All of matt's Comments + Replies

matt*20

I'm trying to apply the ITT to your position, and I'm pretty sure I'm failing (and for the avoidance of doubt I believe that you are generally very well informed, capable and are here engaging in good faith, so I anticipate that the failing is mine, not yours). I hope that you can help me better understand your position:

My background assumptions (not stated or endorsed by you):
Conditional on a contribution (a post, a comment) being all of (a) subject to a reasonably clear interpretation (for the reader alone, if that is the only value the reader is optimi... (read more)

There is an important class of claims detailed enough to either be largely accurate or intentional lies, their distortion can't be achieved with mere lack of understanding or motivated cognition. These can be found even in very strange places, and still be informative when taken out of context.

The claim I see here is that orthonormal used a test for dicey character with reasonable precision. The described collateral damage of just one positive reading signals that it doesn't trigger all the time, and there was at least one solid true positive. The wording ... (read more)

matt60

I've read this comment several times, and it seems open to interpretation whether RyanCarey is mocking orthonormal for presenting weak evidence by presenting further obviously weak evidence, or whether RyanCarey is presenting weak evidence believing it to be strong.

Just to lean on the scales a little here, towards readers taking from these two comments (Ryan's and orthonormal's) what I think could (should?) be taken from them…

An available interpretation of orthonormal's comment is that orthonormal:

  1. had a first impression of Geoff that was negative,
  2. then b
... (read more)

As in, 5+ years ago, around when I'd first visited the Bay, I remember meeting up 1:1 with Geoff in a cafe. One of the things I asked, in order to understand how he thought about EA strategy, was what he would do if he wasn't busy starting Leverage. He said he'd probably start a cult, and I don't remember any indication that he was joking whatsoever. I'd initially drafted my comment as "he told me, unjokingly", except that it's a long time ago, so I don't want to give the impression that I'm quite that certain.

accumulated 30 points of karma from what seems to me to be… unimpressive as presented?

I upvoted on the value of the comment as additional source data (IIRC when the comment had much lower karma). This value shouldn't be diminished by questionable interpretation/attitude bundled with it, since the interpretation can be discarded, but the data can't be magicked up.

This is a general consideration that applies to communications that provoke a much stronger urge to mute them, for example those that defend detestable positions. If such communications bring yo... (read more)

The culture of Homo Sabiens often clashes pretty hard with the culture of LessWrong, so I can't speak to how this will shake out overall.

But in the culture of Homo Sabiens, and in the-version-of-LessWrong-built-and-populated-by-Duncans, this is an outstanding comment, exhibiting several virtues, and also explicitly prosocial in its treatment of orthonormal and RyanCarey in the process of disagreement (being careful and explicit, providing handholds, preregistering places where you might be wrong, distinguishing between claims about the comments and about t... (read more)

matt20

Schedule a script to nuke your history every X minutes?

matt00

Dropbox broke old public links with no way I could see of preventing the link rot (https://www.dropbox.com/help/files-folders/public-folder). See https://drive.google.com/drive/folders/0BxlExVbPZSCRdUNaZUFId05YY1k for all of my audio tracks.

matt00

Anyone else having trouble with keyboard input on Lesswrong? (Arrow keys and page up & down work for me on OSX Chrome, Firefox & Safari.)

matt40

I'm polyphasic on Everyman 3 since about March 2011 (Jan and Feb spent unsuccessfully trying to make Uberman work). According to my aging Zeo I get approximately the same REM and SWS as I did on 7.4hrs of monophasic sleep before I adapted. Nearly all of the SWS is in my 3hr core. On Uberman I never achieved enough SWS in my naps to get me through. The adaptation was ridiculously hard - both for how very unpleasant it was and for having to get through that while sleep deprived.

2lululu
I could believe that a 3 hour core could contain a lot of SWS, making it definitely better than Uberman. In those little naps, it's easy to jump into REM and hard to jump into SWS. I was under the impression that 3 hours is still less SWS than the minimum to prevent sleep deprivation symptoms, but I also am endlessly impressed by the capacity of the human brain to adapt to any symptom. Did you do any cognitive functioning tests before/after switching to Everyman?
matt50

Actually, I would suggest not focusing your attention on evolutionary anthropology while you're supposed to be piloting a multi-ton vehicle at high speeds.

