Posts

How do humans assign utilities to world states? 2015-05-31T20:40:33.151Z · score: 2 (3 votes)
Quotes Repository 2015-02-10T04:36:19.814Z · score: 1 (12 votes)
Musings on non-stability of mind-states, etc. 2014-11-02T02:43:39.423Z · score: 5 (10 votes)
Mistakes repository 2013-09-09T03:32:20.194Z · score: 25 (26 votes)
Open Thread, June 16-30, 2013 2013-06-16T04:45:04.761Z · score: 3 (4 votes)
Health/Longevity Link List 2013-05-05T03:17:30.281Z · score: 4 (11 votes)
Book Summary: Willpower by Baumeister, Tierney 2012-05-24T23:34:25.446Z · score: 22 (23 votes)
AI Box Log 2012-01-27T04:47:49.429Z · score: 16 (21 votes)
[LINK} Bayes' Theorem in New York Times 2011-08-07T03:55:47.315Z · score: 3 (4 votes)
Get data points on your current utility function via hypotheticals 2011-04-24T18:44:35.928Z · score: 1 (6 votes)
[SEQ RERUN] The Proper Use of Humility 2011-04-22T11:48:48.616Z · score: 8 (9 votes)
[SEQ RERUN] ...What's a bias, again? 2011-04-21T19:18:16.923Z · score: 7 (8 votes)
The right kind of fun? 2011-04-16T23:20:51.350Z · score: 4 (6 votes)
Hunger can make you stupid 2011-04-13T16:17:49.897Z · score: 11 (16 votes)
Ithaca Meetup? 2011-03-30T00:51:11.793Z · score: 2 (3 votes)
When to scream "Error!" 2011-02-26T17:40:35.688Z · score: 10 (13 votes)
Tool for combating undue hesitation 2011-02-20T05:19:16.058Z · score: 2 (3 votes)
A Possible Solution to Parfit's Hitchiker 2011-01-28T19:21:15.400Z · score: -5 (13 votes)

Comments

Comment by dorikka on Open Thread, Aug 29. - Sept 5. 2016 · 2016-09-04T18:55:49.404Z · score: 0 (0 votes) · LW · GW

Thanks. I haven't used liquid products much before. Anything you've noticed that's significantly different in terms of onset time, effect duration, etc?

Comment by dorikka on Open Thread, Aug 29. - Sept 5. 2016 · 2016-09-04T18:01:19.435Z · score: 1 (1 votes) · LW · GW

Anyone know where I can find melatonin tablets <300 mcg? Splitting 300 mcg into 75 mcg quarters still gives me morning sleepiness, thinking smaller dose will reduce remaining melatonin upon wake time. Thanks.

Comment by dorikka on Attention! Financial scam targeting Less Wrong users · 2016-03-02T22:14:02.116Z · score: 2 (4 votes) · LW · GW

Surely, as rationalists, we should

So awkward it hurts that this is even a thing.

Comment by dorikka on Attention! Financial scam targeting Less Wrong users · 2016-03-02T07:14:22.411Z · score: 3 (3 votes) · LW · GW

Thanks for noting this.

Comment by dorikka on Open Thread, Dec. 28 - Jan. 3, 2016 · 2015-12-27T16:26:33.281Z · score: 1 (3 votes) · LW · GW

Aren't new open threads usually posted on Mondays? Today is the 27th, not 28th.

Comment by dorikka on Open thread, Nov. 23 - Nov. 29, 2015 · 2015-11-24T02:18:33.983Z · score: 2 (2 votes) · LW · GW

404: Generalized model not found

Comment by dorikka on Open thread, Oct. 19 - Oct. 25, 2015 · 2015-10-20T02:49:15.045Z · score: 2 (2 votes) · LW · GW

Any particular evidence in favor of this approach, anecdotal or otherwise?

Comment by dorikka on Crazy Ideas Thread - October 2015 · 2015-10-08T13:01:39.494Z · score: 1 (1 votes) · LW · GW

Seem to be implying that you are more likely to be in a simulation if historixcally impt. Interesting

Comment by dorikka on Wear a Helmet While Driving a Car · 2015-09-26T16:13:32.804Z · score: 0 (0 votes) · LW · GW

No other source, but keep in mind that helmets are tuned for a certain force level. Too durable and helmet does not reduce peak force as it does not crush. Too weak and it crushes quickly, again with little reduction in peak force. This should just empasize to use the 25% number here though since the forces are more representative.

