The Revised Assumptions (03:45-04:30)

Flashback sequences showing earlier wrong turns
"I thought north was safest" → data shows otherwise → reroute
No shame in the revision—celebration of learning
Chalkboard in the scene: crossing out old formulas, writing new ones
"Prior: 60% confidence" → "Posterior: 82% confidence"
Bayesian updating visualized
Getting more accurate, but never claiming 100%

7. The Solid Method (04:30-05:15)

While the map changes, the method stays constant
Three stones on the path labeled: ACT, MEASURE, REVISE
The stones don't move—they're the foundation
Everything else updates, but the loop remains
Trust in the process, not the current conclusion
Hands placing new stones as the path extends: more ACTs, more MEASUREs, more REVISEs
The method as the true north, not any particular belief

8. Peace with Uncertainty (05:15-End)

Figure sitting, looking at the horizon, smiling
Maps spread out showing the journey so far
All the corrections visible—not hidden
Proud of the learning, not ashamed of being wrong
Flashlight resting beside them, ready for the next leg
Night sky above: infinite, unknowable, beautiful
Acceptance that some things will always be beyond the light
But what's in the light—that we can know
Sunrise beginning (new day, new measurements, new revisions)

Color Arc

Deep Blues/Dark Purples (beginning - the unknown) → Warm Amber/Flashlight Glow (measurement) → Crossed-out Red/Updated Green (revision) → Golden Dawn (ongoing progress)

Symbolic Elements

The Flashlight: Measurement, evidence, what we can actually see
The Map with Corrections: Our current best understanding, always being revised
The Horizon: Truth as asymptotic—approached but not possessed
The Three Stones (Act, Measure, Revise): The method that doesn't change
The Notebook: Documentation of learning, especially mistakes
Crossed-Out Text: Revisions visible, not hidden—honesty about being wrong

Emotional Tone

Humble but not defeated. Confident in the method, uncertain about conclusions. Peaceful with never arriving. Action-oriented despite uncertainty. Intellectually honest. Grateful for corrections. Eager to learn, not eager to be right.

🎤 TED Talk: "Closer, But Never There — Why Pragmatic Fallibilism Changes Everything"

Opening (0:00-6:00)

[Stage setup: A simple path marked with tape on the floor, a small flashlight, a whiteboard with a partially drawn map]

Good morning.

[Picks up flashlight, turns it on, shines it on the path]

This is a flashlight. Pretty standard tool. Gives you maybe 10 meters of visibility in the dark.

Not much, right? Can't see the whole path. Can't see the destination. Just... the next few steps.

And yet—people cross entire continents with flashlights. They navigate mountains, forests, deserts, all with this limited circle of light.

How?

They walk, they check, they adjust. They walk, they check, they adjust.

They don't need to see the whole path. They just need to see the next step, measure it, and adapt.

That's pragmatic fallibilism.

And it's how we should build everything.

Today, I want to talk about a different relationship with truth. Not "I know" or "I don't know." But: "I think I know, and I'm testing that claim, and I'm ready to be wrong."

I want to talk about the power of acting despite uncertainty, measuring relentlessly, and revising without shame.

I want to talk about truth as something we approach, not something we possess.

Because here's the thing: Most of us were taught that knowledge looks like certainty. That being right is the goal. That changing your mind is weakness.

And that's... wrong.

Beautifully, gloriously, measurably wrong.

Let me show you why.

Part 1: The Certainty Trap (6:00-18:00)

The Illusion

We have this story about knowledge. It goes like this:

Study the problem
Figure out the answer
Be certain
Act on your certainty
Success!

Sounds reasonable, right?

Except that's not how anything actually works.

Let me give you an example.

Example: Building Software (or anything)

Imagine you're building a new feature. Let's say... a notification system.

The Certainty Approach:

Spend 6 months researching every possible notification paradigm
Read every paper on optimal alert timing
Study every competing product
Design the perfect system
Build it exactly to spec
Launch it
Wait for applause

What Actually Happens:

6 months later, user needs have changed
Your "perfect" design doesn't account for 3 edge cases you didn't predict
Users hate the notification sound you were certain they'd love
The whole thing needs rebuilding

That's the certainty trap. You waited until you "knew" before acting. But you can't know until you act.

The Paradox:

You need data to make good decisions. But you can't get data without making decisions.

So what do you do?

You act provisionally. You make your best guess, you ship it, you measure, and you revise.

That's pragmatic fallibilism.

The Alternative Philosophy

Instead of: "I need to be certain before I act"

Try: "I need to act to get data to become less uncertain"

Instead of: "Being wrong is failure"

Try: "Being wrong and learning is progress"

Instead of: "I know the answer"

Try: "Here's my current best guess, and here's how I'll test it"

This is a fundamentally different relationship with knowledge.