When you're driving a daily commute your mind is going to wander unless you have extraordinary focus control / mindfulness training. It's not obvious to me that it's more dangerous to have it directed to evolutionary anthropology than to what you're going to do when you get home (or wherever else it wandered).

3Baughn
There's a difference between accidental mental wanderings and deliberately focusing on a sped-up textbook. If you're listening to a sped-up textbook without focusing on it, I'd say you won't get much out of the experience.
matt40

people with late stage cancers often have enough trouble eating as is (a large fraction actually die of starvation), and getting them to eat anything is an accomplishment. So at that level, for a lot of post-metastasis patients, this will be happening naturally anyways.

Starvation does not equal ketosis. If cancer patients are suffering from nausea and lack of motivation to eat anything, they and their carers may not select high fat low carbohydrate foods that would promote and sustain ketosis and may instead choose simple and easy to digest carbohydrates and sugary treats.

(Your comment upvoted.)

matt170

At TrikeApps our job ads say "Choose an appropriate file format for your resume – we’ll draw conclusions about you from the tools you use". Anyone who expects us to prefer a proprietary file format over LaTeX or PDF is probably applying to the wrong place :)

matt80

They're bold enough to punch through unendorsed aversions, they're not afraid to make fools of themselves, they don't procrastinate, they actually try stuff out, and they push on without getting easily discouraged.

For what it's worth, I'm a pretty successful entrepreneur and I'd say this more like:

They manage on the whole to punch through many of their unendorsed aversions (at least the big ones that look like they're getting in the way), they're just as afraid to make fools of themselves as you are but they have ways of making themselves act anyway m

... (read more)
1John_Maxwell
Thanks! Yes, I agree that it's possible to get better at most of those things through deliberate effort, which includes system-building, and it's a good point that people shouldn't be dissuaded just 'cause it doesn't seem to come to them naturally.
matt290

singularity.org, recently acquired after a rather confused and tacky domain-squatter abandoned it

I would not have described the previous owner of singularity.org as a domain squatter. He provided a small amount of relevant info and linked to other relevant organizations. SI/MIRI made more of the domain than he had, but that hardly earns him membership in that pestilential club.

He sold the domain rather than abandoning it, and behaved honestly and reasonably throughout the transaction.

lukeprog200

Agree.

JoshuaFox120

Thanks, and apologies if I wronged the previous owner. I have edited the post.

matt00

Judging by https://groups.google.com/forum/?fromgroups=#!topic/melbourne-less-wrong/2dFTXTJRHZY and posts here (thanks Maelin) it's going to be a quiet one. I'll bring a couple of games in case we don't get critical mass for a raging storm of rational self improvement.

matt10

If you're doing anything else, you may also want to speed up the playback and shift down the pitch. To achieve that you may use a tool like Audacity (open source, many platforms, Effect > Change Tempo…) or SoundStretch. I use this to automate podcast shifting on my mac.

matt00

(duplicate comment removed)

matt20

Hmm… you're not being moderated. I'll followup on possible causes by PM.

matt30

There is, of course: http://hpmor.com/
Eliezer recommends The World of Null-A (which I've not yet read) and Eliezer and I recommend David's Sling.
Eliezer recommends Lawrence Watt-Evens's fiction. I merely point it out (it's not particularly well written or engaging, but it is nice to watch a protagonist be completely derailed from a quest to set up a business because he sees an opportunity).

matt00

We're planning on discussing ways CfAR (and anyone else) might measure practical rationality to provide feedback for their training.

