LessWrong FAQ

This is a new FAQ writ­ten LessWrong 2.0. This is the first ver­sion and I apol­o­gize if it is a lit­tle rough. Please com­ment or mes­sage with fur­ther ques­tions, ty­pos, things that are un­clear, etc.

The old FAQ on the LessWrong Wiki still con­tains much ex­cel­lent in­for­ma­tion, how­ever it has not been kept up to date.

Ad­vice! We sug­gest you nav­i­gate this guide with the help on the table of con­tents (ToC) in the left side­bar. You will need to scroll to see all of it. Mo­bile users need to click the menu icon in the top left.

The ma­jor sec­tions of this FAQ are:

About LessWrong

What is LessWrong?

LessWrong is a com­mu­nity ded­i­cated to im­prov­ing our rea­son­ing and de­ci­sion-mak­ing. We seek to hold true be­liefs and to be effec­tive at ac­com­plish­ing our goals. More gen­er­ally, we want to de­velop and prac­tice the art of hu­man ra­tio­nal­ity.

To that end, LessWrong is a place to 1) de­velop and train ra­tio­nal­ity, and 2) ap­ply one’s ra­tio­nal­ity to real-world prob­lems.

LessWrong serves these pur­poses with its library of ra­tio­nal­ity writ­ings, com­mu­nity dis­cus­sion fo­rum, open ques­tions re­search plat­form, and com­mu­nity page for in-per­son events.

See also: Wel­come to LessWrong!

What is ra­tio­nal­ity?

Ra­tion­al­ity is a term which can have differ­ent mean­ings to differ­ent peo­ple. You might already as­so­ci­ate with a few things. On LessWrong, we mean some­thing like the fol­low­ing:

  • Ra­tion­al­ity is think­ing in ways which sys­tem­at­i­cally ar­rive at truth.

  • Ra­tion­al­ity is think­ing in ways which cause you to sys­tem­at­i­cally achieve your goals.

  • Ra­tion­al­ity is try­ing to do bet­ter on pur­pose.

  • Ra­tion­al­ity is rea­son­ing well even in the face of mas­sive un­cer­tainty.

  • Ra­tion­al­ity is mak­ing good de­ci­sions even when it’s hard.

  • Ra­tion­al­ity is be­ing self-aware, un­der­stand­ing how your own mind works, and ap­ply­ing this knowl­edge to think­ing bet­ter.

See also: What Do We Mean By “Ra­tion­al­ity”?, Why Spock is Not Ra­tional, What are the open prob­lems in Hu­man Ra­tion­al­ity?

What is the his­tory of LessWrong?

In 2006, Eliezer Yud­kowsky and oth­ers be­gan writ­ing on Over­com­ing Bias, a group blog with the gen­eral theme of how to move one’s be­liefs closer to re­al­ity de­spite bi­ases such as over­con­fi­dence and wish­ful think­ing. In 2009, Eliezer moved to a new com­mu­nity blog, LessWrong. Eliezer seeded LessWrong with a se­ries of daily blog posts which be­came known as The Se­quences. Th­ese writ­ings at­tracted a large com­mu­nity of read­ers and writ­ers in­ter­ested in the art of hu­man ra­tio­nal­ity.

See also: A Brief His­tory of LessWrong

What makes LessWrong differ­ent from other dis­cus­sion fo­rums?

A com­bi­na­tion of traits makes LessWrong dis­tinct among on­line com­mu­ni­ties.

  1. We have un­usu­ally high stan­dards of dis­course. We em­pha­size cu­ri­os­ity, truth-seek­ing, crit­i­cal self-re­flec­tion, in­tel­lec­tual col­lab­o­ra­tion, and the long at­ten­tion spans re­quired to ac­tu­ally think through com­pli­cated ideas.

  2. We are open to un­usual ideas and are will­ing to doubt con­ven­tional wis­dom. Cu­ri­os­ity and truth-seek­ing re­quire a will­ing­ness to some­times con­sider po­si­tions which are strange by or­di­nary stan­dards, and in some cases, these po­si­tions will turn out to be cred­ible. As a re­sult of this open­ness, some un­con­ven­tional ideas are preva­lent on LessWrong and many more are en­ter­tained.

  3. We make in­tel­lec­tual progress by build­ing on a large num­ber of com­mu­nally-shared back­ground ideas and con­cepts.

Why the name? It is a bit odd . . .

I (Ruby) per­son­ally wasn’t there when the name was cho­sen so I’m not cer­tain of the his­tor­i­cal thought pro­cess, but I in­ter­pret the name “LessWrong” as ex­press­ing two im­por­tant points:

  1. A hum­ble recog­ni­tion that no hu­man is ever go­ing to at­tain perfectly true be­liefs and be right about ev­ery­thing. We should always be­lieve that some of our be­liefs are mis­taken, we just don’t know which ones.

  2. A bold recog­ni­tion that notwith­stand­ing the im­pos­si­bil­ity of be­ing perfectly right, there is still the pos­si­bil­ity of be­ing less wrong. Every­one be­lieves false things, but some be­lieve a lot fewer wrong things than oth­ers.

And so the as­pira­tion of LessWrong is that by ded­i­cat­ing our­selves to learn­ing how to think in ways which more sys­tem­at­i­cally lead to truth (what we suc­cinctly call ra­tio­nal­ity), we can mean­ingfully re­duce our mis­taken no­tions and have far more ac­cu­rate mod­els of re­al­ity.

Who is this Eliezer guy I keep hear­ing about?