Redacting "won't do much of anything" except as implied by 25%, but keep in mind that if peak accelerations are much higher than the given case, the helmet will be less effective due to the above. This may or may not be the case in car crashes depending on speed.

Comment by dorikka on Wear a Helmet While Driving a Car · 2015-09-26T15:58:48.827Z · score: 1 (1 votes) · LW · GW

Interesting - thanks for checking this. If the Severity Index is claiming no significant damage below 1200, I think it may be incorrect or may have a different criterion for severe damage. Some helmet standards seem to be fairly insensitive, only accounting for moderate or severe brain injury whereas MTBI can have long lasting effects. Yes, I discount Severity claims as the metric does not appear to give reasonable results. 188g is a crapload of linear acceleration, but metric puts it under threshold...I dont buy it, so am left to judge on peak linear accel instead (shame that rotational accel was not measured...)

Comment by dorikka on Wear a Helmet While Driving a Car · 2015-09-21T11:29:00.249Z · score: 0 (0 votes) · LW · GW

The data is posted above, unlikely to get around to Dropboxing it so I can link (as it was from an email). I agree with you re body movement in a vehicle collision. However, at some point your body would stop. If your head hit something while your body was in motion, thr impacted object would likely have enough strength to bring the head to an abrupt halt. (Contrast with a knife being punched through paper mache - I would expect the force on the lnofe to be much lower than if hitting concrete, as it would go through the paper mache without much velocity change.)

Comment by dorikka on Open Thread August 31 - September 6 · 2015-09-03T16:00:04.815Z · score: 2 (4 votes) · LW · GW

I am curious about your terminal goal here.

Comment by dorikka on Yudkowsky's brain is the pinnacle of evolution · 2015-08-25T04:33:55.495Z · score: 4 (3 votes) · LW · GW

shrug The pdf for sincerity looks bimodal to me.

Comment by dorikka on Yudkowsky's brain is the pinnacle of evolution · 2015-08-25T04:32:07.556Z · score: 6 (5 votes) · LW · GW

This is the most tantalizing thread on the page.

Comment by dorikka on Yudkowsky's brain is the pinnacle of evolution · 2015-08-25T04:31:14.007Z · score: 3 (2 votes) · LW · GW

What is this, and why is it here?

(Original response was remarkably vehement, rather like I found a pile of cow dung sitting on my keyboard. Interesting.)

Comment by dorikka on Open Thread - Aug 24 - Aug 30 · 2015-08-25T04:20:26.219Z · score: 0 (0 votes) · LW · GW

Thanks. How does one go about learning more about this, preferably while encountering minimal bullshit on the way?

Comment by dorikka on List of common human goals · 2015-08-24T16:27:16.824Z · score: 1 (1 votes) · LW · GW

Thanks for posting this. Just a quick note, many of the things listed above I would consider may be "common" terminal values, not goals. Might just be a wording thing, but I think of goals as instrumental, with values propagating to actions via the hierarchy values->strategies->campaigns->goals->actions.

Convergent instrumental goals might be an interesting collection as well.

Comment by dorikka on Open Thread - Aug 24 - Aug 30 · 2015-08-24T16:21:25.425Z · score: 1 (1 votes) · LW · GW

A few nutrition-related questions:

  • Why does Soylent 2.0 have so much fat? They appear to be going for 45% of calories from fat, whereas the typical recommendation is 10%-35%.

  • Why does the Bulletproof stuff include so much saturated fat? It appears that the consensus is that saturated fat significantly increases blood cholesterol and arterial plaque formation - curious why such a deviation here.

Comment by dorikka on Open Thread - Aug 24 - Aug 30 · 2015-08-24T12:57:54.105Z · score: 1 (1 votes) · LW · GW

Personal experience that it is useful or just from the indirectly linked papers?

Also, note that it may potentiallly insta-fuck your liver.

