Ranking Webpages by Argument Quality, Not Ad Revenue

The problem with Google: Their algorithm optimizes for clicks and ad revenue, not truth. Engaging misinformation ranks higher than boring accuracy.

The ISE solution: Rank webpages by evidence quality, logical coherence, and argument strength. Make the scoring transparent. Let users debate when they disagree.

---

Why Google's Algorithm Fails for Truth-Seeking

What Google Optimizes For

Not truth. Revenue.

• Engagement (clicks, time on site, shares)
• Ad revenue potential
• SEO gaming success
• Viral spread

The result:

• Climate denial blogs outrank peer-reviewed studies (more engaging)
• Conspiracy theories beat scientific consensus (more clicks)
• Outrage journalism beats investigative reporting (more shares)
• "Secrets they don't want you to know" beats "boring scientific consensus"

What Google Doesn't Tell You

Their algorithm is secret because:

• If you knew the formula, you'd game it (SEO industry proves this)
• Transparency would reveal profit motive over truth
• No accountability for harmful rankings
• No way to contest when good sources get buried

When something ranks lower:

• No explanation given
• No appeal process
• No transparency about why
• Silent demotion, no recourse

---

How the ISE Ranks Webpages Differently

Open Scoring on Five Criteria (0-100 scale)

1. Evidence Quality

• Tier 1 sources (peer-reviewed studies, official data): High score
• Tier 4 sources (blog posts, anecdotes): Low score
• Citations to primary sources: Bonus points
• Cherry-picking detected: Penalty

2. Logical Coherence

• Internally consistent arguments: High score
• Logical fallacies detected: Score reduced
• Conclusions follow from evidence: High score
• Non sequiturs and leaps: Penalty

3. Relevance to Belief

• Directly addresses the claim: High score
• Tangentially related: Lower score
• Linkage strength matters (see Linkage Scores)

4. Position Clarity

• Clear about supporting or opposing belief: High score
• Ambiguous or hedged: Lower score
• Steel-mans opposition: Bonus points
• Straw-mans opposition: Penalty

5. Cultural Influence

• How much this webpage actually affects belief adoption
• Measured by: Citations in other arguments, shares weighted by quality, impact on downstream beliefs
• NOT measured by: Raw clicks, engagement bait, viral outrage

---

The Key Difference: Transparency + Debate

When Users Disagree on a Webpage's Score

Google's approach:

• Algorithm decides
• No explanation
• No appeal
• Done

ISE approach:

1. Disagreement gets flagged (when scores vary by >30 points)
2. Structured debate opens:
• User A: "This webpage cherry-picks evidence" [provides examples]
• User B: "No, it addresses the main studies" [links to coverage]
• Both must provide reasoning, not just votes
3. Community evaluates the debate:
• Which side has better evidence?
• Which reasoning is stronger?
• Scores adjust based on argument strength
4. If still unresolved:
• Trigger jury vote (mix of users and domain experts)
• 60% threshold required
• All reasoning public and logged

Result: Scores aren't arbitrary. They're justified. And contestable.

---

Example: Climate Change Webpages

Belief: "Governments should take aggressive action on climate change"

High-Scoring Supporting Sources

IPCC Assessment Reports

• Evidence Quality: 95/100 (Tier 1: Peer-reviewed, thousands of studies)
• Logical Coherence: 90/100 (Systematic methodology, clear conclusions)
• Relevance: 95/100 (Directly addresses policy implications)
• Cultural Influence: 85/100 (Cited in policy worldwide)
• Overall: Strongly supports, high confidence

Britannica Climate Overview

• Evidence Quality: 85/100 (Synthesizes peer-reviewed consensus)
• Logical Coherence: 90/100 (Clear, systematic presentation)
• Relevance: 80/100 (Educational, covers mechanisms)
• Overall: Supports, educational foundation

The Guardian Climate Coverage

• Evidence Quality: 75/100 (Mix Tier 1-2: Journalism citing studies)
• Logical Coherence: 80/100 (Generally sound)
• Relevance: 85/100 (Policy-focused)
• Cultural Influence: 80/100 (Wide readership, frequently cited)
• Overall: Supports with strong reach

Low-Scoring Opposition Sources

Heartland Institute Reports

• Evidence Quality: 25/100 (Tier 4: Non-peer-reviewed, funded by fossil fuel interests)
• Logical Coherence: 40/100 (Cherry-picks outlier studies, ignores consensus)
• Relevance: 60/100 (Addresses topic but misrepresents science)
• Cultural Influence: 45/100 (Some reach but declining credibility)
• Overall: Opposes, but weak evidence and logical flaws
• Tagged: "Contested - Cherry-picking detected"