Eliezer Yud­kowsky was the origi­nal founder of LessWrong back in 2009. His writ­ings on ra­tio­nal­ity at­tracted to the site a large num­ber of peo­ple en­thu­si­as­tic about learn­ing to think bet­ter. Eliezer’s best-known works are The Se­quences, (later re­named Ra­tion­al­ity: From AI to Zom­bies) and Harry Pot­ter and the Meth­ods of Ra­tion­al­ity. Th­ese texts are part of LessWrong’s philo­soph­i­cal foun­da­tion, and so un­sur­pris­ingly, you will see men­tions of Eliezer not in­fre­quently.

How does LessWrong make money?

We don’t. The LessWrong or­ga­ni­za­tion is a non­profit funded by dona­tions.

This hope­fully has the benefit of re­duc­ing our in­cen­tives to op­ti­mize for clicks and pageviews. In­stead, we can fo­cus on our stated pur­pose.

I have feed­back, bug re­ports, or ques­tions not an­swered in this FAQ. What should I do?

You have sev­eral op­tions.

  1. Mes­sage the LessWrong team via In­ter­com (available in the bot­tom right). En­sure you don’t have Hide In­ter­com set in your ac­count set­tings.

  2. Send a pri­vate mes­sage to a mem­ber of the LessWrong team (see these on the team page)

  3. Open an is­sue on the LessWrong Github repos­i­tory.

  4. Ask a question

  5. For com­plaints and con­cerns re­gard­ing the LessWrong team, you can mes­sage Vaniver.

Oh no! I think I lost my post/​draft/​san­ity! What can I do?

LessWrong stores re­vi­sions of posts as you’re draft­ing them. If you think you have lost con­tent, please mes­sage the team via In­ter­com and we’ll see what we can do.

Get­ting Started

I’m new. Where do I start?

We en­courage new users to read for a while be­fore div­ing into dis­cus­sions or mak­ing their own posts. This is helpful for new users to un­der­stand the site’s cul­ture and back­ground.

  • Our wel­come page offers a high-level de­scrip­tion of LessWrong and in­cludes a list of sam­ple posts. It is a great way to get a feel for what LessWrong is like.

  • For new mem­bers who want to get up to speed, we di­rect you to­wards our core read­ings which can be found on the Library page and are de­scribed el­se­where in this FAQ.

  • At the same time, feel free to browse more re­cent con­tent. This an­swer de­scribes all the way you can lo­cate con­tent on LessWrong.

  • Un­like other places on the In­ter­net, it is of­ten worth­while to read the com­ment sec­tions on posts. Our com­ment­ing guidelines state that is prefer­able:

    • Aim to ex­plain, not per­suade.

    • Pre­sent your own per­spec­tive rather than state group con­sen­sus or in­vok­ing au­thor­i­ties.

    • Get cu­ri­ous. If you dis­agree with some, try to figure out what they’re think­ing. What’s their model? Don’t just as­sume they’re dumb or evil.

If you’re very new and you be­gin post­ing or com­ment­ing, you might find that you are quickly down­voted. This doesn’t mean you’re bad or un­wel­come! But you are prob­a­bly vi­o­lat­ing a norm or ig­nor­ing ex­pected knowl­edge on the site. We sug­gest you read up a bit more be­fore try­ing again later.

What’s a good and fast way to learn about how the web­site works?

LessWrong ex­ten­sively uses tool-tips and con­tent pre­views to help users un­der­stand how the site works and see what con­tent is even be­fore they click.

We en­courage you to mouse over most el­e­ments of the site to see what pops up. You will find:

  • Items in the left side­bar have tool-tips.

  • Hover­ing over post ti­tles dis­plays an ex­cerpt, read­ing time, and other meta info.

  • Hover­ing over user­names dis­plays karma, join date, num­ber of posts and com­ments, and a bio if the user has set one.

  • Hover­ing over karma scores dis­plays the num­ber of votes (in our karma sys­tem, karma does not usu­ally equal the num­ber of votes).

How do I cre­ate an ac­count? (And why should I?)

Although not re­quired to use the LessWrong web­site; we recom­mend cre­at­ing an ac­count so that you can:

  • Sub­scribe to users and differ­ent classes of posts.

  • Save your user settings

  • Vote and com­ment on posts.

  • Store your read­ing his­tory, en­abling tai­lored recom­men­da­tions and po­ten­tially new fea­tures such as view­ing your read­ing his­tory and cre­at­ing cus­tom read­ing lists.

Creat­ing an ac­count takes un­der 30 sec­onds. Click lo­gin in the top right and en­ter a user­name, email, and pass­word.

Once you have cre­ated an ac­count, feel free to in­tro­duce your­self in the lat­est Open/​Wel­come thread. Let oth­ers know how you found LessWrong, your back­ground, and what you’re hop­ing for from LessWrong. This al­lows ex­ist­ing mem­bers to point you in the di­rec­tion of ma­te­rial which you might es­pe­cially like.

How do I Ask Ques­tions/​Make Posts/​Go to My Pro­file/​Pri­vate Mes­sage/​Log Out?

For logged-in users, you can ac­cess all these op­tions via the drop-down menu ac­cessible by click­ing your user­name.

The star to the right of your user­name is the karma no­tifier (star icon) and but­ton for no­tifi­ca­tions panel (bell icon).

How do I edit my ac­count set­tings? What can I do?

By click­ing on your user­name and click­ing Edit Ac­count, you ac­cess your ac­count set­tings. There you can:

  • Set a bio for your ac­count to let other LessWrong mem­bers know about you. If you set one, it will show up when they mouse-over your user­name.

  • Hide or show In­ter­com (mes­sag­ing ser­vice with the LessWrong team mem­bers).

  • Ac­ti­vate the mark­down ed­i­tor.

  • Tog­gle com­ment col­lapse set­tings.

  • Opt into beta fea­tures (new)

  • Ad­just set­tings for no­tifi­ca­tions of re­sponses to your posts and comments

  • Ad­just set­tings for the karma no­tifier.

