Less Wrong Study Hall - Year 1 Retrospective

by Error11 min read18th Mar 201430 comments


Postmortems & Retrospectives

Some time back, a small group of Less Wrongers collected in a video chatroom to work on…things. We’ve been at it for exactly one year as of today, and it seems like a good time to see what’s come of it.[1] So here is what we’ve done, what we’re doing, and a few thoughts on where we’re going. At the end is a survey taken of the LWSH, partly to be compared to Less Wrong proper, but mostly for fun. If you like what you see here, come join us. The password is “lw”.

A Brief History of the Hall

I think the first inspiration was Eliezer looking for someone to sit with him while he worked, to help with productivity and akrasia. Shannon Friedman answered the call and it seemed to be effective. She suggested a similar coworking scheme to one of her clients, Mqrius, to help him with akratic issues surrounding his thesis. She posted on Less Wrong about it, with the intent of connecting him and possibly others who wanted to co-work in a similar fashion. Tsakinis, in the comments, took the idea a step further, and created a Tinychat video chatroom for group working. It was titled the Less Wrong Study Hall. The theory is that it will help us actually do the work, instead of, say, reading tvtropes when we should be studying. It turned out to be a decent Schelling point, enough to form a regular group and occasionally attract new people. It’s grown slowly but steadily.

Tinychat’s software sucks, and there have been a couple of efforts to replace it. Mqrius looked into OpenMeetings, but it didn’t work out. Yours truly took a crack at programming a LWSH Google Hangout, but it ran aground on technical difficulties. Meanwhile the tinychat room continued to work, and despite nobody actually liking it, it’s done the job well enough.

Tinychat is publicly available, and there have been occasional issues with the public along the way. A few people took up modding, but it was still a nuisance. Eventually a password was placed on the room, which mostly shut down the problem. We did have one guy straight out guess the password, which was a…peculiar experience. He was notably not all there, but somehow still scrupulously polite, and left when asked. I don’t think I’ve ever seen that happen on the Internet before.

A year after the Hall opened, we have about twenty to twenty-five regulars, with an unknown number of occasional users. We’re still well within Dunbar’s number, so everybody knows everybody else and new users integrate quickly. We’ve developed a reasonably firm set of social norms to guide our work, in spite of not having direct technical control nor clear leaders.

The story thus far has been a pretty good one. I’m looking forward to Year Two.

Current MO and Social Norms

No one went out of their way to come up with these and nobody is really enforcing them, but it’s how the group currently behaves.

We adopted the Pomodoro Technique early on to organize our work. We run 32-minute pomodoros instead of the usual 25 (I have no idea how this custom originated; I'm told it was Dentin's idea though), interspersed with eight minute breaks that tend to run over as often as not. Sometimes, we’ll run a longer or shorter pomo if someone requests it. When you enter the room, ask for the current time. Usually at least one person will notice and respond with the endpoint of the current pomo, expressed in minutes-past-the-hour to be applicable across time zones.[2]

Don’t talk during pomos without good cause. Asking for the current time is good cause; so is saying a brief goodbye when you leave. If someone else is talking enough to be distracting, it’s acceptable to ask them to save it for the next break. Otherwise, insert nose in grindstone.

During breaks, talk as much as you want. Talking about what you’re working on is explicitly encouraged. Bragging about something you’ve just completed is also explicitly encouraged. In the absence of those two things, though, it’s okay to just be social.[3]

Anyone who notices that the break has ended can start the next pomo just by stating its length and end time. Usually one or two self-selected people will keep the rest of the room on track in this regard. If the break runs over and you’re the first to notice, speak up even if you don’t want to be responsible for the clock; it’s good form to cut the chitchat quickly once someone has noted that it’s time to get back to work.

Most of us keep our cameras on in the room, and I encourage new users to do the same. Having someone watch you work seems to improve motivation and reduce the temptation to slack. Your presence will be more obvious, which helps if you happen to be new and are scrolled off the bottom of the user list. It also encourages others to interact with you; socialization during breaks is easier and works better when you can see people's expressions. As noted in the survey below, that sort of social reinforcement is a big plus. In lieu of a camera, some users display their working desktop instead, but this is rare and seems to require a third party program. If you like privacy, keeping everything off is okay too; just be aware that you’re avoiding a useful tool.

