Dr Larry Sanger07 November 2019

Why Is Wikipedia Broken? - Dr Larry Sanger - #118

0Frameworks

19Insights

Insights & moments

The myth-busts, hot takes, explainers, and tools worth keeping.

Myth Buster· 3

Myth Buster00:00

Wikipedia's 'Consensus' Isn't Real Consensus

Sanger argues Wikipedia has no formal process for resolving disputed editorial questions. What gets called "consensus" is actually just the view of whoever has the most seniority or allies on a topic, dressed up as if it were a genuine agreement.

No formalized way to arrive at decisions on disputed questions
What Wikipedia calls "consensus" is decided by whoever has the most seniority or allies on a topic
Sanger calls this "a cynical description of what is not a consensus at all"
He proposes reducing disputes to specific editorial decisions with legitimate votes instead

“It's not a consensus. It's just putting a cynical description on what is not a consensus at all.”
Larry Sanger · 00:30

#wikipedia#governance#misinformation

Myth Buster30:00

Locking Articles Isn't the Problem — Abusing the Lock Is

Sanger clarifies that protecting popular articles from vandalism is reasonable, since cleaning up after vandals is a genuine chore. The real problem is when that same lockdown power is abused to exclude legitimate outside contributors whose edits simply disagree with the incumbent editors.

Protecting popular articles from constant vandalism is legitimate and not itself the complaint
The abuse is locking out outsiders with legitimate contributions simply because insiders disagree with them
Wikipedia effectively runs on informal "fiefdoms" of editors controlling sets of articles rather than centralized rules

“The complaint that I would have is the abuse of such a policy... those people are excluded. That's a problem.”
Larry Sanger · 31:00

#wikipedia#moderation#governance

Myth Buster41:30

Some 'Conspiracy Theories' Were Just True Before They Were Confirmed

When asked whether an open encyclopedia risks legitimizing disinformation, Sanger points out that some past conspiracy theories are now accepted as true — meaning what counts as mainstream truth is a moving target, part of his argument for tracking multiple viewpoints rather than a single enforced consensus.

Some historical conspiracy theories have since been confirmed as true
What counts as "true" versus "conspiracy" shifts over time, so a single fixed consensus view is risky
His proposed system approximates a "mainstream view" by letting users choose which groups' ratings to follow

“Any conspiracy theories of the past... are now regarded as having been discovered, or are no longer conspiracy theories, they're just true.”
Larry Sanger · 42:00

#truth#misinformation#encyclosphere

Hot Take· 4

Hot Take27:30

Try Editing Wikipedia's 'Racism' Article — You'll Get Banned

Sanger makes a confident, testable claim: if a listener tried to edit Wikipedia's article on the definition of racism, they would almost certainly be banned just for attempting it, illustrating how locked-down contentious topics have become.

Sanger says he's "pretty damn sure" anyone editing the racism article would be banned just for trying
He jokingly warns listeners not to actually test it on that specific article
Frames it as evidence of how inaccessible editing on contentious topics has become

“If you were to go and edit the definition of racism in the article about racism, I'll bet you'd be banned just for trying.”
Larry Sanger · 28:00

“It has to be something that is controversial. Just don't do racism, jeez.”
Larry Sanger · 28:30

#wikipedia#editing#censorship

Hot Take36:30

Why One 'Correct' Encyclopedia Article Per Topic Is the Wrong Model

Sanger, describing himself as an objectivist about truth, argues encyclopedias shouldn't force a single consensus article per topic. Instead, his Encyclosphere vision would let readers filter and rank competing articles by the demographic or ideological background of the raters.

Sanger identifies as an objectivist: a mind-independent reality exists, but no political entity can be trusted to arbitrate what it is
Proposes rating and ranking articles by user-declared demographic and ideological categories
Example: compare the top-rated "jihad" article among American college professors vs. imams in the Middle East
Creates a decentralized, leaderless competition to write the best article from every point of view

“It would be really, really interesting to compare the top-rated article about jihad according to American college professors versus imams living in the Middle East.”
Larry Sanger · 39:00

#encyclosphere#truth#bias

Hot Take69:30

Apple's Privacy Reputation Is Overrated — Sanger's Buying a Linux Phone Instead

Pushing back on the idea that Apple is the best privacy option, Sanger says Apple talks a better game than it delivers, and reveals he's pre-ordered the Librem 5 — the first mass-produced Linux phone — as a genuine third smartphone ecosystem free of app-store gatekeeping.

Sanger says Apple "talks a good game" on privacy but isn't as strong on privacy and security as its reputation suggests
He pre-ordered the Librem 5 from Purism, calling it the first mass-produced Linux phone
Positions it as a third smartphone ecosystem, with full command-line/desktop access and no Apple-style app gatekeeping

“I am not that impressed. I mean I agree that Apple talks a good game.”
Larry Sanger · 69:30

#apple#privacy#linux-phone

Hot Take71:00

Linux Is Now Easier to Use Than Windows, Sanger Says

Contrary to Linux's old reputation as a geeks-only system, Sanger claims modern distributions like Ubuntu are now easier to install than Windows and could be handed to a non-technical user without ever touching the command line.