Comment by dorikka on Open thread, Aug. 17 - Aug. 23, 2015 · 2015-08-23T03:47:14.599Z · score: 0 (0 votes) · LW · GW

A few nutrition-related questions:

  • Why does Soylent 2.0 have so much fat? They appear to be going for 45% of calories from fat, whereas they typical recommendation is 10%-35%.

  • Why does the Bulletproof stuff include so much saturated fat? It appears that the consensus is that saturated fat significantly increases blood cholesterol and arterial plaque formation - curious why such a deviation here.

Comment by dorikka on A list of apps that are useful to me. (And other phone details) · 2015-08-23T03:00:44.829Z · score: 0 (0 votes) · LW · GW

Thanks for writing this. A few notes:

  • I find Evernote to be an exceptionally great notetaking app.

  • If you end up using Google Calendar, I like Smooth Calendar as a widget that shows a few appts and lets you click through to the full calendar.

  • I previously had an S4, now an S5. I use the InvisibleShield Glass screen cover - people seem to keep finding ways to damage the glass on their phone screen, so the durable cover might pay dividends. (And already did on my S4, when I dropped it about a meter onto slate. I currently have a BodyGlove phone case.

  • I am totally mystified as to how you "go through" a phone case every month - I tend to use rubber ones or semi-flexible plastic, so the phone electronics would probably be mauled by the shock before the case suffered significant damage. Do you use a very different type of case?

Comment by dorikka on Instrumental Rationality Questions Thread · 2015-08-23T02:48:25.906Z · score: 0 (0 votes) · LW · GW

I haven't heard of psuedoephedrine having nootropic effects before - what have you heard? (On a related note, it mostly makes me unable to sleep. :( )

Comment by dorikka on Soylent has been found to contain lead (12-25x) and cadmium (≥4x) in greater concentrations than California's 'safe harbor' levels · 2015-08-17T01:51:20.488Z · score: 0 (0 votes) · LW · GW

Lower threshold on safe doses of lead and other contaminants for frequent and infrequent consumption. Mostly just wanted to check if you knew of any such guidelines that you considered sane. :P

Comment by dorikka on Soylent has been found to contain lead (12-25x) and cadmium (≥4x) in greater concentrations than California's 'safe harbor' levels · 2015-08-16T04:59:20.061Z · score: 0 (0 votes) · LW · GW

Got it. Do you know of a sane set of guidelines that I can reference?

Comment by dorikka on Soylent has been found to contain lead (12-25x) and cadmium (≥4x) in greater concentrations than California's 'safe harbor' levels · 2015-08-16T04:10:26.370Z · score: 2 (2 votes) · LW · GW

Citations not really necessary, but would like to know why you have that opinion. I don't know much about contaminant quantities.

Comment by dorikka on Soylent has been found to contain lead (12-25x) and cadmium (≥4x) in greater concentrations than California's 'safe harbor' levels · 2015-08-16T01:37:55.252Z · score: 3 (3 votes) · LW · GW

How do the California guidelines compare to other recommendations?

Comment by dorikka on Peer-to-peer "knowledge exchanges" · 2015-08-10T13:30:13.920Z · score: 2 (2 votes) · LW · GW

What topics might you be able to teach others about?

Comment by dorikka on Peer-to-peer "knowledge exchanges" · 2015-08-10T13:28:59.460Z · score: 2 (2 votes) · LW · GW

What would you like to learn about?

Comment by dorikka on Peer-to-peer "knowledge exchanges" · 2015-08-10T13:27:52.434Z · score: 3 (3 votes) · LW · GW

List of participants:

Comment by dorikka on Stupid Questions August 2015 · 2015-08-10T03:35:00.931Z · score: 1 (1 votes) · LW · GW

only 25% are younger than 31.

I think you may have flipped this one, (so it's actually 75%). Median is something like 26 or 28.

Comment by dorikka on Stupid Questions August 2015 · 2015-08-10T03:31:53.679Z · score: 0 (0 votes) · LW · GW

As an off-the-cuff thought, much LW-content is useful for determining values and bringing strategies in alignment with those values. (I think of a hierarchy: values->strategies->campaigns->goals->actions.) It's also useful for some bug-catching and as some general tools of thought, but amassing lots of wealth is a convergent instrumental goal of many, many value setups. So value-discrimination may not really affect that, since either set of values would lead you to amass wealth.

