What is a "Claim"?

The basic unit of interaction on Whysaurus is the Claim. Each claim makes a statement about the world.

Claims can be:

  • Supported or countered by other claims
  • Rated with agree/disagree
  • Re-used (along with their networks of evidence!)

A collection of supporting and counter-claims are called Evidence.

A claim with evidence is an Argument.

What are Sources in Whysaurus?

Each claim can link to sources. But remember — IRL and on Whysaurus — arguments work best when you explain the best claims made by your sources.

“Because they said so” doesn’t work when we don’t agree on who to trust. Plus, even the best sources are wrong sometimes.

The good news is that we can all get behind “show me the evidence!”

I want to state a “fact” — how do I do that in Whysaurus?

A fact is a claim with real strong evidence. So make your claim and back it up! No claims are sacred cows — we must protect the right to argue against any claim.

How are Claims scored in Whysaurus?

A claim’s score is based on:

  • The scores of its Evidence (multiplied by Relevance)
  • Agrees and Disagrees
  • Sources

Supporting evidence makes scores go up; counter evidence makes scores go down. The algorithm uses "Bayesian Thinking", and it's a work-in-progress. Currently it only “looks” one level down. In the future, it will be possible to defend a claim by attacking its counter points!

Also, importantly, if a claim has many "agrees" but weak evidence, each additional agree will have a smaller and smaller impact on the claim's score. So a swarm of upvotes won't significantly change a claim’s score, but well supported counter-evidence will.

And yes, we've got plans for trolls.

What is Relevance and how’s it different from Agree/Disagree?

Relevance is one of the most powerful weapons in Whysaurus.

Claims are rated with agree/disagree — and the links between claims are rated with Relevance.

If someone claims “Tom Brady is a cheater” and supports it with “Star Wars stole ideas from Isaac Asimov’s Foundation series”, you might agree with both, but rate the relevance as 0%.

Claims with low relevance are hidden by default.

How does Relevance impact scores?

Relevance impacts scoring dramatically. For example:

a +10 Supporting claim with 20% Relevance adds only +2 to the claim above it (10 × 20% = 2).

When a piece of evidence is 0% Relevant, it has no effect on scores.

How is Evidence sorted?

By Relevance. So the most relevant claims rise to the top. In the case of ties, we look at scores.

Is "Agree", "Disagree" and "Relevance" voting anonymous?

Yes. This way, peer pressure from friends, family, employers and governments can't influence votes. We keep count so you can't vote twice.

What if the score of a claim is wrong?

Set the record straight! Add evidence or counter evidence. Vote on relevance and click agree/ disagree. Share the argument with friends and ask them to do the same!

Why are claims reusable and why is Whysaurus collaborative?

So you can build good arguments faster!


  • 1. Gus argues “No one ever died from overdosing on marijuana” and adds evidence.
  • 2. Later, Susannah argues “Marijuana is safer than alcohol” and links to Gus’s claim.
  • 3. All of Gus’ supporting evidence comes in automatically.

Why should we start from scratch when we can stand on the shoulders of the best work done so far?

How do Comments work in Whysaurus?

Comments are where we discuss how to make each claim better.

Click the little speech bubble icon. That opens the Comments area, where you can work with other users to hammer out the best phrasing for each claim.

If you don’t agree with how other users are framing a discussion, start your own argument!

Oops I messed up! Can I delete my claim?

If you created a claim and it has no other contributors, you can delete it. Click the icon and the delete button should be in the menu. Once multiple contributors are involved, only admins can delete. If you see a claim you think should be trashed, contact us.

What About The Trolls!?

Why do you think this will work?

Research shows that when you ask people to explain a problem in detail, they take more moderate positions than they would otherwise. The act of explanation shatters The Illusion Of Explanatory Depth. In other words, arguing with evidence on Whysaurus often reveals to us that we understand less than we thought we did! In practice, we’ve found that asking for evidence makes folks argue in good faith.

Why should I trust what I read on Whysaurus?

There’s a “See Evidence” button on every claim! Be skeptical on Whysaurus, online and IRL. (But on Whysaurus at least the evidence — or warning that there’s none — is always close at hand).

What about swarms of nutters upvoting stuff that isn’t true?

If a claim has many "agrees" but weak evidence, each additional agree will have a smaller and smaller impact on the claim's score. So a swarm of upvotes won't significantly change a claim’s score, but well supported counter-evidence will. Also, users will be able to apply a "Needs Evidence" flag, clearly identifying baseless claims.

What if the nutters add evidence?

For those that do: if their evidence deserves to be flagged too, it will be; and if their evidence is valid, it’s worth arguing. Maybe they’ll learn something from the experience — and maybe you will too!

Why should I trust the crowd? Why should I trust moderators?

Wisdom-of-the-Crowds projects have proven track-records of accuracy and self-moderation — see Wikipedia, StackOverflow, Quora and Genius (all projects we admire!)

What about harassment and personal attacks?

Unlike Twitter, claims about users are not allowed (and these are easy to identify).

What about hate? / What about free speech?

Hateful activities are not allowed. We like the idea of allowing all speech — and using Whysaurus to rebut vile claims — but we don’t yet have the resources to moderate those kinds of conversations. Until we have bags of money, our moderators will do their best to protect arguments made in good faith while also flagging hateful activities. We use The Southern Poverty Law Center's definition of "hateful activities": “activities that incite or engage in violence, intimidation, harassment, threats, or defamation targeting an individual or group based on their actual or perceived race, color, religion, national origin, ethnicity, immigration status, gender, gender identity, sexual orientation, or disability.”

Ultimately, we hope to offer a format in which you can respond productively to claims you disagree with!

Best Practices

Community rules

  • This is a civil community — be respectful.
  • No claims about other users.
  • Have fun!

Is it cool to click “agree” or “disagree” on my own claims?

We think so… check out our on-going argument about this.

Divide big ideas into multiple bite-sized claims

Make it easy for everyone to agree with the agreeable parts of your argument, and zero in on the juicy stuff — the places where there’s conflict. Plus, your argument score will grow higher faster with many short claims than it would with a few bloated ones.

Be Specific

Make sure everyone knows what you mean.

Use simple words

Only use big words when shorter synonyms don’t exist.

Be Concise

Life is short.

Write claims so they can be re-used in multiple arguments (when you can)

Certain fundamental claims contribute to many big ideas. If your claim is written to "stand alone" — i.e. it doesn't require the context of other claims to be understood — then other users can link to it in their own arguments.

Contest semantics only when a user's intent is truly unclear

When a grammatical error or hasty word choice prevents a clear interpretation of a user’s intent, use the comments section to iron it out. But if the intent of a claim can be understood by a reasonable person, there's no need to call in the semantics police.


I’ve got an idea that could make Whysaurus better, how can I share it with you?

Contact us here. Bonus points if you create a Whysaurus argument to back-up your suggestion!

What's up with with dinosaurs?

Dinosaurs are awesome.