« Dark Side Epistemology | Main | Protected From Myself »

October 18, 2008

Comments

I request a short post on turingtrade.org. It's a contest for chatbots to see which ones act the most human, as well as a contest for humans to see who's best at distinguishing humans from bots. Both contests are scored with a prediction market.

I'd like to see your thoughts on prediction markets within the larger context of information markets.

Also, your thoughts on markets supporting and rewarding effective prediction with increasingly deep, complex, or "in principle" applicability, in contrast with today's nearly complete focus on specific, spot predictions. In other words, a shift from the market rewarding effective predictions, to rewarding increasingly effective methods of prediction.

Of course it's easy to show that this shift occurs naturally already in the background, as competition inexorably leads to ratcheting forward of increasing efficiencies and synergies, but to make this explicit...?

It'll be sad to see you go, Robin, I'll miss your contributions here. Perhaps you could come back every so often with an occasional post? Anyway, I think a last post of yours on theoretical prediction markets in various arenas would be a great read. Maybe a hypothetical suggestion on how the US government could use a prediction market to plan for future military actions?

One last post defending/distinguishing Naturalism vs. Supernaturalism would be epic.

I'd like to second Aaron's hope that you'll drop in with the occasional post [although I have most of the archives still to go through :-)].

"I'm strongly tempted to quit blogging for a while, to free up time for more ambitious projects"

NOOOOOOOOOOOOOOOOOOOOO!

But if you're really going, then perhaps something about what communities choose to subsidize and why. ( I notice in your Econ 350 class syllabus, you look at subsidizing sidewalks v. wireless internet. In my town, property taxes have gone up dramatically as more and more kids have left our town's schools for private schools or schools in nearby towns.)

I'd also love to read any new thoughts you might have on signalling and on consumer behaviour in general, and/or on the value of health care and medicine in the U.S.

Awww.... My friends and I recently discussed our favorite feeds and when asked to restrict myself to 3, you made the cut. I look forward to your ambitious projects and hope to see a guest post now & then as opportunity strikes.

I hope that your ambitious projects include a book on health care. Count me among those who will miss your posts.

Robin,

I enjoy reading your posts. After reading for over a year, I am beginning to think of the following:

Cognitive bias while limiting cognition, by the same token, completes your perception. Without approximation, you could never complete a perception.

So, thoughts can be thought as low-resolution thoughts, high-resolution thoughts and so forth and each serve their own function.

So, we could classify biases that are expected to be present in a low-resolution thought and others can exist in a medium-resolution thought and so forth.

I think, just like we have a hearing range, seeing range, we do have a perception range and it is bounded by cognitive bias.

What do you think?

http://blog.amusecorp.com
Vasu Srinivasan

You will be missed.

Perhaps something on the topic of how to make truth-based decisions in a social-signalling-based environment? (Especially on the topic of health care?)

Dear Robin,

Your blogging has been admirable; and in a way, your decision to give it up is equally admirable and honest.

What I would be most interested to see is post detailing your thought process when presented with an interesting argument or piece of evidence.

All the best!

Hi,

I greatly enjoyed your posts. Wish you success in future endeavours.

The topic i would like discussed is

Would restricting franchise to taxpayers be a better political decision making system? What abuses may appear in such a system?

Case1 - Those who pay taxes get one vote. National service of a minimum period could be considered equivalent to a tax.
Case2 - Vote is proportional to tax paid.

This assumes a scenario of only direct, attributable taxes. No corporate taxes or VAT.

I forbid it. You may not stop posting.

Also, my favorite topic of yours is health care, so anything you think you've missed in that area I'd be interested in hearing about.

Why choose to quit rather than simply reduce your posting frequency?

I really appreciate you and Eliezer engaging each other. That is what I would miss the most.

I'd also be interested to hear about any personal/health/career advice you have.

You've been a must-read from the very beginning when you posted occasional guest posts over at MR, to your two years of blogging here. Best of luck on your new project(s)!

How about getting together w Cowen for that bloggingheads?? That was much anticipated.

Post about your new project(s).

Aaargh! You're one of my favorite bloggers.

Since you love meta, I suggest you do a post on what topic/s you OUGHT to be focusing on.

Let's not call this quitting Robin. There's no need for such promises. How about an indefinite break from posting? It is interesting that you mention a sense of completion. I figured blogging was generally more of a journey than a destination. I also suspect it is a bit of a time-waster. Your time may indeed be better spent on something with a more clear destination.. good luck.

I'd hate for you to go, I've enjoyed many of your posts and have had my mind expanded by your insights.

My last request would be for you to finish that story of individuals searching for the final complete theory of the universe. You started the story with students in a classroom in the distant future being given an assignment to find the theory of everything by their mischievous teacher. The story left off with them musing over different paths and ideas to work on...

It was a delicious beginning to a good story, would love for you to finish it.

I other words you're addicted and you think the only solution is to stop cold turkey? Because blogging has no prerequisite minimum time investment. And this blog already has over a dozen co-bloggers.

Compliments on an excellent run Robin. My question is: once we develop an awareness of our biases and their various manifestations, how do we best modify or tweak our analytic-engines or thought process?

What biases warp our judgement causing us to keep blogging too long or to give up too soon?

I'd really like a post on when (if ever) it would be rational to disregard market (whether prediction or traditional) prices for a fear of a speculative bubble (or "Keynesian beauty contest") when one has the option to participate in said market. This is something I haven't seen brought up by many (or any?) commenters on the current financial situation.

Wow... unsettling that some readers don't seem to have kept Robin and myself distinct in their minds... that composite must be one strange and scary fellow.

Will you be replaced? Unbroken Eliezer is not something my GReader can endure.

Clearly the two of you have merged to become Nicolas Hanskowsky.