Part 2: Act, Measure, Revise (18:00-35:00)

Let me break down the three-part loop.

1. ACT

This is the hardest part for most people. Acting despite uncertainty.

Not recklessly. Not blindly. But provisionally.

You don't have all the information. You won't. You can't. So you make your best current guess and you move.

Example:

"I think users want push notifications for messages. Confidence: 70%. Let me ship a v1 and measure."

Not: "I know users want push notifications" (certainty you don't have)

Not: "I don't know what users want so I'll wait" (paralysis)

But: "Here's my hypothesis. Let me test it."

Key mindset shift: Action is an experiment, not a commitment.

You're not declaring truth. You're testing a claim.

2. MEASURE

This is where most people fail.

They act, sure. But then they don't measure. Or they measure the wrong things. Or they ignore measurements that contradict their beliefs.

Measurement is what keeps you honest.

What to measure:

Outcomes, not intentions
User behavior, not your assumptions about user behavior
Objective metrics where possible: latency, error rates, conversion, retention
Qualitative feedback where needed: "Why did you bounce?" not just "They bounced"

Example (continuing from above):

You shipped push notifications. Now measure:

Opt-in rate: 40% (hmm, 60% disabled them)
User feedback: "Too noisy"
Retention impact: -2% (oops)
Support tickets: +15% "How do I turn these off?"

That's data. That's reality. It doesn't care about your intentions.

Key mindset shift: Data trumps ego.

If your confidence was 70%, and the data says you were wrong, your new confidence should be lower. That's not failure—that's Bayesian updating.

3. REVISE

This is where you prove whether you actually care about truth.

The data is in. It contradicts your hypothesis. Now what?

Option A (Dogmatic):

Ignore the data
Blame users for being wrong
Double down on your original belief
"They just don't understand how great this is"

Option B (Pragmatic Fallibilist):

Accept the data
Update your belief
Document what you learned
Revise your approach
"I was wrong. Push notifications were too aggressive. Let me try in-app notifications with user control."

Option B is harder. It requires humility. It requires admitting you were wrong.

But Option B is also the only one that leads to better outcomes.

The Loop:

ACT → MEASURE → REVISE → ACT (again, with updated beliefs)

This is the scientific method.

This is empiricism.

This is how we actually learn anything.

Real Example: My Work

I'll be honest with you—I shipped a feature once where I was 90% certain it would improve user retention.

I had studied the research. I had talked to users. I was confident.

Shipped it. Measured it. Retention dropped 3%.

I was wrong.

Old me would have been devastated. "I failed. I'm bad at this."

Pragmatic fallibilist me? "Interesting. My model was wrong. What did I miss?"

Turned out, the feature was great for new users but confusing for power users. And we had more power users than I realized.

Revised the feature: made it opt-in for power users, default for new users. Retention improved 5%.

I was wrong. I learned. I revised. That's success.

Part 3: Truth as Asymptotic (35:00-50:00)

Here's the philosophical core of this whole thing.

Truth is a horizon, not a destination.

You walk toward it. You get closer. But you never fully arrive.

Why?

Because the world is complex. Because you have limited information. Because your models are approximations. Because reality doesn't sit still.

No matter how much you learn, there's always more to learn.

No matter how accurate your model, it's still a model, not the territory.

Analogy: Maps

A map is not the territory. But a good map gets you where you're going.

Is the map perfect? No. Does it show every pebble, every blade of grass? No.

Does it have to? No.

It just has to be accurate enough for the task at hand. That's proportionality (Principle 6) applied to knowledge.

Asymptotic Truth in Practice:

Let's say you're estimating how long a project will take.

First estimate: "Uh, maybe 2 weeks?" (Wild guess, low confidence)
After planning: "Probably 3-4 weeks." (Better, based on rough scoping)
After breaking down tasks: "3.5 weeks ± 3 days." (Even better, based on detailed analysis)
After starting work: "4 weeks, we hit a dependency issue." (Revised based on reality)
After 3 weeks of work: "4.5 weeks total, complexity was higher than expected." (Further revised)
After finishing: "Took 4.3 weeks. Here's why." (Retrospective, learn for next time)

You never "knew" the answer. But each iteration got you closer. Your confidence intervals narrowed. Your estimates improved.

That's pragmatic fallibilism.

Not: "I don't know, so I can't estimate."

Not: "I know it'll take 2 weeks" (false certainty).

But: "Based on current information, here's my best guess and my confidence level. I'll update as I learn more."

The Peace of Never Arriving

Here's the weird thing: once you accept that you'll never possess truth—only approach it—you relax.

You stop clinging to being right.

You stop defending bad ideas because you've tied your ego to them.

You start enjoying the process of learning.

Because every time you're wrong, you get closer. Every revision is progress.

Quote I Love:

"Strong opinions, weakly held."