0toner
Notes from discussion about finding better metrics to measure CfAR's effectiveness [Discussion about what CfAR is trying to teach.] Have to avoid CfAR just measuring how much people are paying attention in their courses Beforehand: write down goals Six months later: measure them against their pre-camp goals Control group Just attending a camp might make it feel like you're making more progress towards your goals. Some goals hard to measure How much throughput are they expecting? 3 camps/month. Measuring income: a lot of people might decide during the camp to change career, or devote their life to effective charity, so income may well drop when your rationality goes up. Also income has a long lag before interventions take effect epicwin: gamify your life. We're going to stop proposing solutions for a while Suggestion to generate ideas individually. Desirable qualities: low stat noise, objective, quantitative etc. What sorts of measures: e.g. based on other's opinions, performance at a task, other classes Could talk about what goals people have? Then we could think about how to measure those particular goals. Money, happier, healthier, etc. Whose perspective are we taking? The consumer of the minicamps. In self-help field, typically sell books or lecture tours based on how good your stuff sounds. Most of it doesn't come with metrics. If there's some of that works (e.g. GTD), then CfAR should be teaching it. [Tangent about quality of scientific literature.] Problem with finding material to teach by looking for stuff whose effectiveness has already been well measured in the literature is that you're setting the bar too high. We don't need to published, so can make do with messier non-laboratory non-ethics-committee-approved measures. Come up with a test you like. Try to teach to that. Break up to discuss: 1. Desirable qualities of metrics 2. Types of test 3. Goals that people might have [We break into groups] I'm with yurifury (in hammock) +
matt00

All of the materials from the July minicamp are available at https://github.com/CfAR/core-materials … for those with access to that private repository. The modules are all in Markdown format and the project includes build scripts that make HTML and PDF "books" that select some or all of the material. The formatting needs some work, and the project needs an owner.

I think the CfAR brass are happy that I give access to Alumni of past minicamps, but I'll need to confirm that before I add anyone. If you're interested in having access to the materials,... (read more)

matt10

It seems not hard to implement naively.

Discussion threads would truncate for new users from new user comments (experienced user comments on new user comments would be invisible to new users).
Our caching gets more complicated.
Many candidate tests for "experienced" seem obvious, but some might be very easy to game (funny comments on HPMOR posts qualify you).

matt60

Sorry people - I should have posted when we did this. Leaving y'all in the dark was unkind.

matt330

Downvoted for putting more than one suggestion in a single comment.

Punish me for this anti-social act if you must, but as one of the dudes who tries to act after reading these suggestions (and tries hard to discount his own opinion and be guided by the community) this practice makes it much harder for me to judge community support for ideas. Does your comment having a score of 10 suggest 2.5 points per suggestion? ~10 points per suggestion? 15 points each for 3 of your suggestions and --35 for one of them (and which one is the -35?)?

Can we please adopt a community norm of atomicity in suggestions?

6JoshuaZ
Sorry, yes, that was obviously a bad action on my part. (Kindly's highly upvoted comment suggests that #1 is of the four the only one of these that people seem to like.)
matt20

And, your body repartitions your sleep on a polyphasic schedule. My sleep really isn't like yours any more. See the bar charts waaaay down the page here: http://trypolyphasic.com/forum/post/8455/#p8455

matt20

We get about as much REM and SWS (deep sleep) as monophasic sleepers - about 90mins each per 24hrs. This is one hypothesis to explain why so many people (me included) have so much trouble adapting to the original Uberman schedule (which, properly adapted, gives you 50+ mins each).

0Risto_Saarelma
Hm, right. So the really deep SWS sleep seems to mostly happen during the first 3 hours of sleep, and the rest is alternation between REM and lighter sleep. Based on that, the Everyman cycle does look a lot more sustainable than full-on Uberman.
matt20

I think 10hrs awake, especially while adapting, is going to be very tough. I think you want to aim for 4 to 6:30 hr periods awake. I know that that requires a nap during normal working hours - as I said in my minicamp unconference presentation (unconference: polyphasic sleep isn't endorsed by CfAR) I think you're going to have to try talking to your employer about it, or sneaking off during a break.

Duplicate http://bit.ly/poly-schedule-tool and play with the times in blue for my advice - the blue cells will turn red if I think what you're attempting is going to be hard to make work.

0Viliam_Bur
Thank you for your answer! If I succeed to have a nap at work, then I have no reason for having awake periods outside of the 4-6 hours range. I would go with this schedule: core 02:45 -- 06:00 (5:00 hours awake: breakfast, commute, work) nap 11:00 -- 11:20 (4:40 hours awake: lunch, work, commute) nap 16:00 -- 16:20 (5:40 hours free time) nap 22:00 -- 22:20 (4:25 hours free time)
matt10

Yah - Wozniak is fairly well known in the polyphasic community for having very strongly held views that are directly contradicted by the experience of polyphasic sleepers. See for example http://www.puredoxyk.com/index.php/2006/11/01/an-attack-on-polyphasic-sleep/.