  • Un­sub­scribe from your email sub­scrip­tions.

Read­ing Content

What are all the ways to ac­cess con­tent on LessWrong?

Ah, there are many ways!


LessWrong’s home­page has the fol­low­ing con­tent sec­tions:

  • Core Read­ing (shown only to logged-out users)

  • Curated

  • Lat­est Posts

  • Re­cent Discussion

Core Readings

The core read­ings sec­tion pro­vides links to texts which de­scribe the in­tel­lec­tual foun­da­tions of LessWrong. They are de­scribed here.


Each week, LessWrong’s mod­er­a­tion team se­lects on av­er­age three posts which seem to us to be es­pe­cially well-writ­ten, in­sight­ful, in­struc­tive, or oth­er­wise im­por­tant. Th­ese are tagged as cu­rated posts and ap­pear with a star icon next to the ti­tle.

The three most re­cently cu­rated posts ap­pear in the Cu­rated sec­tion. You can view more Cu­rated posts by click­ing View All Cu­rated Posts or se­lect­ing the Cu­rated filter on the AllPosts page.

Be­neath the Cu­rated sec­tion is a but­ton to sub­scribe via email or RSS to cu­rated posts (~3/​week).

Lat­est Posts

The Lat­est Posts sec­tion dis­plays all* re­cent posts to LessWrong. Th­ese sorted mag­i­cally** to bal­ance be­tween re­cency and qual­ity (as in­di­cated by karma score), i.e. more up­voted posts re­main higher up in the Lat­est Posts sec­tion for longer.

*By de­fault, only Front­page posts are dis­played in the Lat­est Posts sec­tion. To en­able Per­sonal blog­posts to ap­pear as well, check the check­box be­neath the sec­tion. See more in What’s the differ­ence be­tween Per­sonal Blog­posts and Front­page Posts?

**LessWrong uses the fol­low­ing for­mula to rank posts in Lat­est Posts:

This is same the for­mula as used by Hacker News. You can read about it here.

Re­cent Discussion

This sec­tion is a purely-time based feed of the most re­cent com­ment ac­tivity hap­pen­ing on posts. Cur­rently, all posts (both Per­sonal blog­posts and Front­page) are shown. Dis­cus­sion is grouped by post but re­stricted to only show­ing a few com­ments per post.

All Posts Page (aka Archive)

Whereas the home­page dis­plays posts or­dered with a mag­i­cal al­gorithm, the All Posts page gives you com­plete con­trol over which posts are in­cluded and how they or­dered.

The All Posts page can be ac­cessed via the left side­bar and drop-down menu (desk­top); but­tons on the bot­tom of the screen (mo­bile); or di­rectly via www.less­wrong.com/​allPosts

The gear icon al­lows you to se­lect which posts:

  • All Posts (ab­solutely ev­ery­thing)

  • Front­page (pages given Front­page sta­tus by the mod­er­a­tion team)

  • Cu­rated (pages given Cu­rated sta­tus by the mod­er­a­tion team)

  • Ques­tions (from our Open Ques­tions plat­form)

  • Events (from the Com­mu­nity Events Page)

  • Meta (de­p­re­cated cat­e­gory con­tain­ing posts about the LessWrong web­site and similar)

Th­ese can then be sorted by: Daily, Magic, has Re­cent Com­ments, New, Old, and Top.

The Library

The Library page is ac­cessible from the left side­bar/​drop-down menu (desk­top) or the but­tons at the bot­tom of the screen (mo­bile). The Library page con­tains se­quences (or­dered sets of posts) and col­lec­tions (or­dered sets of se­quences) of LessWrong’s best writ­ings. Th­ese are split into Core Read­ings, Cu­rated Se­quences, and Com­mu­nity Se­quences.

LessWrong’s de­vel­op­ers have put effort into mak­ing the read­ing ex­pe­rience in The Library as con­ve­nient and en­joy­able as pos­si­ble.

Core Readings

Th­ese are Ra­tion­al­ity: From AI to Zom­bies (formerly The Se­quences), The Codex, and Harry Pot­ter and the Meth­ods of Ra­tion­al­ity. They are de­scribed in What are LessWrong’s core read­ings?

Cu­rated Sequences

Similar Cu­rated posts, Cu­rated se­quences are sets of posts which LessWrong’s mod­er­a­tion team think are es­pe­cially valuable and ought to be in­cluded in LessWrong’s in­tel­lec­tual canon.

Top cu­rated se­quences in­clude:

Com­mu­nity Sequences

Any LessWrong site mem­ber, not just mod­er­a­tors, can cre­ate post se­quences. Th­ese ap­pear in the Com­mu­nity Se­quences sec­tion.

Stand­out men­tions in­clude:

Se­quences have qual­i­ta­tive benefits over posts in that an au­thor can build to­wards a larger point or ex­plain more nu­anced con­cepts than is pos­si­ble in sin­gle (even quite long) blog posts.

You can also cre­ate your own se­quence on this page.

User Page

Lastly, you can ac­cess a User’s posts and com­ments di­rectly from their user page.

Note that you have the same op­tions available for sort­ing and fil­ter­ing a user’s posts as you do on the All Posts page.

What’s the differ­ence be­tween Front­page posts and Per­sonal blog­posts?

Although LessWrong’s fo­cus is on the de­vel­op­ment and ap­pli­ca­tion of ra­tio­nal­ity, we in­vite posts on al­most any topic. To en­sure that the de­fault ex­pe­rience is still one cen­tered on ra­tio­nal­ity, LessWrong clas­sifies posts into Front­page posts and Per­sonal blog­posts.