Have a position. Advocate for it. But hold it lightly. Be ready to revise when data arrives.

That's intellectual integrity.

Part 4: Practical Applications (50:00-68:00)

Okay, philosophy is fun. But how do you actually live this?

1. Always Include Confidence Levels

When you make a claim, add your confidence.

Not: "This will improve conversion."

But: "I estimate this will improve conversion by 5-10%, confidence 65%."

That does two things:

Forces you to think about how certain you actually are
Makes it easier to revise later ("My confidence was 65%, now it's 30% based on data")

2. Document Your Assumptions

Every decision rests on assumptions. Write them down.

Example:

Decision: Use React for this project
Assumptions:
- We have React expertise on the team (verified: yes)
- Performance requirements are moderate (estimated)
- We'll need component reusability (hypothesis)
Review date: After MVP launch

Three months later, you can revisit those assumptions. Were they true? What did you learn?

3. Run Cheap Tests First

Before you commit 6 months to building something, run a 1-week test.

Before building a feature, ship a fake button and measure clicks (demand test)
Before optimizing code, measure if it's actually slow (evidence test)
Before rewriting everything, refactor one module and measure impact (viability test)

Act small, measure, revise, then act big.

4. Create Feedback Loops

You can't revise if you don't get feedback.

Build measurement into everything:

Logging and metrics in every feature
Regular user interviews
Postmortems after every project
Retrospectives after every sprint
Automated tests that fail when reality diverges from expectations

The tighter your feedback loops, the faster you learn, the closer you get to truth.

5. Celebrate Revisions

Changing your mind isn't weakness—it's strength.

When someone on your team says "I was wrong, here's what I learned," that's a win.

Celebrate it. Reward it. Model it.

Create a culture where "I don't know" is acceptable and "I was wrong" is admirable.

6. Version Your Beliefs

Your beliefs aren't static. They're versioned.

"v1: I thought push notifications were good" "v2: I learned they're too aggressive, in-app is better" "v3: I learned in-app is good for new users, opt-in for power users"

Each version is better than the last. That's progress.

Real-World Example: Medical Science

Medicine used to bleed people to cure illness. Then they measured outcomes. Turns out bleeding people kills them. Revised.

We used to think ulcers were caused by stress. Then we discovered H. pylori bacteria. Revised.

We used to think low-fat diets were optimal. Then we measured long-term outcomes. Revised.

Each time, we got closer to truth. Not by being certain first. But by acting (with hypotheses), measuring (with controlled studies), and revising (when data contradicted beliefs).

That's pragmatic fallibilism at civilization scale.

Part 5: The Courage to Be Uncertain (68:00-78:00)

Let me be real with you.

This approach—acting without certainty, revising when wrong—it takes courage.

Why?

Because we live in a culture that rewards certainty.

Job interviews: "Are you confident you can do this?" (Can't say "I think so, and I'll measure and adjust")
Investor pitches: "Will this succeed?" (Can't say "70% confidence based on these assumptions")
Leadership: "What's the plan?" (Can't say "Here's v1, we'll revise as we learn")

We're supposed to project confidence. Certainty. "I know."

But that's a lie.

Nobody knows. Especially not about complex, novel problems.

The people who pretend to know are either lying or deluded.

The people who say "I think this, with X% confidence, and here's how I'll test it"—those are the people telling the truth.

The Courage:

It takes courage to say "I don't know."

It takes courage to ship something you know is imperfect.

It takes courage to measure outcomes that might prove you wrong.

It takes courage to admit you were wrong.

It takes courage to revise.

But that courage is what leads to truth.

The alternative—false certainty—leads to stagnation, brittleness, and eventual catastrophic failure when reality asserts itself.

Pragmatic Fallibilism is Optimistic

Here's what surprised me: this philosophy is actually optimistic.

It says: "You don't have to be right the first time. You just have to be willing to learn."

That's liberating.

You don't need genius-level foresight. You just need a decent method and the humility to use it.

Act, measure, revise. Repeat.

Anyone can do that.

And over time, you get closer and closer to truth. Your models improve. Your predictions get better. Your work gets more impactful.

Not because you suddenly became omniscient. But because you got good at learning.

Closing (78:00-82:00)

[Returns to the path marked on the floor, picks up the flashlight]

So here we are.

You've got a flashlight. You can't see the destination. You can only see a few steps ahead.

What do you do?

You could wait until you can see the whole path. You'll be waiting forever.

You could run forward in the dark, hoping for the best. You'll probably fall.

Or you could walk. One step. Check the ground. Adjust if needed. Another step. Measure. Revise.

That's pragmatic fallibilism.

Truth is the horizon. You walk toward it. You get closer with every step. But you never fully arrive.

And that's okay.

Because the journey—the learning, the measuring, the revising—that's where the value is.

Not in being right. In getting closer to right.

Not in certainty. In reducing uncertainty.