Bullet point version of the above:

  • Say hello and ask for the current time when entering.
  • Don’t talk during pomos.
  • Do talk during breaks.
  • Talking about work is encouraged.
  • Bragging about work is encouraged.
  • Don’t turn your mic on.
  • But do turn your camera or desktop view on if you want.

Personal Reflections

It’s been an interesting first year. I’m part of a tribe as it begins to cohere, and I like it.

Is it effective? Good question. I enjoy the environment but that doesn’t mean it’s succeeding at what it’s supposed to do: help akratics Get Stuff Done. I feel like it’s effective, but being sure of that is hard. I've certainly written more in the last year than the previous ten. I do use a couple of other productivity hacks, especially Beeminder, and in fact I frequently work on Beeminded tasks in the Hall. Someone (jkadlubo?) mentioned that they didn’t like Beeminder because it was all stick and no carrot. For me, the study hall is the carrot. I like working alongside other people.

We have a lot of students, which is unsurprising given that we splintered off Less Wrong and Less Wrong also has a lot of students. This makes me wonder what will happen when our users begin to graduate.

Tinychat, as a technology, sucks. Unfortunately it’s the best existing software that does what we want, by virtue of being the only existing software that does what we want (Google Hangouts, the most obvious alternative, does not work for our purposes as mentioned above), and at least two attempts to replace it have failed. I believe this is because, while Tinychat is annoying, it’s not annoying enough for anyone to put in the relatively large effort that seems to be required to replace it – especially since the room is self-selected for akratic problems in the first place. The most promising alternative tech I can find is XMPP Muji, a standard for multi-user video chat services, but the standard’s approval is deferred and I’m not sure why. A cursory Google search does not turn up any software that implements it.

One of the problems brought up when the room was first proposed was that people might end up using the Hall to socialize rather than work. The concern was justified, at least partially; we routinely run over breaks talking. But we seem to have countered that with a fairly strong norm of stopping the break as soon as someone notices. As we’ve all gotten to know each other better, the atmosphere during the breaks has become quite social. I would blame our break-overruns on this, except I’m pretty sure that has been happening since the start. Nevertheless there has been some concern that the more-social atmosphere has been distracting people from actually working. I asked about this in the survey, but the answers were heavily slanted towards “not terribly distracting.” (no higher than a 3 on a 1–5 scale, and only three 3’s) This leads me to believe that there are more people concerned that others might be bothered than there are others who are actually bothered. It is possible there is a selection effect whereby the people irritated leave and never took the survey; but the change in atmosphere has been relatively recent, and I haven’t noticed an exodus of regulars. Additionally, the single largest draw reported is social reinforcement for working. I think the social atmosphere is probably a net positive, although it would be good if we could technologically enforce break limits. Several other people noted that in the suggestion box; unfortunately it is not possible without replacing Tinychat. Which was also a common item in the suggestion box.

We’ve lost some users along the way (including the original creator, I think), but a near-majority of the room has been around since almost the beginning. That seems to suggest that people who try it out for more than a few cursory sessions stick around. We’ll begin to see if that’s the case next year, with repeat survey-takers.

Survey and Results

Yvain’s Less Wrong 2013 Survey results came out as the one year mark was approaching, and it seemed to me that it would be fun, if maybe not all that useful, to collect the same sort of information about the Hall. So, I did. The survey was open for about three weeks and publicized during Hall breaks by myself and a few other people. You can see the questions used here. I flagrantly plagiarized many of the questions from Yvain’s survey (with permission, and thank you), because it seemed like comparing the Hall to Less Wrong proper would be interesting. I took out a few sections that I didn’t feel were relevant and added a couple of Study Hall specific sections. There is also some silliness at the end.