Front­page posts must meet the crite­ria of be­ing broadly rele­vant to LessWrong’s main in­ter­ests; time­less, i.e. not about re­cent events; and are at­tempts to ex­plain not per­suade. In con­trast, Per­sonal blog­posts can be on any topic of in­ter­est to the au­thor in­clud­ing di­vi­sive top­ics (which we gen­er­ally keep off the front­page), dis­cus­sions about the com­mu­nity, and meta posts about LessWrong it­self.

Front­page posts have visi­bil­ity by de­fault. Per­sonal blog­posts can be viewed by: i) check­ing the “show Per­sonal blog­posts” check­box on the home­page, ii) via the All Posts page if “All Posts” filter op­tion is se­lected, iii) via a user’s pro­file page, iv) in the Re­cent Dis­cus­sion sec­tion of the home­page.

See also: Site Guide: Per­sonal Blog­posts vs Front­page Posts

What are Cu­rated posts?

Each week, LessWrong’s mod­er­a­tion team se­lects on av­er­age three posts which seem to us to be es­pe­cially well-writ­ten, in­sight­ful, in­struc­tive, or oth­er­wise im­por­tant. Th­ese are tagged as cu­rated posts and ap­pear with a star icon next to the ti­tle.

All Cu­rated posts will also be Front­page posts.

The three most re­cently cu­rated posts ap­pear in the Cu­rated sec­tion. You can view more Cu­rated posts by click­ing View All Cu­rated Posts or se­lect­ing the Cu­rated filter on the AllPosts page.

Be­neath the Cu­rated sec­tion is a but­ton to sub­scribe via email or RSS to cu­rated posts (~3/​week).

What are LessWrong’s core read­ings?

The fol­low­ing texts lay the philo­soph­i­cal foun­da­tions of the LessWrong web­site and com­mu­nity. They are widely re­garded as ex­cel­lent, and, even when the ideas are not uni­ver­sally agreed upon, they are still com­monly as­sumed back­ground knowl­edge in the com­mu­nity.

Ra­tion­al­ity: AI to Zom­bies (aka “the Se­quences”)

In 2006, Eliezer be­gan post­ing on a pre­cur­sor to LessWrong, the shared blog, Over­com­ing Bias be­fore the cur­rent site was launched in 2019. He posted nearly daily for sev­eral years and those writ­ings be­came known as the Se­quences. Later they were ed­ited into a book, Ra­tion­al­ity: A-Z (or RAZ).

Ra­tion­al­ity: A-Z is a deep ex­plo­ra­tion of how hu­man minds can come to un­der the world they ex­ist in—and all the rea­sons they so com­monly fail to do. The com­pre­hen­sive work:

Eliezer cov­ers these top­ics and oth­ers through alle­gory, anec­dote, and sci­en­tific the­ory. He demon­strates the ideas by ap­ply­ing them to de­bates in ar­tifi­cial in­tel­li­gence (AI), physics, metaethics, and con­scious­ness.

To start read­ing R:A-Z, visit www.less­wrong.com/​ra­tio­nal­ity or visit Ama­zon to pur­chase the e-book or au­dio­book.

The Codex

Scott Alexan­der’s, one of LessWrong’s ear­liest and most pro­lific con­trib­u­tors, wrote many es­says on good rea­son­ing, learn­ing from the in­sti­tu­tion of sci­ence, and differ­ent ways so­ciety has and could be or­ga­nized. Th­ese have been or­ga­nized into the Codex. Scott’s se­quences in­clude:

His ex­em­plary es­says in­clude:

Harry Pot­ter and Meth­ods of Ra­tion­al­ity (HPMOR)

A side pro­ject of Eliezer’s grew to be one of the most highly rated Harry Pot­ter fan­fic­tions of all time and an ex­cel­lent primer on ra­tio­nal­ity. Eliezer imag­ined an al­ter­nate-uni­verse Harry Pot­ter who grew up with lov­ing adopted par­ents, one of them an Oxford sci­en­tist. In this ver­sion, Harry en­ters the wiz­ard­ing world with En­light­en­ment ideals and the ex­per­i­men­tal spirit.

We recom­mend HPMOR to in­ter­ested in an in­tro­duc­tion to ra­tio­nal­ity via a highly en­ter­tain­ing nar­ra­tive. Click here to read HPMOR through LessWrong or try the au­dio­book.

What’s with all the AI and math posts?

For both his­tor­i­cal rea­sons and be­cause these top­ics are rele­vant to hu­man ra­tio­nal­ity, many mem­bers of the LessWrong com­mu­nity are in­ter­ested in AI, de­ci­sion-the­ory, math, and re­lated top­ics.

  • His­tor­i­cally: LessWrong’s founder and au­thor of its foun­da­tional works, Eliezer Yud­kowsky, is a co-founder of the Ma­chine In­tel­li­gence Re­search In­sti­tute and ma­jor propopent for AI safety. His writ­ings on LessWrong at­tracted many peo­ple who were in­ter­ested in both ra­tio­nal­ity and AI/​AI safety, caus­ing these to be on­go­ing over­lap be­tween LessWrong an AI safety com­mu­ni­ties.

  • Rele­vancy: Ar­tifi­cial in­tel­li­gence is very much the study of in­tel­li­gence and how “minds” work. Even if you are more in­ter­ested in how hu­man minds work and in im­prov­ing your hu­man ra­tio­nal­ity, there is much to learn from think­ing gen­er­ally about how in­tel­li­gence works (for hu­mans or non-hu­mans). In par­tic­u­lar, the fields of AI of­ten brings tech­ni­cal pre­ci­sion and rigor to think­ing to the gnarly, com­pli­cated top­ics of in­tel­li­gence and op­ti­mal de­ci­sion-mak­ing.