WattsUpWithThat Blog

• Evidence Quality: 20/100 (Tier 4: Blog posts, no peer review)
• Logical Coherence: 35/100 (Frequent logical fallacies detected by community)
• Relevance: 55/100 (Discusses climate but often straw-mans positions)
• Cultural Influence: 60/100 (High traffic but low citation by quality sources)
• Overall: Opposes, but poor reasoning
• Tagged: "High traffic, weak logic - See debate thread"

---

How This Integrates with ISE Features

Connection to Evidence Tiers

From Evidence Quality Framework:

Webpages classified by source tier:

• Tier 1 webpage (peer-reviewed journal, official data) → Automatic high evidence score
• Tier 4 webpage (opinion blog) → Starts with low score, can improve if logic is exceptional

The scoring shows this explicitly:

• "This argument cites 3 Tier 1 webpages and 2 Tier 2"
• vs.
• "This argument cites 15 Tier 4 webpages (possible Gish gallop)"

Connection to Logical Validity

From Logical Validity Scoring:

Webpages can be flagged for logical fallacies:

• "This webpage commits false dilemma" [requires reasoning]
• Community debates the fallacy claim
• If confirmed, logical coherence score drops
• Explicitly shown: "Logical validity reduced due to confirmed slippery slope fallacy"

Connection to Argument Trees

From Argument Scoring:

Webpages support specific arguments:

• Argument: "Carbon tax reduces emissions"
• Supporting webpage: [British Columbia carbon tax study - Tier 1]
• Score boost: +15 points (high-quality evidence)
• Argument: "Carbon tax will destroy economy"
• Supporting webpage: [Industry lobbying blog - Tier 4]
• Score boost: +3 points (weak evidence)

Recursive scoring:

• Strong webpages strengthen arguments
• Strong arguments strengthen parent beliefs
• Weak webpages expose weak foundations

---

Why This Matters: The Misinformation Problem

The Current Disaster

Google's engagement-driven algorithm:

• "Vaccines cause autism" blog ranks higher than CDC (more engaging)
• "Election was stolen" gets amplified (more shares)
• "Climate change is a hoax" outcompetes boring science (more clicks)

The mechanism:

1. Outrage is engaging
2. Engagement signals "quality" to algorithm
3. Algorithm amplifies outrage
4. More people see outrage
5. Outrage spreads faster than truth

The damage:

• Public health crises (vaccine hesitancy)
• Political instability (election denial)
• Policy paralysis (climate inaction)
• Epistemic collapse (can't agree on basic facts)

The ISE Solution

Evidence-driven algorithm:

• Peer-reviewed study ranks higher than blog (better evidence)
• Scientific consensus beats cherry-picked outliers (logical coherence)
• Boring accurate information gets proper ranking (not penalized for being boring)

The mechanism:

1. Evidence quality matters most
2. Logical coherence required
3. Fallacies get flagged and penalized
4. Community debate resolves disputes
5. Truth gets proper ranking regardless of engagement

The result:

• Misinformation visible as weak (low scores)
• Strong arguments properly elevated
• Users can see WHY rankings differ
• Disputes resolved through reasoning, not algorithms

---

The Transparent Scoring Process

Step 1: Submit Webpage

User provides:

• URL
• Position (supports/opposes which belief?)
• Classification (academic, journalism, opinion, advocacy)
• Brief justification: "Why is this relevant?"

System checks:

• Duplicate detection
• Auto-classification suggestion
• Links to potentially relevant arguments
• Flags for community review

Step 2: Community Scores

Multiple users rate on 0-100 scale:

• Evidence quality
• Logical coherence
• Relevance
• Position clarity

Aggregate scoring:

• Median (not mean) to resist outliers
• Weighted by evaluator credibility
• Confidence interval based on agreement level
• Display: Score ± confidence

Example:

• Evidence Quality: 85 ± 5 (high agreement, high confidence)
• Evidence Quality: 60 ± 30 (low agreement, needs debate)

Step 3: Link to Arguments

Users connect webpage to specific arguments:

• "This webpage supports argument X"
• "This webpage contradicts argument Y"
• Must explain the connection

System updates:

• Argument scores adjust based on webpage evidence tier
• Recursive scoring propagates up to parent beliefs
• Linkage strength matters (see Linkage Scores)

Step 4: Resolve Disputes

When scores disagree (>30 point variance):

Option A: Structured Debate

• Users present reasoning for their scores
• Others evaluate the reasoning (not just vote)
• Strong reasoning wins
• Scores adjust when 60% consensus reached

Option B: Jury Vote (if debate doesn't resolve)

• Mix: 50% general users, 50% domain experts
• All must provide reasoning
• 60% threshold required
• Decision logged and public