We had 23 responses. I counted 22 unique users in the room over the period the survey was open. Assuming there are at least some people I did not see during that time (owing to differing schedules) I would guess our actual recurring population is around 30, of which perhaps 20 show up regularly. I have noticed an influx of new faces in the last week or two; I sort of wish I’d run this a month later.

With such a small population it would be fairly easy for community members to de-anonymize people. Respondents were warned of that, but I would still appreciate it if people didn’t go out of their way to try and connect responses to names.

The results are below. My comments are in brackets.



Mean 23.4
Stddev 4.2
Min 17
1st quartile 20.5
2nd quartile 23
3rd quartile 26.5
Max 32

[[ Our youngest user is 17 and our oldest is 32, at least among respondents. I feel old now. ]]


Australia 2 9%
Canada 1 4%
Finland 0 0%
France 0 0%
Germany 8 35%
Israel 0 0%
New Zealand 0 0%
Poland 2 9%
Russia 0 0%
United Kingdom 1 4%
United States 4 17%
Other 5 22%

[[ I don’t know why we’re so hugely concentrated in Germany. I think it might be one cluster of people all referring each other. ]]


Asian (East Asian) 0 0%
Asian (Indian Subcontinent) 0 0%
Middle Eastern 0 0%
Black 0 0%
White (non-Hispanic) 21 100%
White (Hispanic) 0 0%
Other 0 0%

[[ I’m sure I’ve seen at least one non-white person…but still. ]]

Sex, Gender, Relationships

I considered not including this section as irrelevant and possibly touchy, especially given how easy it would be to de-anonymize people. Someone (I think tkadlubo) pointed out that Less Wrongers in general are interested in relationship hacking. I ended up making it optional, but everybody answered it anyway.


Male 18 78%
Female 5 22%
Other 0 0%


Male (cisgender) 18 78%
Female (cisgender) 5 22%
Male (transgender f->m) 0 0%
Female (transgender m->f) 0 0%
Other 0 0%

Sexual Orientation

Heterosexual 16 70%
Homosexual 0 0%
Bisexual 2 9%
Asexual 1 4%
Other 4 17%

Relationship Style

Prefer monogamous 5 22%
Prefer polyamorous 8 35%
Uncertain / no preference 10 43%
Other 0 0%

[[ All but two of our polyamorous folks were from Europe, and one of those two was me. Clearly I need to move to Europe. ]]

Number of Current Partners

0 13 62%
1 3 14%
2 5 24%

Relationship Goals

...and currently looking for more relationship partners 13 59%
...and currently not looking for more relationship partners 6 27%
Other 3 14%

[[ Bonus points to whoever answered “that’s complicated.” Also, if I’d been thinking I would have included an option for “not looking, but open to the possibility.” ]]

Relationship Status

Single 14 61%
Relationship 6 26%
Married 3 13%
Other 0 0%


0 20 87%
1 1 4%
2 2 9%

More Children

Yes 6 27%
No 8 36%
Uncertain 8 36%

Work and Education

Work Status

Student 14 61%
Academics (on the teaching side) 0 0%
Self-employed 2 9%
Independently wealthy 0 0%
Non-profit work 0 0%
For-profit work 5 22%
Government work 0 0%
Homemaking 1 4%
Unemployed 0 0%
Other 1 4%


Philosophy 0 0%
Finance / Economics 2 9%
Statistics 0 0%
Computers (AI) 0 0%
Computers (other academic, computer science) 1 4%
Computers (practical: IT, programming, etc.) 8 35%
Physics 1 4%
Biology 3 13%
Medicine 1 4%
Mathematics 2 9%
Neuroscience 2 9%
Other "hard science" 0 0%
Psychology 0 0%
Other "social science" 1 4%
Engineering 2 9%
Business 0 0%
Law 0 0%
Art 0 0%
Other 0 0%


None 1 5%
High School 10 45%
2 year degree 2 9%
Bachelor's 2 9%
Master's 6 27%
PH D. 1 5%
MD/JD/other professional degree 0 0%
Other 0 0%

Less Wrong

I wanted to get an idea of how our splinter tribe relates to Less Wrong as a whole, so I included enough of a subset of that section from the main survey to do so.