I did not gather objective evidence of the differences in my cognition before and after polyphasic sleep, but any differences are small enough that they're invisible to me and those who live with me.

2ryjm
I think Wozniak is only evangelical about the Uberman schedule being a horrible idea. He states in his 2010 update that the Everyman 3-hour core sounds "pretty sustainable".
matt110

(Note that there are a few LWers attempting or contemplating polyphasic sleep right now. If you are considering it seriously we'd love your participation in a data collection effort on before and after cognitive performance.)

Polyphasic Sleep
How to have 19-22hrs of fun every day

https://dl.dropbox.com/u/107056/Minicamp2012/PolyphasicSleep/index.html
https://dl.dropbox.com/u/107056/Minicamp2012/PolyphasicSleep.zip

which includes at slides 114 and 115

Theory:
http://trypolyphasic.com/forum/post/8455/#p8455
http://trypolyphasic.com/forum/topic/876/uber-and-everyma

... (read more)
0MalcolmOcean
The nap tracks are no longer available from Matt's dropbox, but fortunately I saved all except the 15min one and have made them available here: https://drive.google.com/drive/folders/0BwP3_2aUw0uPczVqVHpKUFhMd1E
0Paul Crowley
https://dl.dropbox.com/u/107056/Minicamp2012/PolyphasicSleep/index.html This is now a 404. (Came here to re-download the nap tracks, which still work fine :))
0Risto_Saarelma
Another random fun thing I realized. I addition to being a lot like the first half of the reported archaic human bimodal sleep, an Everyman core sleep is also a lot like the first phase of the night's sleep done in the Wake-Back-To-Bed lucid dream induction technique. Just tried a WBTB with 5 hours of initial sleep, and it led to a lucid dream (slept another 3 hours afterwards). 5 hours is pretty close to the Everyman-2 core sleep. I wonder if, in addition to being potentially nice for lucid dreams, adopting a bimodal schedule for a full night's sleep could be used as a stepping stone to adopting an Everyman sleep schedule, since you're training yourself to sleep no more than 4+ hours at a time.
2Risto_Saarelma
Going to mention my crackpot theory on why polyphasic sleeping might end up killing you again: Polyphasic sleep drastically reduces the amount of deep sleep you get. Sleep helps the body heal. Human aging and mortality seem to be modeled well by a Gompertz curve, where the thing that kills you at old age is the body's diminished healing ability, which lets cancer precursors in the body stay unfixed long enough to grow into something that kills you. So for all we know, throwing out 4 hours of deep sleep daily(*) for years on end might make you look a lot more like an 80-year-old, as far as the Gompertz curve modeled mortality is concerned. ETA: (*) All non-REM sleep isn't deep (SWS) sleep, see below. According to online hypnograms of sleep stages, the deepest sleep stages mostly happen during the first three hours of sleep, so a sleep cycle that maintains a 3-hour core sleep should be significantly better than a sleep cycle like Uberman that runs solely on power naps. The 4 hours of deep sleep bit is probably an oversimplification. ETA2: If the first three hours of sleep are the most important for making the body heal, would a sleep cycle where you have two core sleeps daily, for example 3 to 6 AM and 3 to 6 PM, keep you more healthy than a single 8-hour sleep?
0ryjm
I wish I had that schedule calculator earlier - I must have spent a couple of hours googling (#1 failure of my rationality skills) for one because I was sure someone had to have made it, given that all these polyphasic sleepers have oodles of free time.
1aelephant
Thanks for sharing all of the info. You mentioned the effects on cognitive performance, which is my main concern. There is an article here that is skeptical / critical of polyphasic sleep, claiming that it will have negative effects on cognition. I'm curious, do you have data for yourself? Just a subjective assessment? Either would be welcome, although of course the former would be more valuable.
matt10

I'd point out that being a polyphasic sleeper is a major confound here

Agreed.

… we all know that sleep is necessary for learning & long-term memory formation...