Everything is transparent:

• All reasoning visible
• All votes recorded
• Version history tracked
• Appeal process available

Step 5: Ongoing Updates

System monitors:

• Weekly reassessment based on new arguments
• Monthly aggregate score updates
• Dispute flags when variance increases
• Archive after 12 months of no activity

User actions:

• Submit updates when webpage content changes
• Flag new arguments that cite the webpage
• Refine scoring as understanding improves

---

Key Differences from Google (Summary Table)

Feature	Google	Idea Stock Exchange
Primary Goal	Ad revenue, engagement	Evidence quality, truth-seeking
Algorithm	Secret, proprietary	Open, transparent
Success Metric	Clicks, time on site, shares	Evidence tier, logical validity, argument strength
User Role	Passive consumer (SEO target)	Active evaluator and contributor
Ranking Basis	Engagement signals, SEO optimization	Evidence quality, logical coherence, relevance
Dispute Resolution	None (silent demotion)	Public debate with reasoning
Accountability	None (no explanation given)	Full transparency (all reasoning visible)
Gaming Resistance	Constant SEO arms race	Multiple overlapping signals, debate-based
Misinformation	Amplified if engaging	Flagged and downranked
Example	Vaccine denial blog ranks high (engaging)	Vaccine denial blog ranks low (weak evidence)

---

Integration with Other Media Types

Current focus: Webpages

Future expansion: Same principles apply to:

Books

• Evidence quality: Peer-reviewed academic > opinion book
• Cultural influence: Citations in other works, adoption in courses
• Logical coherence: Community evaluates argument structure
• See: Books

Videos

• Evidence quality: What sources cited? Which tier?
• Logical coherence: Fallacy detection in presentation
• Production quality: Separate from argument quality
• See: Videos

Podcasts

• Evidence quality: Guests' credentials, sources cited
• Logical coherence: Episode-level fallacy detection
• Influence: Citations in other arguments
• See: Podcasts

Academic Papers

• Already Tier 1 evidence baseline
• But: Can still have logical flaws or narrow relevance
• Replication status matters
• Citation impact tracked

Same core principles:

1. Evidence quality tiers
2. Logical coherence scoring
3. Relevance to specific arguments
4. Transparent community evaluation
5. Dispute resolution through debate

---

Why This Approach Works

1. Multiple Signals Prevent Gaming

Can't game all simultaneously:

• Good evidence quality requires Tier 1 sources
• Logical coherence requires fallacy-free reasoning
• Relevance requires actual connection to claim
• Cultural influence requires quality citations, not just clicks
• Community consensus required across all metrics

Unlike SEO:

• Can't fake peer-reviewed status
• Can't hide logical fallacies (community detects)
• Can't buy rankings (no ad revenue motive)
• Can't game through engagement bait

2. Transparency Creates Accountability

Everything visible:

• Why each score was given
• Who evaluated what
• What reasoning was provided
• How disputes were resolved

Result:

• Bad evaluations get challenged
• Good evaluations get defended
• Gaming attempts visible
• Quality emerges through scrutiny

3. Debate Resolves Real Disagreements

Not all disputes are gaming:

• Sometimes smart people disagree
• Sometimes context matters
• Sometimes evidence is genuinely ambiguous

Structured debate handles this:

• Both sides provide reasoning
• Community evaluates reasoning quality
• Conclusion based on argument strength
• Not just popularity contest

4. Expert Input Without Expert Gatekeeping

Experts matter but don't dominate:

• Domain expertise weighted in scoring
• But must provide reasoning (can't just decree)
• Community can challenge with evidence
• Experts judged by track record, not credentials alone

Balance:

• Prevents ignorant mob rule
• Prevents expert capture
• Meritocracy based on reasoning quality

---

Related Pages

Core ISE Framework:

• Main Blueprint
• Evidence Quality Tiers
• Logical Validity Scoring
• Linkage Scores
• Argument Scoring Algorithm

Other Media Types:

• Books
• Videos
• Podcasts
• Images
• People

Supporting Framework:

• One Page Per Topic
• Objective Criteria
• Reasons

---

The Bottom Line

Google ranks for profit. Engagement drives revenue. Truth is incidental.

ISE ranks for evidence. Argument quality drives score. Revenue is irrelevant.

The difference matters:

• Misinformation gets exposed (low scores for weak evidence)
• Quality sources get elevated (regardless of excitement level)
• Disputes get resolved through reasoning (not algorithm decree)
• Users can see why rankings differ (transparency)

We're building the infrastructure Google should have built: Ranking information by quality, not clicks.

---

Start evaluating webpages | Technical implementation | Contact me