Less Wrong Use

Doesn't use LW 1 5%
Lurker (no account) 6 29%
Lurker (account) 1 5%
Poster (comments only) 5 24%
Poster (discussion) 8 38%
Poster (Main) 0 0%

Time in Community (LW, years)

Mean 1.9
Stddev 1.06
Min 0.2
1st quartile 1
2nd quartile 2
3rd quartile 2.75
Max 4

Karma Score

Mean 183.1
Stddev 265.5
Min 0
1st quartile 0.5
2nd quartile 60
3rd quartile 251
Max 800


Yes, regularly. 3 15%
Yes, once or a few times. 6 30%
No. 11 55%

Community (LW)

Yes, all the time. 4 20%
Yes, sometimes. 8 40%
No. 8 40%

Took the main LW survey?

Yes 12 60%
No 8 40%

Less Wrong Study Hall

The meat lies here.

Time in Community (LWSH)

Less than a month. 3 13%
1 - 3 months 4 17%
4 - 6 months 3 13%
6 - 9 months 4 17%
9 - 12 months 9 39%


Every day 8 35%
Several times a week 9 39%
Once or twice a week 1 4%
Less than once a week 1 4%
It varies 3 13%
I haven't been here long enough to form a pattern 1 4%

Time in the Hall (per visit, in minutes)

Mean 226.1
Stddev 137.1
Min 60
1st quartile 120
2nd quartile 180
3rd quartile 300
Max 540

[[ NINE HOURS??! O_O I salute you. ]]


Academic studies 18 27%
Personal projects 21 32%
Deliberate practice (e.g. learning guitar) 4 6%
Work for an employer 5 8%
Chores, paperwork, or other necessities 17 26%
Other 1 2%

[[ I was surprised so many people use it for chores and the like. I do, but I thought I was just about the only one. On the plus side, my home actually gets cleaned up regularly now… ]]

Usage 2 (most important)

Academic studies 12 55%
Personal projects 5 23%
Deliberate practice (e.g. learning guitar) 0 0%
Work for an employer 3 14%
Chores, paperwork, or other necessities 2 9%
Other 0 0%

[[ Unsurprisingly, academics is the big thing. ]]


Social reinforcement for working. 22 33%
Social punishment for not working. 7 10%
Camera-induced self-consciousness when working. 15 22%
Distraction reduction via group pomodoros. 18 27%
Other 5 7%

Draw 2 (most important)

Social reinforcement for working. 14 61%
Social punishment for not working. 0 0%
Camera-induced self-consciousness when working. 1 4%
Distraction reduction via group pomodoros. 7 30%
Other 1 4%

[[ The carrot is more appreciated than the stick. ]]


Yes, always. 11 48%
Yes, sometimes. 10 43%
Rarely 1 4%
Never 1 4%

[[ I should have asked about desktop sharing too. I have a suspicion that it works even better than cameras, but only one or two people do it. ]]

Camera Off (why?)

This was a freeform response about why users leave their cameras off. Most users do leave their cameras on so there wasn’t a lot to go on here. Included themes were privacy (for oneself or others nearby), temporary non-working states (as when someone is eating), technical difficulties, or just generally not feeling sociable.

Time Zone

UTC+10:00 2 9%
UTC+02:00 1 4.5%
UTC+01:00 13 60%
UTC+00:00 1 4.5%
UTC-05:00 5 23%

[[ Apparently all our American users are on the east coast? Perhaps the West Coast people that might be interested already have access to the main real-life Less Wrong community/CFAR people, and thus don’t need the Hall. But I thought we had at least one guy in the West. I am confused. ]]

Temporal Habits (Weekdays)

Mornings (6am-12pm) 6 15%
Afternoons (12pm-5pm) 9 22%
Evenings (5pm-10pm) 18 44%
Late night/very early morning (10pm-6am) 6 15%
Too variable to say 2 5%
I don't use the room during the week 0 0%