Sanger installed Red Hat Linux back in 2002 when it was "for geeks only"
He now says Linux distros like Ubuntu are easier to install than Windows
Claims a non-technical daily user could use it without ever needing the command line

“It's easier to install Linux, first of all, than it is to install Windows.”
Larry Sanger · 71:30

“Linux rocks now. I'm telling you.”
Larry Sanger · 72:00

#linux#technology#operating-systems

Explainer· 3

Explainer16:00

The Timeline of Wikipedia's Ideological Drift

Sanger lays out when he saw Wikipedia's neutrality policy start eroding: a visible ideological tilt appeared around 2004-2005, became pronounced by 2010, and in the last three or four years has tracked mainstream media's abandonment of objectivity.

Ideological tilt in some articles first visible around 2004-2005
Became "really pronounced" starting around 2010
In the last 3-4 years Wikipedia has "more or less followed suit" as mainstream media abandoned objectivity
Sanger says the neutrality policy has been increasingly ignored since a couple of years after he left

“Wikipedia has more or less followed suit as far as I can tell.”
Larry Sanger · 17:00

#wikipedia#media-bias#neutrality

Explainer18:30

Citizendium: Sanger's Real-Names Wikipedia Rival

Sanger describes Citizendium, the competing wiki encyclopedia he launched in 2006-2007, which required contributors to use real names, made everyone agree to a strict neutrality statement upfront, and ran a genuine constitutional governance system with real elections.

Citizendium launched 2006-2007, during Wikipedia's steepest growth period, and never took off
Required real names; pseudonyms only for vetted cases like political dissidents
Contributors had to agree in advance to strict neutrality and other core policies
Had a genuine constitutional system with real elections, unlike Wikipedia's "mob rule"

“That I think just takes care of a heck of a lot of the problems.”
Larry Sanger · 20:00

#citizendium#wikipedia-alternatives#governance

Explainer29:00

Why Wikipedia Vandalism Rarely Wins: The Revert Asymmetry

Sanger explains that Wikipedia's open-editing model survives trolling because reverting a bad edit takes far less effort than making one — an automated revert button means destructive changes get undone almost instantly, so vandalism is actually rare.

A revert button makes undoing bad edits far easier than making them
There's a work asymmetry: destroying content takes more effort to sustain than restoring it
As a result, vandalism is "actually quite rare" despite the open-editing model

“There's a lot more work, so there's an asymmetry in terms of work that someone who wants to destroy something has to do.”
Larry Sanger · 29:30

#wikipedia#editing#moderation

Story· 6

Story01:00

Why Larry Sanger Calls Himself Wikipedia's 'Ex-Founder'

Sanger explains he coined the term "ex-founder" himself after Jimmy Wales began denying him co-founder status around 2005, and after growing hostility from Wikipedia fans who blame him for the platform now being "out of control."

Sanger made up "ex-founder" a couple of days before the interview after polling Twitter on it
Jimmy Wales originally credited him as co-founder, then started denying it around 2005
People online have recently gotten hostile when he mentions being co-founder, since Wikipedia is "out of control"
He left Wikipedia in 2002 and permanently distanced himself in 2003

“So it's a little bit of a dig at him for that.”
Larry Sanger · 02:00

#wikipedia#larry-sanger#jimmy-wales

Story04:00

How a Failed 7-Step Editorial Process Led to Wikipedia

Sanger recounts meeting Jimmy Wales through an Ayn Rand mailing list, being recruited to run Nupedia's arduous peer-review encyclopedia, and how its slow output pushed them toward wiki software in January 2001, which became Wikipedia.

Sanger met Wales in the mid-1990s via an Ayn Rand mailing list
Nupedia used a 7-step editorial process inspired by "The Cathedral and the Bazaar"; too slow, only a few dozen articles in year one
A friend told Sanger about wiki software on Jan 2, 2001; he pitched it to Wales the same day
Nupedia's PhD editors refused to be associated with anything called "Wiki," so Sanger created the separate Wikipedia brand, launching internally Jan 10 and publicly Jan 15, 2001
Sanger wrote the original Wikipedia neutrality policy and other founding policies

“I ran home actually and made a one-page proposal to Jimmy Wales.”
Larry Sanger · 07:30

#wikipedia#nupedia#origin-story

Story11:00

How Wikipedia Got Taken Over by 'Trolls' Within a Year

Sanger says the biggest change since Wikipedia's founding is that, by the end of its first year, people he'd have considered trolls began dominating the project, driving away many of the best original contributors and eventually pushing him to give Wales an ultimatum.

Takeover by difficult, agenda-driven editors began toward the end of year one
Sanger left after Wales couldn't pay him once dot-com funding dried up
A year later Sanger gave Wales an ultimatum to fix the problems; Wales denied the problems existed
Sanger has been a Wikipedia critic ever since, one of its earliest critics

“You are absolutely driving away some of the very best people that we recruited for Nupedia.”
Larry Sanger · 13:00

#wikipedia#community#larry-sanger

Story56:00

Jack Dorsey Publicly Said Yes to Decentralizing Twitter

Sanger recounts publicly asking Twitter CEO Jack Dorsey three questions at South by Southwest about data portability, cross-platform import, and algorithm control — getting a public "yes" to all three, followed by a private half-hour call where Dorsey described what sounded like a real internal plan. None of it became public or was reported on.