With some sleep phases more important than others. High quality evidence is thin on the ground here, but what is available says I'm getting a normal amount of REM and slow wave sleep, and nearly none of the other phases. Wiki (and other sources I've found) suggest that those are the sleep phases important in memory formation. (Note some studies listed on that wiki page have found napping t... (read more)

0gwern
Oh well. Nah; my advice would be simply to start spaced repetition in advance, and look into getting a Zeo for recording sleep data. Not complex, but also not advice they're likely to take.
0JGWeissman
I have emailed them to point them at this thread.
matt30

I tried Bacopa, found in some studies to improve learning and memory. It made me very sleepy in the day following taking it:

10 Aug  Bacopa  Good

11 Aug  Bacopa  Lethargic all day

12 Aug  Bacopa  Lethargic all day

13 Aug  -       Lethargic all day

14 Aug  -       Good

15 Aug  Bacopa  Good

16 Aug  Bacopa  Morning lethargy, clearing after 3hrs

17 Aug  Bacopa  Lethargic all day

18 Aug  Bacopa  Lethargic all morning

19 Aug  Bacopa  Lethargic, less than other days

20 Aug  -       Good

I've stopped.

Important: I'm a polyphasic sleeper: 3hr core, 3x 20mi... (read more)

0Viliam_Bur
Matt, how important is it in Everyman 3 to have your naps distributed evenly through the day? Basicly I am thinking about optimizing for too many things at once -- less sleep time, 8-hour job without sleep (plus commute time), and being somehow synchronized with my significant other; so here are the two schedules I was considering: a) core 03:00 -- 06:00; nap at 16:00; nap at 19:00; nap at 22:00 b) core 22:00 -- 01:00; nap at 6:00; nap at 16:00; nap at 19:00 Do you think any of them would be sustainable? If yes, which would be better: would having the core sleep right before the long wake make it easier?
3gwern
I'd point out that being a polyphasic sleeper is a major confound here: we all know that sleep is necessary for learning & long-term memory formation... Incidentally, do you do spaced repetition? I and Wozniak would be interested in your statistics/database if you started it before the polyphasic sleeping.
1aelephant
Sorry to sidetrack, but is there any chance you could share your experience with polyphasic sleep? I did a search of your submissions & only found that you were interested in it but no history of how you came to actually start doing it successfully.
matt60

Erm… that's security by obscurity in the same way that Wikipedia relies on security by obscurity, right?

0John_Maxwell
Fair enough.
matt70

Ouch - downvoted, presumably because it's a dup. For the record, I raised http://code.google.com/p/lesswrong/issues/detail?id=327 and worked with John to fix the cause of the dup.
Deep down, under the annoying double post character defect, I try to be a good person.

0Matt_Simpson
Tell that to the cops in Alamo...
matt50

Bayesian reformulations welcome.

1albeola
Apologies — I should have taken reinforcement into account and noted that the new algorithm is probably still a lot better than the previous one.
0jsalvatier
This seems like a neat problem. Would it be hard to go from a python function that takes a set of comment upvote downvote counts and returns a ranking to a comment sorting option? If I don't know much about the reddit internals? Also, would it be difficult to get a real dataset of comment counts from LW?
0Kaj_Sotala
Thanks, John and Wes!
matt20

I think "Popular" adds weight to recent comments. This seems to be a much worse way of achieving what "Best" shoots for.

0Elund
Not necessarily. Someone who has already seen the best comments and returns a while later to see what new but good comments have been posted may have a use for it.
matt10

The page expressly says "Supplement your rating by leaving a comment. Comments provide more information, but do not affect the reputation."

If you click "Rate this website" you can rate each scale as you wish. Surely some users choosing different values on the scales is a much simpler explanation than that the site programmers built in a more complicated rating system then lied about it?!

matt00

Can this be true? I don't know how to check it; googling "link:singularity.org" reveals nothing (but the functionality of "link:" seems broken or something; I'd be glad if someone could explain me how it works).

Google Webmaster Tools isn't helping here either: screenshot of webmaster tools

(Webmaster Tools Links to Your Site shows "No data available")

matt00

Your comments don't count, your ratings do: screenshot of WOT page showing relevant controls and explanatory text

(look for the green "Rate this website" link above right of the rating graphic)

-2Viliam_Bur
When you write a comment, you also have to select one of options, such as "Good site", "Useful, informative", "Malicious content, viruses", "Phishing or other scams" etc. I wouldn't be surprised if this kind of data was also used somehow. Perhaps this would explain the small differences in rating -- for example currently "singularity.org" has "Trustworthiness" and "Vendor reliability" 60, and "Privacy" 58, and I suppose that people click the same answer to all of these. (Alternative explanation is that the rating is on the scale, and they simply clicked a different pixel.)
matt00