Temporal Habits (Weekends)

Mornings (6am-12pm) 4 10%
Afternoons (12pm-5pm) 11 26%
Evenings (5pm-10pm) 13 31%
Late night/very early morning (10pm-6am) 6 14%
Too variable to say 7 17%
I don't use the room on weekends 1 2%

[[ I meant to use time zone information and logon windows to figure out when the room is populated and not, but never got around to it. Maybe next year. Or someone else could do the honors. ]]


The initial announcement 9 39%
Other comments or posts on Less Wrong 7 30%
Referred by a friend or partner 6 26%
Referred by a non-LW blog 0 0%
Referred by a search engine 0 0%
Other 1 4%

Interaction (outside the Hall)

Yes, regularly. 9 39%
Yes, sometimes. 7 30%
No. 7 30%

Interaction 2 (in person only)

Yes, regularly. 5 22%
Yes, sometimes. 5 22%
I've met a few people in person once or twice. 2 9%
No. 11 48%

[[ It looks like more people here get involved in-person than Less Wrong as a whole, but it looks like it might be because all our Germans know each other in real life. ]]


Yes. 5 22%
I didn't meet them through the Hall, but they come there now. 2 9%
No. 16 70%

[[ That Yes% is much higher than Less Wrong proper, even though our gender balance isn’t much better. Shame our sample size is far too small to tell if it matters. I predict not. ]]

Base Akrasia (1–7 scale, 1 being no akrasia, 7 being functionally paralyzed)

1 1 5%
2 0 0%
3 4 18%
4 5 23%
5 6 27%
6 5 23%
7 1 5%

Akratic Impact (Akrasia in the Hall, same scale as previous)

1 1 5%
2 9 41%
3 8 36%
4 3 14%
5 1 5%
6 0 0%
7 0 0%

[[ Self-reporting being what it is, these numbers aren’t terribly reliable. But they’re all we’ve got for the moment and they do show an improvement. My thanks to whoever suggested that two absolute scales were a better idea than absolute + improvement. ]]

Hedonic Impact (Hedonic improvement in the Hall, 1–5, 1 being no/negative impact, 5 being large impact)

1 0 0%
2 0 0%
3 3 13%
4 7 30%
5 13 57%

[[ If I’d been thinking I would have measured hedons in the same manner as akradons. Oh well, next year. ]]

Distractions (percentage)

Mean 29%
Stddev 15
Min 10%
1st quartile 18.5%
2nd quartile 30%
3rd quartile 35%
Max 70%

[[ I am not sure if these numbers are low or high. I would have to compare them to distractibility outside the Hall, but that is much harder to notice and track unless you are explicitly looking for it. ]]

Distraction Cause

People talking or otherwise drawing attention in the Hall during the pomo 0 0%
Digital interruptions (email or IM) 4 17%
In-person interruptions (family or friends wanting attention) 7 30%
Spontaneous web browsing or other computer use. 11 48%
Other 1 4%

[[ I should have done the same checkbox/most-important thing here that I did with the Draw and Usage questions. Unsurprisingly, spontaneous web browsing topped the list by a large margin. ]]

Overwork (percentage)

Mean 34.7
Stddev 24.8
Min 0%
1st quartile 12.5%
2nd quartile 33%
3rd quartile 55%
Max 90%


Yes 15 65%
No 8 35%

[[ Since we’re here for a reason, it seemed good to find out whether people are succeeding at their purpose. Hence this question and the next one. ]]

Accomplishment Examples

This was a freeform question and not everyone answered it. Most answers again involved academic work and study, including one brave soul who studied for and passed final exams for classes on which he’d skipped most of the lectures. I’m not certain if I should applaud that or not. Artistic endeavors were the second most common response, and showed up in the form of painting, manuscripts, fanfiction, and blog posts. One person has been using the Hall to translate HPMoR chapters, and another wrote an open source library. And one of our users landed a new job, for which they’d written their CV using the Hall. Congratulations!