At SXSW, Sanger publicly asked Dorsey three questions about decentralizing Twitter
Dorsey answered "yes" to all three on the record
Dorsey later called Sanger for a half-hour and described what sounded like a genuine internal plan
No news outlet ever reported on the exchange, and nothing has shipped

“He was on the record on Twitter saying yes to the decentralization of Twitter.”
Larry Sanger · 58:00

“Nobody ever wrote any news stories about it. Ever.”
Larry Sanger · 59:30

#twitter#jack-dorsey#decentralization

Story60:00

The July 4th Social Media Strike and the Traffic Data After It

Sanger describes organizing a two-day social media strike on July 4-5, inviting people to sign his "Declaration of Digital Independence." Thousands participated and it drew national media coverage, followed by a striking data point: major platforms' Alexa traffic rankings dropped sharply in the weeks after.

Social media strike held July 4-5, tied to a "Declaration of Digital Independence" still postable on his blog
Got coverage on CNN and Fox News; thousands participated
Post-strike: Facebook's Alexa rank fell from 3 to 5, Twitter from 11 to 30, Instagram down 25-30 places, Wikipedia from 5th to 9th most popular site
Winners in the same window: Bing, Microsoft Office, Stack Overflow/Exchange, Apple (new iPhone launch), and mainstream news sites like NYT, WaPo, Fox News, Breitbart, WSJ

“Facebook has gone down from an Alexa rank of three to five. Twitter has declined from 11 to 30 today.”
Larry Sanger · 61:30

#social-media-strike#decentralization#traffic-data

Story64:00

Quora Took Down Jordan Peterson's Famous '40 Rules' Post

Sanger recounts how Jordan Peterson's viral Quora post — the direct origin of his book "12 Rules for Life" — was removed for violating community standards, publicly appealed by Peterson, initially upheld after "manual review," and only reinstated after a Quora staffer privately intervened, all while Quora's traffic ranking collapsed.

Peterson's list of life-advice maxims began as a Quora answer and became one of Quora's most-engaged posts ever
Quora removed it for violating community standards, then upheld the removal after "manual review"
It was reinstated only after a Quora employee DMed Peterson privately and intervened
Quora's Alexa rank fell from 82 to over 260 in the same period Sanger discusses

“We're very sorry. We're not going to reinstate your post... and you are never getting it back.”
Larry Sanger · 65:30

#quora#jordan-peterson#censorship

Tool· 2

Tool22:00

Everipedia: The Blockchain Encyclopedia of Everything

Sanger explains his former role as chief information officer of Everipedia, a Wikipedia fork built on the blockchain that allows articles on literally any topic, aiming to scale to billions of articles rather than Wikipedia's roughly 5 million.

Everipedia is a Wikipedia fork, described as the first blockchain encyclopedia
Open to all contributors and all topics, unlike Wikipedia's increasingly closed approach
Sanger wrote an article about his own left thumb to prove any topic is allowed
Vision: not 5 million English articles but potentially billions covering "all proper names"

“I have an article about my left thumb, just to demonstrate that it is possible to have an article about anything.”
Larry Sanger · 23:30

#everipedia#blockchain#wikipedia-alternatives

Tool32:30

The Encyclosphere: An RSS-Style Feed for Encyclopedias

Sanger describes his new nonprofit, the Knowledge Standards Foundation, and its project the Encyclosphere — a shared metadata/content standard, modeled on RSS and the blogosphere, that would let anyone publish encyclopedia articles into a public commons pulling from Wikipedia, Britannica, Everipedia, Ballotpedia, and individual writers alike.

Sanger resigned from Everipedia in September to start the Knowledge Standards Foundation
Encyclosphere is modeled directly on RSS/the blogosphere: a shared feed format, not a single platform
Goal: let millions of people publish encyclopedia articles outside Wikipedia's editorial culture
Would aggregate metadata (and full articles where licensing allows) from Wikipedia, Everipedia, Ballotpedia, Britannica, the Stanford Encyclopedia of Philosophy, and individuals

“We need to build that kind of system for encyclopedias. That's what needs to exist.”
Larry Sanger · 35:00

#encyclosphere#knowledge-standards-foundation#decentralization

Takeaway· 1

Takeaway45:30

No Corporate or Government Money for the Knowledge Standards Foundation

Sanger states the Knowledge Standards Foundation will refuse corporate and government funding entirely, vetting even individual and foundation donors, because he believes a successful open knowledge standard would become powerful enough to attract corrupting influence.

No corporate money, no government money accepted
Individual and family foundation donations will be vetted ("make sure they're not another Epstein")
Reasoning: "knowledge is power," and a successful standard could attract undue influence seeking special favors

“We want the money to be clean and we want to make sure that there is no quid pro quo involved.”
Larry Sanger · 46:00

#knowledge-standards-foundation#funding#independence