    • Be­cause of this rele­vance, many writ­ings about hu­man ra­tio­nal­ity on LessWrong (from Eliezer and oth­ers) make refer­ence to con­cepts from AI and for­mal de­ci­sion-the­ory.

See also: What is the AI Align­ment Fo­rum (AIAF) and what does it have to do with LessWrong?

What is the AI Align­ment Fo­rum (AIAF) and what does it have to do with LessWrong?

The AI Align­ment Fo­rum is an on­line hub for AI Safety (aka AI al­ign­ment) re­searchers to dis­cuss top­ics in the field. The AI Align­ment fo­rum is an­other pro­ject of the LessWrong team’s and re­sul­tantly shares some in­fras­truc­ture with LessWrong proper, i.e. shared user ac­counts.

Be­cause of the over­laps be­tween the LessWrong and AI Safety com­mu­niites and rele­vance of AI con­tent to ra­tio­nal­ity, posts made to the AI Align­ment fo­rum are au­to­mat­i­cally cross-posted to LessWrong.

  • Th­ese posts will have the AIAF sym­bol (Omega/​ 𝛀) shown next to the ti­tle and con­tain a warn­ing that the con­tent may es­pe­cially tech­ni­cal.

I (Ruby) am ad­vo­cat­ing strongly for there to be an easy way to filter these out for users who are not in­ter­ested in AIAF con­tent.

What is that Omega sym­bol I see on some posts? Oh, it’s AIAF karma.

Posts and com­ments which been cross-posted from the Align­ment Fo­rum will dis­play their Align­ment Fo­rum karma (sym­bol: Omega/​ 𝛀). When users with the abil­ity to vote on Align­ment Fo­rum con­tent vote on cross-posted AIAF on LessWrong, this will cause both the con­tents or­di­nary LessWrong karma and Align­ment Fo­rum karma to up­date.

Post­ing & Commenting

What can I post on LessWrong?

Posts on prac­ti­cally any topic are wel­comed on LessWrong. I (and oth­ers on the team) feel it is im­por­tant that mem­bers are able to “bring their en­tire selves” to LessWrong and are able to share all their thoughts, ideas, and ex­pe­riences with­out fear­ing whether they are “on topic” for LessWrong. Ra­tion­al­ity is not re­stricted to only spe­cific do­mains of one’s life and nei­ther should LessWrong be.

How­ever, to main­tain its over­all fo­cus while still al­low­ing posts on any topic, LessWrong clas­sifies posts as ei­ther Per­sonal blog­posts or as Front­page posts. See more in the post on Per­sonal Blog­post vs Front­page Posts.

The Editor

LessWrong’s ed­i­tor is what use you to en­ter posts and com­ments.

How do I use Mark­down? (And not the Draft.js de­fault ed­i­tor)

By de­fault, LessWrong uses an im­ple­men­ta­tion of Draft.js, how­ever, if you pre­fer, you can switch to en­ter­ing your text with mark­down syn­tax. To do, check Ac­ti­vate Mark­down Edi­tor check­box in your ac­count set­tings.

With the Mark­down ed­i­tor ac­ti­vated, you can use Mark­down syn­tax for for­mat­ting.

How do I in­sert images?

If you are us­ing the Draft.Js ed­i­tor, se­lect some text (or whites­pace) and click the image icon in the toolbar that ap­pears your text. Insert a URL to a hosted image. The image must be hosted! Use a free on­line ser­vice like Imgur or similar. En­sure you use the url to the hosted image it­self, not the page dis­play­ing up­loaded image (com­mon mis­take).

Note: image in­ser­tions are only en­abled for posts, not com­ments.

If you are us­ing the Mark­down ed­i­tor, us­ing the Mark­down syn­tax for in­sert­ing images. It is:

![image text](https://​​www.ex­am­ple.com)

As above, the link must be to a hosted image.

How do I in­sert spoiler pro­tec­tions?

LessWrong gives you a way to “avoid spoiling” your read­ers. Text is con­cealed un­til a user mouses over it (it works a bit less well on mo­bile right now). This func­tion­al­ity is use­ful for cre­at­ing ex­er­cises in your posts, e.g. ask a ques­tion in your post and con­ceal with an­swer be­neath spoiler pro­tec­tion so users don’t ac­ci­den­tally see it. See this post as an ex­am­ple.

In the Draft.js ed­i­tor type `>!` on a new line, then press space, then a spoiler box should appear

In the Mark­down ed­i­tor, sur­round your text with `:::spoiler` at the be­gin­ning, and `:::` at the end.

How do I in­sert foot­notes?

At the pre­sent time, foot­notes with link high­light­ing and re­turn but­tons can only be in­serted by us­ing the Mark­down ed­i­tor. To do use, use the syn­tax de­scribed here.

How do I use La­tex?

If us­ing the Draft.js ed­i­tor, press Cmd-4 for in­line and Cmd-M for block-level. (Ctrl on Win­dows).

If us­ing Mark­down, sur­round your LaTeX text with $, for ex­am­ple:

$<LaTeX text>$

How do I add mul­ti­ple au­thors to a post?

Cur­rently, only ad­mins can add mul­ti­ple to au­thors to a post. Send us a mes­sage on In­ter­com (bot­tom right) or email us at team@less­wrong.com and we’ll do it for you.

Karma & Voting

How do I vote?

Posts and com­ments have but­tons for up­vot­ing and down­vot­ing them dis­played around the posts cur­rent karma score.

Fur­ther, you have the op­tion to strong up­votes or down­vote posts and com­ments. On desk­top: hold the vote but­ton un­til you see the dou­ble bars ap­pear. On mo­bile: dou­ble-tap the vote but­ton (ig­nore a tool-tip tel­ling you to hold).

What should my votes mean?