The above isn't really something that I'm confident of, and belongs in the "only posted because the alternative is not posting" bucket.

Comment by dorikka on Tell me what hat to buy · 2015-08-08T16:56:12.038Z · score: 3 (3 votes) · LW · GW

Here is a bit of miscellaneous data on TBI accelerations. Here are some results and discussion re the Crasche hat. In short, I wonder whether the Crasche hat might put you at a bit more risk than otherwise, if you alieve that your head is protected but the hat will only reduce ~25% of the impact.

Comment by dorikka on Wear a Helmet While Driving a Car · 2015-08-07T03:26:41.320Z · score: 0 (0 votes) · LW · GW

I got some info regarding this, linking in case you're interested.

Comment by dorikka on Wear a Helmet While Driving a Car · 2015-08-07T03:26:14.583Z · score: 0 (0 votes) · LW · GW

I got some info, linking in case you're interested.

Comment by dorikka on Wear a Helmet While Driving a Car · 2015-08-07T03:24:46.785Z · score: 10 (10 votes) · LW · GW

I sent the Crasche folks an email asking for data documenting the performance of their product. I received the following results as part of a brochure:

LACROSSE CROSS-IMPACT TEST RESULTS (these appear to be stick-to-head):

Sample # Impact Location Cross Velocity (MPH) Severity Index Peak G

Bare Headform Side 38.19 1051 188

Size Small Side 39.94 68 49

Size Large Side 38.64 8 24

LACROSSE BALL IMPACT TEST RESULTS (these appear to be ball-to-head):

Sample # Impact Location Ball Velocity (MPH) Severity Index Peak G

Crosse BXX 1 68.34 278 179

Crosse BXX 2 69.3 208 146

Crosse BXX 3 70.75 294 186

The following sentence was also included in the reply: "Note - these tests are for impact coming onto the head. In a drop test, which simulates a head on collision, a 9 mph event showed an impact reduction of about 25 %."

For reference, I remember reading somewhere that football helmets tend to increase impact duration from about 3 ms to about 8 ms. Assuming uniform force distribution over the duration of the impact, this amounts to ~60% reduction in peak acceleration (over whatever force domain that acceleration is correct for.) Crasche seems to guard effectively against forces similar to that of a lacrosse stick being swung at the head. Unfortunately, the results of the second test don't really tell much about the usefulness of the hat, as the Crasche folks seemed to be satisfied with concluding that it's effective based on seeing a Severity Index (I assume this is based on the head injury criterion?) <300 when taking a lacrosse ball to the head, despite the huge accelerations.

Regarding the drop test - 9 mph gives around 137-82g of acceleration against a hard surface, assuming impact duration of 3-5ms. So we're looking at 6700-4000N (human head weighs around 5kg). In this domain, we can compare the 25% advertised reduction against the ~60% ballparked estimate for force reduction due to a football helmet.

In summary, the performance appears to be inferior (in terms of pure linear acceleration reduction) to a sports helmet, as one would expect. The hat appears to be more effective against object-to-head impacts (involving smaller forces) relative to impacts that result in the head being brought to a halt from motion (involving larger forces) - perhaps the crushable elements providing the resistance are crushed by the larger forces? My gut says that the performance in a vehicle collision will probably bring the head to a halt against a relatively immobile object, so the hat won't do much of anything as the crushable bits are crushed too fast to be effective.

Comment by dorikka on Wear a Helmet While Driving a Car · 2015-08-06T03:51:17.683Z · score: 0 (0 votes) · LW · GW

Hmph. I sent them an email asking for some quantification of performance, or studies if they have them. Will see what they say.

Comment by dorikka on Wear a Helmet While Driving a Car · 2015-08-06T03:41:19.222Z · score: 0 (0 votes) · LW · GW

Most of the studies that I've seen either seem to be looking at high-acceleration impacts or are trying to quantify impacts received in "daily life". I have repeated impacts flagged in my head as "result unknown" - I haven't come across anything that I can remember that would give legit thresholds for how hard a repeated impact has to be before it would cause damage (other than the 10g figure noted above). People seem to agree that repeated impacts have the capacity for great harm, and I remember seeing that people with a certain gene appeared to be more prone to symptoms if subjected to repeated impacts, but that's pretty much all I remember. Let me know if you find anything.