I'd like to hear justifications for some of your contrarian principles from Politics isn't about Policy. I find it frustrating when people make controversial assertions without any sort of explanation.

@Eliezer: You might be able to change the typepad settings so the author's name would appear right under a post's title, instead of at the bottom. Maybe that would make you and Robin harder to mix up.

A post saying that this one was a joke, that you will continue blogging.

I second joaco's request.

This one is difficult. One shouldn't go on too long. But one shouldn't stop too short. And a two year anniversary is just an accident of time - surely a bias, if this should decide the event.

I think, Robin, you underestimate your value. For you, you are just thinking the same old thoughts. But for us they are often deeply and originally strange and interesting. Which I think is quite an obligation that should weigh heavy in your Bayesian decision.

I would also think, that this forum reaches parts no other endeavour would ever reach.

I would regard blogging as a temporary phenomenom - and this is still the Golden Age where newness roams and the growing audience is listening.

Two years isn't enough, Robin, But thanks for all the fish.

At some point the talking has to stop and action begin.

I'd be interested in knowing what you were going to do next as well.

So you abandon the Hansonians to their fate at the hands of the Yudkowskyites, do you?

Well, if you must. But I'll say I'm not happy about it. I vote for one last post on why Eliezer's view of the Singularity is nonsense and yours is the best. I found those highly entertaining.

First, Robin, have you ever thought of going on a twice-a-week blogging schedule? That's what I've done, and it's done wonders for my ability to spend time on other things. I create the illusion of continuity by putting timers on when my posts will go up.

But if you do choose to stop, I'd like to see you expand on your "politics isn't about policy" stuff, perhaps with recommendations on how to learn more about it.

@Eliezer: I do check the names on OB posts, I tend to think of you and Robin as the "Overcoming Bias guys"--at risk of offending other team members, you're the best guys on the site, losing either of you would be a big loss.

@John Maxwell: That post of Robin's was a summary/outline. Read his other posts for the detail you're seeking, or offer a specific examples of claims he has not explained well enough.

A clarification of your view on advanced nanotechnology, if you really don't mean to dismiss it.

Ways of improving your life through overcoming bias.

An evaluation of Eliezer's views of the Singularity and friendly AI.

Biases of economists.

I'd be interested in hearing more about why you think blogging limits your ability to pursue your ambitious projects. Is it just a matter of time?

I'll be sorry to see you go, though I look forward to your longer projects. Perhaps you could post occasionally, when inspired, to keep the OB community?

As to last post topics:

(1) I'd like your summary of: (a) what aspects of bias and rationality OB most hasn't hit, and (b) what lines of evidence or avenues of attack might help here.

(2) I'm interested in the thoughts in your Matrix essay and in your exchanges with Eliezer on hypocrisy, and I'm puzzled by the underlying issues. I'd love to hear anything more you have to say along those lines.

(3) OB has mostly discussed rationality as individually accurate map-making. I'd love to hear any mechanisms (e.g., Idea Futures, or guarding against group polarization) for good map making as a community of realistic rationalists, or any starting-points for thinking up such mechanisms.

(4) You've discussed how beliefs (and actions) are like clothes. I'd love your summary of the main ways clothes behave. I.e., if beliefs aren't about having accurate maps, and actions aren't about moving instrumentally toward goals, what are they about? What are the main dynamics by which they propagate?

Robin, sorry to see you go..

My request: how about an update on your old piece: Fourteen Wild Ideas, Five Of Which Are True!


I'd be interested in
- how would you rank these ideas now (in terms of likelihood of being true)
- what do you think, now, is the probability that 5 or more are true?
- if you wrote that piece again, would you use the same 14, or are there any you'd drop and, if so, what would you replace them with?

I second botogol, although I'd like to see numerical probabilities assigned to each item individually.

Anna Salamon: (3) OB has mostly discussed rationality as individually accurate map-making. I'd love to hear any mechanisms (e.g., Idea Futures, or guarding against group polarization) for good map making as a community of realistic rationalists, or any starting-points for thinking up such mechanisms. (4) You've discussed how beliefs (and actions) are like clothes. [. . .]
Bayesians, and perhaps other types of rationalists, aspire to accurate and precise map-making assuming that two or more accurate and precise maps are qualitatively identical, i.e., identical in every way except for being one and the same. (Of course, some Bayesians might aspire to have accurate and precise maps that properly contain other maps, to be domain masters.) This, I'm fairly sure, is more "realistic" than a community of rationalists wearing shirts expressing the falsehood, if their psychological world lines are distinguishable, "My map is numerically identical [one and the same] with every other rationalist."

I would hate to see you stop blogging.

I would like to see the point/counterpoint with Tyler C.

@Sociology Graduate Student:

Robin doesn't appear to believe that he's supported his claims:

I won't support all these claims here; for today, let's just talk politics.

I'd be interested in seeing the following claims supported:

  • Bedrooms aren't about Sleep
  • Marriage isn't about Romance
  • Charity isn't about Helping
  • Medicine isn't about Health
  • Consulting isn't about Advice
  • School isn't about Learning
  • Research isn't about Progress

If Robin supports his "School isn't about Learning" claim, I would also be interested in hearing how that affects his role as a college professor.

@Sociology Graduate Student:

This is pretty much unrelated, but this is an interesting blog post on the applications of sociology.

What do you think of drawing a parallel between the various philosophies of probability: the frequentist, subjectivist, logical, or propensity theory, with the debate about on whether economics is a science. In other words, are there parts of economics that are more frequentist, other more subectivist, etc...?

Dear Robin Hanson:

It has been bothering me more and more lately that there isn't a real-money, large-scale idea futures/prediction market.

Do you have any ideas or plans for how to make such a thing real?

Regards,

Zooko

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31