We en­courage peo­ple to vote such that up­vote means “I want to see more of this” and down­vote means “I want to see less of this.”

What’s the re­la­tion­ship be­tween votes and karma? Why aren’t they the same?

Posts and com­ments have a karma score. A sin­gle vote will in­crease or de­crease the karma by an in­te­ger value. Upvotes in­crease the karma, down­votes de­crease—and these can can­cel out.

Fur­ther, users have karma scores too. A user’s karma score is the sum of all the karma on their posts and com­ments. The votes of users with more karma have more power un­der LessWrong’s vot­ing sys­tem, en­sur­ing that users who have earned the com­mu­nity’s re­spect and trust have more in­fluence than new sign-ups. Be­cause some users have votes which are worth more than a sin­gle point, the karma score of a post is usu­ally greater than the num­ber of votes on it.

What’s the map­ping be­tween users’ karma and vot­ing power?

A user’s vote power is de­ter­mined by the code im­ple­mented in this file.

What about re­acts and other di­men­sions of re­sponse?

We’re think­ing about it! See FB/​Dis­cord Style Re­acts.

No­tifi­ca­tions & Subscriptions

The no­tifi­ca­tion and sub­scrip­tions sys­tem are cur­rently un­der­go­ing a sig­nifi­cant up­grade. Ex­pect the func­tion­al­ity to be ex­panded in the next week or two. We will up­date this doc­u­men­ta­tion then.

Where do I get no­tifi­ca­tions?

See the bell icon in the up­per right-hand cor­ner. There are four tabs.

Bell: com­bined re­sponses to your posts and com­ments + pri­vate mes­sage notifications

  1. Paper/​Doc: New post notifications

  2. Speech Bub­ble: No­tifi­ca­tions of com­ments on your posts

  3. Two Speech Bub­bles: No­tifi­ca­tions of pri­vate mes­sages on.

What can I get no­tifi­ca­tions for?

In your ac­count set­tings you can tog­gle no­tifi­ca­tions on and off for re­sponses to your posts and com­ments.

Can I sub­scribe by email? What can I sub­scribe to?

Right now, you can sub­scribe to re­ceive Cu­rated posts by email or RSS. See the sub­scribe but­tons be­neath the Cu­rated posts sec­tion on the home­page or in your ac­count set­tings.


How do I sent pri­vate mes­sages to other users?

Nav­i­gate to a user’s page by click­ing on an ap­pear­ance of their user­name or find­ing them via search. Click send mes­sage.

To read your mes­sages, click on the no­tifi­ca­tion icon (bell icon, top right) > click the two speech bub­bles on the right. Or visit www.less­wrong.com/​in­box.


What do you mean, ques­tions?

The LessWrong team is ac­tively de­vel­op­ing a new ex­per­i­men­tal Open Ques­tions Re­search Plat­form. The vi­sion is to build a sys­tem which al­lows the LessWrong com­mu­nity to ap­ply its high stan­dards of rea­son­ing and schol­ar­ship to solv­ing large, im­por­tant ques­tions.

We ex­pect LessWrong’s Open Ques­tions to be valuable be­yond ex­ist­ing plat­forms, e.g. Quora and Stack­Ex­change, for mul­ti­ple rea­sons. Among them:

  • The LessWrong com­mu­nity’s fo­cus on good rea­son­ing and com­mit­ment to truth

  • The de­sign of our tool to be for large [dis­tributed] re­search ques­tions.

The LessWrong teams thinks this is an ex­cel­lent way to train and ap­ply ra­tio­nal­ity.

What kind of ques­tions can I ask?

If you have a ques­tion which seems like the LessWrong com­mu­nity could an­swer bet­ter than any other Q&A plat­form, we wel­come you to ask it here.

We will han­dle mak­ing sure ques­tions of the right type are shown in each place, so don’t worry too much about whether your ques­tion is rele­vant. Like with posts, we wel­come ques­tions on most top­ics and then cat­e­go­rize them ap­pro­pri­ately.

Ex­ist­ing ques­tions have been of all the fol­low­ing types:

  • Re­quests for facts

  • Re­quests for an­swers to difficult re­search questions

  • Re­quests for ex­pla­na­tions of difficult topics

  • Re­quests for ar­gu­ments for or against a position

  • Re­quests for opinions and in­sights on a given topic

  • Re­quests for per­sonal advice

  • Recom­men­da­tions, feed­back, or re­quest to hear other’s per­sonal experiences

  • Ques­tions about the LessWrong website

Th­ese are all good. Get a sense of what peo­ple ask on LessWrong by view­ing the ques­tions page.

How do I ask ques­tions?

To ask a ques­tion, click on your User­name (top right, you must have an ac­count), and click Ask Ques­tion [Beta].

It is best to give your ques­tion a ti­tle which suc­cinctly de­scribed what you’re ask­ing (ques­tion with ques­tion marks are bet­ter than sen­tences) and write a longer de­scrip­tion in the body ex­plain­ing what ex­actly it is you want to know and why.

How can I helpfully an­swer ques­tions?

You can prob­a­bly help more than you think! Even if it’s not easy to an­swer a ques­tion out­right, small con­tri­bu­tions of in­for­ma­tion or in­sight can still go a long way.

We en­courage you to look through the ques­tions page to find ques­tions that ei­ther have ex­ist­ing knowl­edge about or catch your cu­ri­os­ity about. Read through ex­ist­ing an­swers and then see what you can add. All of the fol­low­ing can be use­ful con­tri­bu­tions in ad­di­tion to di­rect an­swers:

  • A link or recom­men­da­tion to a re­source which might help an­swer the ques­tion.

  • A recom­men­da­tion of who might know the an­swer that you could talk to.

  • A sug­ges­tion for what things, if ob­served, would be ev­i­dence about a ques­tion one way or an­other.