Comment by dorikka on Help Build a Landing Page for Existential Risk? · 2015-08-06T03:37:27.850Z · score: 5 (5 votes) · LW · GW

Do you know anyone who has done website design, like as an actual job? May want to ask them. I can really just say whether something does or doesn't look right to me - honestly wouldn't know where to start recommending fonts and stuff.

Comment by dorikka on Help Build a Landing Page for Existential Risk? · 2015-07-31T05:44:23.023Z · score: 5 (5 votes) · LW · GW

Thanks for your effort. As with many PR efforts, I would classify this one as either positive or negative; I would not intuitively expect a neutral result to occur (unless you had very few unique visitors), but that the website would shape visitor's perceptions of x-risk, either positively or negatively. Just something to keep in mind. On a more concrete level, I have trouble parsing the banner shown behind the "Existentially Risky" title in your screenshot. The combination of font and banner seems sketchy to me.

It doesn't appear to be optimized as you mention, but are you familiar with this page? http://www.existential-risk.org/

Comment by dorikka on Wear a Helmet While Driving a Car · 2015-07-31T05:38:47.570Z · score: 11 (11 votes) · LW · GW

I've looked up a few studies over the past few months in an attempt to understand direct causes of TBI and what kind of accelerations are "safe" - those under which a TBI (or diffuse axonal injury, DAI) is unlikely to occur, and to get a sense of what types of activities different accelerations correspond to. I'm dumping some info below without many citations as it's much lower effort than the alternative. If there's anything in particular that one is interested in, I may be able to find the relevant article. if you see anything that is wrong, please let me know.

  • TBIs appear to be caused by a combination of linear and rotational acceleration; these appear to cause injury via different mechanisms. Linear force/acceleration appears to create a high-pressure volume on the side of impact and a low-pressure volume on the opposite side of the head, with injuries possibly resulting from either. Rotational acceleration can, broadly speaking, cause brain tissues to stretch and twist in a way that they're not resilient to. However, this only appears to cause direct damage in the most severe of cases. In most cases, the majority of damage appears to be caused by a chemical cascade initially triggered by the damaged tissues. Bleeding of the brain is also possible, though I am not very clear on the causes of this.Regarding the chemical cascade, calcium ions spring to mind as one of the damaging species - curiously, repeatedly recovering from chronic alcohol abuse seems to be contraindicated also due cumulative damage from increased release of calcium ions, and they also may be rather indirectly linked to Alzheimers. More detail on the impacts and causes of calcium ion release in the brain is on my to-learn list.

  • In severe cases, it appears that hospitals sometimes administer treatment to control intracranial pressure or manage the chemical cascade to reduce damage. (I am not sure whether the focus is on inhibiting the cellular mechanism perpetuating the cascade or neutralizing the offending species themselves.)

  • Concussions seem to start happening around linear accelerations of ~50g (though one article places this at around 20-30g if I remember correctly, and expresses concern regarding repeated impacts as low as 10g) and/or a rotational acceleration of ~2000 rad/sec^2. Impact durations appear to range from ~3 milliseconds (against a hard surface) to 8 milliseconds (wearing a football helmet). As expected, lateral impacts (from sides, or front/back of head) produce greater rotational acceleration compared to impacts to the top of head. I would personally expect that injuries involving cessation of motion of the head would be hazardous at much lower peak velocities than those involving an object striking the head, as the 1) head is (usually) more massive than the object striking it, resulting in greater net force and 2) neck strength may reduce peak accelerations in the latter case, but perhaps not the former.

  • These links give examples of accelerations that appear to be produced during assorted daily activities. An experiment was done in which people of various ages shook their heads up/down in an "unconstrained manner", resulting in peak linear accelerations of ~4-5g and peak rotational accelerations of ~250-500 rad/sec^2. This appears to be somewhat difficult to reconcile with reports of injuries from headbanging, as these accelerations are very much below typically expressed hazardous levels.This article seems to refer to a similar motion, while reaching radically different results regarding accelerations (though, as far as I tell, it only does modeling, no measurement of actual head movement.)