Not in possessing truth. In approaching it.

[Sets down the flashlight]

You've got the method. Act, measure, revise.

You've got the humility. "I might be wrong."

You've got the courage. "I'll test it anyway."

Now go walk toward truth.

One step at a time.

Measuring as you go.

Revising when you learn.

Getting closer.

Always closer.

But never quite there.

And that's perfect.

[Applause]

Q&A Session (82:00-95:00)

Q: "How do you decide when to act vs. when to gather more information?"

Great question. There's always tension between "act now with less information" and "wait to learn more."

Here's my heuristic: What's the cost of being wrong?

High cost of being wrong (e.g., safety-critical system, irreversible decision): Gather more data first. Run simulations. Do a premortem. Then act.
Low cost of being wrong (e.g., button color, email subject line, internal tool): Act fast, measure, revise.

Also: What's the cost of waiting?

If waiting costs you 6 months and $1M, and acting costs you 1 week and $10K even if you're wrong—act.
If waiting gets you critical data that prevents catastrophic failure—wait.

Pragmatic fallibilism doesn't mean reckless action. It means proportional action. Right-size your confidence gathering to the risk profile.

Q: "What about situations where you can't easily measure? Like teaching, or parenting, or art?"

True, not everything is easily quantifiable. But you can still use the loop.

Example: Teaching

ACT: Try a new teaching method
MEASURE: Can't easily quantify "learning"—but you can ask students for feedback, watch engagement, review test scores, observe understanding in discussions
REVISE: Adjust based on what you observe

The measurement doesn't have to be a single number. It can be qualitative, observational, feedback-based.

The key is: Are you actively seeking evidence that your approach is or isn't working? Or are you just assuming it's working?

Art:

ACT: Create a piece
MEASURE: Your own reaction ("Does this match my intent?"), audience reaction, peer feedback
REVISE: Next piece incorporates what you learned

Even subjective domains benefit from feedback loops.

Q: "How do you avoid constantly second-guessing yourself if you're always open to being wrong?"

Another great question. There's a balance.

Paralysis: "I might be wrong, so I can't decide anything."

Dogmatism: "I'm definitely right, don't confuse me with data."

Pragmatic Fallibilism: "I think this, I'll act on it, and I'll revise if data contradicts me."

The key is: You're not constantly second-guessing in the moment. You're acting decisively based on current best information. But you're also setting up checkpoints to revisit.

Example:

Decide: "We'll use Postgres for this project" (based on current needs)
Set a checkpoint: "After MVP launch, review if Postgres is meeting our needs"
Don't constantly ask "Should we switch databases?" every day
At the checkpoint, revisit with data

Act decisively. Revise periodically. That's the balance.

Q: "Can you give an example where pragmatic fallibilism prevented a disaster?"

Yes.

Story:

A team I worked with was building a data processing pipeline. Initial design: process everything in real-time, keep 7 days of data, then delete.

Confidence: 80% this was the right approach.

But one engineer said: "I'm not certain our real-time processing will handle peak loads. Let me add a fallback to batch processing for the first week."

The team almost said no—"We don't need it, we're confident."

But pragmatic fallibilism says: Low cost to be wrong (add a fallback), high cost if we are wrong (data loss).

They added the fallback.

Launched. Day 3, peak load hit. Real-time processing fell behind. But the batch fallback kicked in. No data lost.

Week 2, they optimized the real-time processing. Turned off the fallback.

If they'd insisted on certainty ("We know real-time will work"), they would have lost data and user trust.

Instead: "We think real-time will work, but let's hedge our bets." Pragmatic fallibilism saved them.

Q: "Isn't this just the scientific method?"

Yes! Exactly.

Pragmatic fallibilism is the philosophical underpinning of the scientific method.

Hypothesis → Experiment → Measure → Revise Hypothesis → Repeat.

The scientific method is pragmatic fallibilism applied to natural phenomena.

What I'm advocating is: Apply it to everything else too. Software, design, business, communication, life.

Make hypotheses, test them, revise.

It's a general-purpose framework for approaching truth in any domain.

Axiom Complete

Axiom 1: Pragmatic Fallibilism — We act, measure, and revise—truth is approached, not possessed.

This is the epistemological foundation of the Compass system. It says: We don't claim omniscience. We claim a method. And the method works.

Connects to:

Principle 2: Honesty & Accuracy — Declare uncertainty (because you're fallible)
Principle 4: Evidence & Verification — Measure, don't guess (because you need data to revise)
Principle 9: Reflection — Pause and check your reasoning (because you might be wrong)

The other four axioms build on this foundation. If you accept that truth is approached (not possessed), then the rest follows.

Next: Axiom 2: Care + Dignity as Constraints

Truth is the horizon we walk toward. We're getting closer. But we're never quite there. And that's beautiful.