  • An ex­pla­na­tion of how the ques­tion is maybe “con­fused” and should be dis­solved.

  • Iden­ti­fy­ing a re­lated or “sub-ques­tion” you think will help an­swer the big­ger ques­tion.

    • Note the Ask re­lated ques­tion fea­ture in ques­tion pages.

An­swer­ing ques­tions is also a great way to prac­tice the ne­glected virtue of schol­ar­ship. A cou­ple of LessWrong mem­bers have writ­ten guides helpful for get­ting started with schol­ar­ship. Luke­prog wrote Schol­ar­ship: How to Do It Effi­ciently and gw­ern wrote a lengthy In­ter­net Search Tips guide.

How do I in­ter­act with ques­tions?

Ques­tion pages might seem con­fus­ing at first. They’re not so bad. Be­neath the ques­tion text you will see a text­book with three op­tions: “New An­swer”, “Ask Re­lated Ques­tion”, and “New Com­ment” as pic­tured.

New An­swer: An an­swer can be any re­sponse which sheds light on the ques­tion be­ing asked, even if it’s not a com­plete or com­pre­hen­sive an­swer. Some users choose to make smaller con­tri­bu­tions as com­ments. There’s a bit of fuzzy line here so don’t worry about it too much. You have the abil­ity to move re­sponses back and forth be­tween be­ing com­ments or an­swers if you change your mind.

New Com­ment: Com­ments on ques­tions can be used to ask clar­ify­ing ques­tions and other thoughts which aren’t re­ally an­swers to the ques­tion asked. You can also com­ment on other peo­ple’s An­swers, al­low­ing for dis­cus­sion of those an­swers.

Ask Re­lated Ques­tion: For large ques­tions, some­times you can’t an­swer a ques­tion di­rectly and in­stead to ask an­other ques­tion first. You can re­spond to a ques­tion by ask­ing what you think is a re­lated ques­tion. Th­ese will then be linked in the Ques­tion UI.

Ask­ing a (smaller) re­lated ques­tion and then mak­ing progress on an­swer­ing it is a great way to help get large re­search ques­tions an­swered by the com­mu­nity.

Com­mu­nity Events Page

What is the LessWrong com­mu­nity event page?

LessWrong is both an on­line and offline com­mu­nity where mem­bers around the globe meet up in per­son for small and large gath­er­ings in­clud­ing lo­cal mee­tups, re­gional re­treats, and con­fer­ences.

The com­mu­nity events page is where LessWrong mem­bers can find each other in the phys­i­cal world and cre­ate events and groups.

You can find the page at www.less­wrong.com/​com­mu­nity, via the left side­bar (desk­top) or bot­tom but­tons (mo­bile).

What are all these cat­e­gories of mee­tups?

The com­mu­nity page dis­plays four non-ex­clu­sive cat­e­gories of events and groups. Th­ese in­clude ex­plic­itly “LessWrong” themed events plus those over­lap­ping and ad­ja­cent com­mu­ni­ties.

Th­ese four in­clude ex­plic­itly LessWrong themed events plus those from over­lap­ping and ad­ja­cent com­mu­ni­ties.

What hap­pens at ra­tio­nal­ity mee­tups?

Depends on the meetup! Some mee­tups fo­cus on for­mal ra­tio­nal­ity prac­tice while oth­ers are just op­por­tu­nity’s for like-minded peo­ple to so­cial­ize—many mee­tups or groups split their time be­tween the two.

What are the larger com­mu­nity events?

The com­mu­nity events page has in­for­ma­tion for large events too. Ex­am­ples in­clude the Bay Area Sum­mer Sols­tice Cel­e­bra­tion, Athena Ra­tion­al­ity Work­shop, Euro­pean Com­mu­nity Week­end, and MIRI Sum­mer Fel­lows Pro­gram.

What re­sources can help me run my lo­cal ra­tio­nal­ity meetup?

There is a re­sources sec­tion on the bot­tom of the com­mu­nity events page. Just scroll to the bot­tom!


What do LessWrong mod­er­a­tors do?

LessWrong aims to be a well-kept gar­den. It is warded by a team of ac­tive mod­er­a­tors who en­sure that dis­cus­sion and con­tent are of high qual­ity, and that be­hav­iors which would diminish the value of LessWrong are pre­vented.

Who can mod­er­ate on LessWrong?

LessWrong has a split mod­er­a­tion sys­tem. Most mod­er­a­tion ac­tivity is performed by LessWrong’s mod­er­a­tion team; how­ever, users who meet cer­tain karma thresh­olds can mod­er­ate their own posts plus set the mod­er­a­tion guidelines that ap­pear on their posts.

Users with over 50 karma can mod­er­ate their own posts when they re­main as Per­sonal blog­posts.

Users with over 2000 karma can mod­er­ate their own posts even when they have been pro­moted to Front­page sta­tus.

What mod­er­a­tion ac­tions can I take on my own posts?

If you meet the karma thresh­olds (50 on Per­sonal blog­posts, 2000 on Front­page posts), you can perform the fol­low­ing mod­er­a­tion ac­tions on your posts:

  • Delete comments

    • Op­tion­ally with a pub­lic no­tice and rea­son.

  • Delete com­ment thread with­out a trace (deletes all com­ments and chil­dren)

    • Op­tion­ally with a pri­vate rea­son sent to the au­thor.