Miscellaneous Feedback

Social Atmosphere

1 9 41%
2 10 45%
3 3 14%
4 0 0%
5 0 0%

[[ Some people have suggested that the recent uptick in sociability and silliness during breaks would distract people. This is an attempt to measure it. More in the personal reflections section. ]]

Suggestion Box

This was a freeform question, and most respondents didn’t answer. The most common response by a fair margin was better enforcement of pomodoros, either technologically or socially. There were also multiple calls to replace tinychat.

Despite my complaining, I would be surprised if tinychat was replaced anytime soon. Better social enforcement of pomos should be possible, though. Remember that anyone can track the clock if they want.


Since we’re here to fight akrasia, measuring it seemed good. I copied almost the entire Akrasia section from Yvain’s survey. A couple of people noted some ambiguous questions. I wanted to keep the data comparable to LW proper, so rather than modifying his questions I added clarified ones in between.


Yes 11 50%
No 11 50%

Akrasia: Current

Yes 2 9%
No 20 91%

Akrasia: Illness

Depression 3 14%
Bipolar disorder 2 10%
Schizophrenia 0 0%
Obsessive-compulsive disorder 0 0%
ADHD 0 0%
PTSD 0 0%
Autism or autism spectrum disorder 2 10%
Generalized anxiety disorder 1 5%
Social phobia 0 0%
Any personality disorder 0 0%
None 12 57%
Other 1 5%

Akrasia: Medicines 1

No 14 67%
Other 7 33%

[[ Someone pointed out that the question as originally stated didn’t specify “medications taken specifically to combat akrasia,” so I repeated the question with that clarification for both this and the supplements question. ]]

Akrasia: Medicines 1.5

No 16 80%
Other 4 20%

Akrasia: Medicines 2

1 2 25%
2 2 25%
3 3 38%
4 0 0%
5 1 13%

Akrasia: Supplements 1

No 11 55%
Other 9 45%

Akrasia: Supplements 1.5

No 14 74%
Other 5 26%

Akrasia: Supplements 2

1 0 0%
2 1 13%
3 4 50%
4 0 0%
5 3 38%

Akrasia: Therapy 1

No 10 63%
Other 6 38%

Akrasia: Therapy 2

1 1 9%
2 3 27%
3 1 9%
4 1 9%
5 5 45%

Akrasia: Meditation 1

No 9 45%
Other 11 55%

Akrasia: Meditation 2

1 2 17%
2 4 33%
3 4 33%
4 1 8%
5 1 8%

Akrasia: Elsewhat 1

No 9 50%
Other 9 50%

Akrasia: Elsewhat 2

1 3 27%
2 3 27%
3 3 27%
4 1 9%
5 1 9%

Akrasia: Communication

Yes 16 76%
No 5 24%


Some things suggested by our users when I was putting the survey together. This is just for fun.

Tinychat Hatred

1 4 18% (thousand burning suns)
2 1 5%
3 7 32%
4 3 14%
5 7 32% (Emperor Palpatine)

Tinychat Screams (number of)

I don’t know how to correctly calculate averages or related statistics when some numbers are written in Knuth notation, so there’s not much to report here. My favorite answer was Pi^^^42.


No 5 24%
Yes, but only one 2 10%
Yes, more than one 13 62%
Tons 1 5%

Stuffies on Camera

Yes 10 48%
No 11 52%

[[ For some reason lots of us have stuffies. ]]

And that's the end of it. My thanks to Lachouette for feedback on a draft of this post, and to the Hall itself for keeping me on track while I was writing about it. Those who want to join us, consider yourselves invited.

  1. I’m posting this to Main, but given that our splinter tribe is a very small fragment of Less Wrong, I’m not sure if it’s appropriate to Main. If a mod wants to move it, I won’t object.  ↩

  2. Someday we’ll have users with bizarre half- or quarter-hour timezone offsets and this will all fall down.  ↩

  3. While there’s no explicit rules about discussion topics, it’s probably a good idea to observe Less Wrong’s anti-politics norm. The few times I’ve witnessed conflicts between regulars, it was when politics came up in some way.  ↩