See https://support.google.com/webmasters/bin/answer.py?hl=en&answer=55281

To find a sampling of links to any site, you can perform a Google search using the link: operator. For instance, [link:www.google.com] will list a selection of the web pages that have links pointing to the Google home page. …

See a much larger sampling of links to a verified site:

  1. On the Webmaster Tools Home page, click the site you want.
  2. On the left-hand menu, click Traffic, and then click Links to Your Site.
matt100

But the important part is this: Someone from SIAI should follow the link "Click here if you own this site", verify the site ownership, and request a review.

Done (with at stretch at the "someone from SIAI" part). Comment above

matt310

I've registered at WoT and requested a reevaluation. We're also making a couple of changes the WoT reevaluation request process seems to suggest are important (like more prominently linking the site's privacy policy).

http://www.mywot.com/en/forum/24394-singularity-org
http://www.mywot.com/en/scorecard/singularity.org

Thanks! So it seems the previous owner of the "singularity.org" domain did some spamming. The other domain "singinst.org" has good rating.

EDIT: See here -- SIAI purchased the domain in April 2011, but the spam complaint was written in January 2012, which is later.

matt100

I want to design a reinforcement schedule in one of our apps. Can anyone link me to some specific guidelines on how to optimise this?

(Reinforce exactly what % of successes (30%? 26%? 8%?)? Reinforce performances in the top 10% of past performances (or the top 12%, or the top 8%?)? How does time factor (if the user hasn't used the app for a week, should I push a reinforcer forward?)?)

0TheOtherDave
I can't, but if you find anything concise and useful, I'd love to hear about it myself. My rule of thumb is to set the threshold so as to reinforce the top 20% or so of performances, and arrange performance frequencies so I'm reinforcing 2-3 times/minute during active training periods. But that's not based on anything. I'll also note that reinforcing higher-tier performances more strongly works really well (though is hard to do consistently by hand), as do very intermittent "jackpots" (disproportional and unpredictable mega-rewards).
matt30

I'm looking for someone to help with me on a paid basis with statistical analysis. I have problems like the following:

1. When to inspect?
I have 10k documents per month steaming to office staff for data entry in offices scattered around the world. I have trained staff at HQ doing inspections of the data entry performed by the office staff, detecting errors and updating fields in which they detected errors. I will soon have random re-checking by HQ inspectors of entries already checked by other HQ staff.
The HQ staff currently detect errors on ~15% of docum... (read more)

matt30

I pushed all the way through. I'm signed up with Alcor, but feel very much as you do about how hard signup was, and how unlikely it is that Alcor will survive very long. I know only one other Australian who tried to sign up, and he also gave up in frustration.

(I've tried to volunteer my time and efforts to Alcor, and they can't organise enough to accept my help.)

matt10

There's some weight behind this proposal. Consider modifing the Anti-Kibitzer (http://lesswrong.com/lw/1s/lesswrong_antikibitzer_hides_comment_authors_and/1hvk) to do what you want (or adding a ticket to request same - http://code.google.com/p/lesswrong/issues/list).

0witzvo
Thanks for mentioning these things. I didn't know the Anti-Kibitzer existed. Shouldn't this be basic information for newbies? Did I miss it? [more when I have time, maybe]
0wedrifid
My proposal was essentially the proposal "keep the status quo, let upvotes hide downvotes, no more information please!". I take it you are considering the extended option "don't let me see my karma score below 0 at all", which is also an idea with potential. I'm sure I'll be tempted to next time I endure an irritating number of downvotes for obnoxious seeming reasons. I've modified the anti-kbitzer script previously to hide my own karma and inbox notifications to reduce the addictive feedback mechanism.
matt00

Test comment:

  • this
  • that
    Next paragraph
matt20

It's too bad that automatic wiki editing privileges don't come with a certain level of karma

Hmmm... you know that wouldn't be too hard to arrange. Keeping the passwords in sync after a change to one account would be much more work, but might be ignorable.

2John_Maxwell
Ideally it seems like you would get your wiki authentication cookie automatically after logging into Less Wrong, so you could log in once and use both. I don't know if that changes things regarding passwords.
Load More