  • Ban users from com­ment­ing on a given post of yours

  • Ban users from com­ment­ing on any of your posts

Be­fore you can mod­er­ate your own posts, you must set mod­er­a­tion style on your post. The fol­low­ing op­tions are available:

  • Easy Go­ing—I just delete ob­vi­ous spam and trolling

  • Norm En­forc­ing—I try to en­force par­tic­u­lar rules (See mod­er­a­tion guidelines)

  • Reign of Ter­ror—I delete any­thing I judge to be an­noy­ing or counterproductive

If you se­lect norm en­forc­ing, you should set your cus­tom mod­er­a­tion policy which will be shown at the top of the com­ment sec­tion and at the bot­tom of the new com­ment form of posts you can mod­er­ate.

We en­courage you to take mod­er­a­tion ac­tions con­sis­tent with the mod­er­a­tion policy you have set on your posts.

What ac­tions and du­ties do the LessWrong team mod­er­a­tors perform?

Moder­a­tors perform the fol­low­ing rou­tine reg­u­lar du­ties:

  • Re­view­ing all new posts and as­signs them spam, per­sonal blog­post, or Front­page (if the au­thor has per­mit­ted Front­page pro­mo­tion).

  • Re­view­ing new users when they first com­ment or post.

  • Delet­ing spam caught by our au­to­matic filters.

  • Select­ing posts for Cu­ra­tion.

  • Keep­ing an eye on dis­cus­sions and en­sur­ing they re­main pro­duc­tive and civil.

Moder­a­tions perform the fol­low­ing less-com­mon ac­tions:

  • Is­su­ing feed­back and warn­ings to users who be­have in ways harm­ful to LessWrong’s dis­course qual­ity and cul­ture.

    • Th­ese will usu­ally start with pri­vate feed­back but es­ca­late to pub­lic warn­ings.

  • Ban­ning users. Usu­ally tem­porar­ily for a few months or a year.

  • Lock­ing com­ment threads (usu­ally tem­porar­ily) if they be­come overly heated and di­vi­sive.

  • Limit­ing the visi­bil­ity of di­vi­sive, heated con­ver­sa­tions on the site to pro­tect the cul­ture and what peo­ple are ex­posed to.

    • In one re­cent in­stance, we moved one com­ment thread on a post to a sep­a­rate post.

    • Moder­a­tors can hide dis­cus­sion threads from the Re­cent dis­cus­sion feed on the home­page.

In ex­tremely ex­treme and ex­ceed­ingly se­vere cases:

  • The LessWrong team may de­cide that we can­not dis­play cer­tain con­tent on the site. In this case, we will likely move that con­tent back to a user’s drafts.

What pow­ers do mod­er­a­tors have?

Moder­a­tors gen­er­ally have ac­cess to the site data, most of this time this is ac­cessed at the re­quest of a user in the pro­cess of de­bug­ging a tech­ni­cal is­sue. We take data pri­vacy se­ri­ously. We don’t just read pri­vate mes­sages.

Note: if a com­ment of yours is ever deleted, you will au­to­mat­i­cally re­ceive a pri­vate mes­sage with its con­tents.

  • The abil­ity to delete comments

    • Usu­ally with pub­lic no­tice and rea­son.

  • The abil­ity to delete com­ment threads with­out trace (deletes all com­ments and all its chil­dren)

    • Usu­ally, with a pri­vate rea­son send to the au­thor.

  • The abil­ity to move con­tent be­tween differ­ent clas­sifi­ca­tions, e.g. Per­sonal blog­post, Front­page post, Cu­rated, Meta (de­p­re­cated cat­e­gory) and drafts.

  • Moder­a­tors can view drafts, but they al­most never will un­less they’re helping you de­bug some­thing.

  • The abil­ity to edit posts.

    • Moder­a­tors usu­ally use this to fix awry for­mat­ting for you, e.g. your LaTeX is screwed up or egre­gious ty­pos, leav­ing a com­ment say­ing they have done so.

  • The abil­ity to ban users from com­ment­ing on given posts or com­ment threads.

  • The abil­ity to ban users from the site (typ­i­cally done tem­porar­ily).

  • The abil­ity to lock com­ment threads.

What is the LessWrong mod­er­a­tion policy/​philos­o­phy?

Un­for­tu­nately, we do not have a re­cent and up to date doc­u­ment that speaks co­her­ently for the whole site, how­ever habryka’s post on Models of mod­er­a­tion is a good start.

Who are the mod­er­a­tors?

The LessWrong core team plus a few oth­ers form the cur­rent mod­er­a­tion team. You can see who they are on the team page.

How do I be­come a mod­er­a­tor?

We are not cur­rently re­cruit­ing any new mod­er­a­tors and there is no cur­rent pro­cess.

That said, mod­er­a­tors would be re­cruited from among those we be­lieve pos­sess ex­cel­lent judg­ment and un­der­stand LessWrong, its pur­pose, its cul­ture, and its val­ues. The best way to demon­strate this would be through con­sis­tently valuable par­ti­ci­pa­tion on LessWrong.

What is LessWrong’s Pri­vacy Policy and Terms of Use?

Our Pri­vacy Policy and Terms of Use can be viewed here.

Note that the Ma­chine In­tel­li­gence Re­search In­sti­tute (MIRI) is the rele­vant le­gal party for this pri­vacy policy and terms of use. When Eliezer Yud­kowsky founded the LessWrong web­site in 2009, he cre­ated it as the prop­erty of MIRI (then named the Sin­gu­lar­ity In­sti­tute for Ar­tifi­cial In­tel­li­gence, aka SIAI).

While we’re at it, we can add that the cur­rent LessWrong team op­er­ates legally as a part of a re­lated or­ga­ni­za­tion, the Cen­ter for Ap­plied Ra­tion­al­ity (CFAR) while re­tain­ing au­ton­omy over its in­ter­nal de­ci­sion-mak­ing and all de­ci­sions about the LessWrong web­site.

For the in­ter­twined his­tory of MIRI and CFAR, see this an­swer to a LessWrong ques­tion.