Comment by dorikka on Wear a Helmet While Driving a Car · 2015-07-31T04:55:27.457Z · score: 0 (0 votes) · LW · GW

Good point. I would be interested in learning what types of impacts (in terms of acceleration, preferably rotational) crasche is designed for, as well as its performance compared to traditional helmets. Do you know of any commonly used metrics that might ve useful for comparing helmets in general? If I remember correctly, the ASME standards use a very high acceptance criteria for linear acceleration (>100g) and do not account for rotational acceleration (which seems to cause most of the damage.)

Comment by dorikka on Wear a Helmet While Driving a Car · 2015-07-31T00:48:24.852Z · score: 4 (4 votes) · LW · GW

Was looking at this in the past, a couple of thoughts. 1. Some articles seem to indicate that repetitive impacts as small as 10g may have cumulative effect. By increasing the effective diameter of the skull, helmets may increase thr likelihood of impact by reducing the necessary angular displacement for the head ro collide with back of seat. As materials of many helmets are tuned for more severe impacts, they may not offer meaningful protection in this domain. 2. Do helmets reduce peripheral vision?

Comment by dorikka on A Federal Judge on Biases in the Criminal Justice System. · 2015-07-09T00:59:15.685Z · score: 0 (0 votes) · LW · GW

Might be a case where posting a summary reduces the chance people will read the article sufficiently that it's best not to post a summary.

Comment by dorikka on Stupid Questions July 2015 · 2015-07-02T00:17:11.222Z · score: 0 (0 votes) · LW · GW

Curious whether, say, 100mg caffeine and 100 mg theanine would have a similar effect.

Comment by dorikka on Top 9+2 myths about AI risk · 2015-06-30T02:52:20.600Z · score: 4 (4 votes) · LW · GW

Nice. This would probably be useful to have near the end of a primer on AI risk, perhaps as a summary of sorts.

Comment by dorikka on Cryonics: peace of mind vs. immortality · 2015-06-24T23:12:52.510Z · score: 1 (1 votes) · LW · GW

Good point. As somewhat of a nitpick, I've heard that with some types of brain injury (especially those resulting from rotational forces and indirect impacts), a large portion of the total damage may be from secondary effects - my understanding is that this results from a chemical cascade that may only appreciably occur 4-30 hours after the event.

No citations and won't even bother for a hypothetical case - just a bit that I've read here and there. Google keywords are probably "secondary injury", "DAI", and "diffuse axonal injury".

Comment by dorikka on Open Thread, Jun. 8 - Jun. 14, 2015 · 2015-06-13T00:31:53.893Z · score: 0 (0 votes) · LW · GW

Value transparency, cross domain translation, and stability.

Comment by dorikka on Salary charts & Projection tool · 2015-06-09T03:27:33.684Z · score: 0 (0 votes) · LW · GW

Cool, glad to hear of the improvements.

Do you mean like when you are looking at Job A, and then move over to look at Job B? If so, would it be more useful if the graph just consistently showed, say, $20,000 a year as the minimum and, say, $200,000 a year as the maximum, regardless of occupation? (Or any other arbitrary min/max)

Yes, and I think that would work.

Comment by dorikka on LessWrong Help Desk - free paper downloads and more (2014) · 2015-06-06T21:29:59.357Z · score: 1 (1 votes) · LW · GW

If anyone could find the following, I'd appreciate it.

Funk JR, Watson RA, Cormier JM, Guzman H, Bonugli E. Kinematics and kinetics of vigorous head shaking. J Appl Biomech. 2015 Jun;31(3):170-5.

Comment by dorikka on Salary charts & Projection tool · 2015-06-05T19:46:26.193Z · score: 1 (1 votes) · LW · GW

The two seem to serve different purposes. I experienced some frustration with 80,000 Hours because I really just wanted to see numerical salary distributions but got lots of words and blog posts instead. I could have set up an advising session, but it felt 1) Cumbersome because I'd need to set up an appt, follow up, etc. and 2) Awkward because I wouldn't trust their judgments to a great extant; they hadn't provided sufficient evidence of their expertise for me to trust their opinions, but I would have trusted them to accurately report data that they had compiled.

Edited to remove word salad.