= 27d567bc2d48d9f40e9ccc20ea370b15)
You can't effectively compete against people who are not accurately evaluating the consequences of their decisions.
So you're saying capitalism could not possibly work, right? X-/
For more clarification, I was thinking this over when considering rental properties in my area. A lot of people have complained that it is near impossible to make a profit on a rental property where I live. I think a lot of that is because there is a huge chunk of people who have bought property as an investment based on potential appreciation instead of based on cash flow. If your model is using only cash flow, but another model has a 5% appreciation of principle built in to it, it is going to be near impossible to be able to compete with them on rates. However, what you also see is a ton of people looking to sell their rental properties after the appreciation they were expecting never occurs (potentially due to the glut of properties on the market from other people doing the same thing). The people exiting the market take a loss, and the people who could have actually made a profit based on a cash-flow model never can get a renter to begin with, due to overly competitive rates. However, since new people are willing to invest under the same assumptions as the old investors (imperfect information), the market keeps going strong, and renters can take advantage of cheap rates, though no one is really profiting from it. That's the type of scenario you can get with an overly competitive marketplace filled with imperfect actors.
Edit: I'm sure people are making money from rentals in my area, or there wouldn't be so much of it. I'm just also sure that a lot of people are losing a ton of money from it, and driving down prices for everyone else.
You can't effectively compete against people who are not accurately evaluating the consequences of their decisions.
So you're saying capitalism could not possibly work, right? X-/
No, I'm saying that capitalism is never purely implemented (with no barriers to entry/perfect information/etc.), and there are cases where due to these inefficiencies, increased competition can cause a poor business model to outcompete a sound business model, leaving nobody standing at the end. This doesn't always happen (hence capitalism mostly works).
In response to
LINK: Competition is toxic
This ties in with a thought chain I had this morning. While we may have an incredibly competitive environment, it is also populated by imperfect actors. You can't effectively compete against people who are not accurately evaluating the consequences of their decisions. This can be seen in sports, where illegal performance enhancers are the norm in many sports, and non-dopers can't really keep up (despite the achievements of the dopers being annulled later). This can be seen in business where someone who is willing to sell at a loss and make up for it in volume, will steal all of the customers from a legitimately profitable operation (though will soon go bankrupt as well). I'm sure there are examples in all sorts of industry where imperfect actors make decisions based on poor analysis that can potentially ruin better plans due to an overly competitive marketplace.
Give the artificial life forms an explicit preference that the external world (if one exists) remain causally pristine.
The AI is given a specific, relatively short time window in which to complete its task.
These two things seem to contradict each other. How should AI both complete a task for you and not influence you causally?
Exactly. Any observations you make on the AI, essentially give it a communications channel to the outside world. The original AI Box experiment chooses a simple text interface as the lowest bandwidth method of making those observations, as it is the least likely to be exploitable by the AI.
In response to
Stupid Questions April 2015
If I cook a fixed amount of raw rice (or couscous, or other things in that genre) in a variable amount of water, what difference does the amount of water make to calories, nutrition, satiety, whatever?
For example, if I want to eat fewer calories, could I cook less rice in more water to get something just as filling but less calorific?
In response to
Open thread, Apr. 01 - Apr. 05, 2015
There's a fairly specific subset of coding theory that I feel should exist, but I don't know what it's called or how to find it. It's best characterised by needing to obliquely embed subchannels of communication in human-readable text.
Here are some examples of problems that exist in this area:
1a) How do I pass arbitrary concealed data in a body of English text? Let's say I have a message of length n bits. What would be the most efficient way of obliquely encoding that message so that it passes for plain English text? For example, if I had the message 11001001, I could use a Markov text generator, and for each word, check and force the parity of a checksum of that word, picking eight words whose parities corresponded to [1, 1, 0, 0, 1, 0, 0, 1]. This wouldn't be very efficient, and it probably wouldn't be astonishingly comprehensible English, but it starts to address the problem.
1b) Say I have an existing body of well-formed English text, and want to somehow transform it so that it encodes the message 11001001. What variety of transformations would still yield well-formed English text?
2) Given a body of English text, how can I obliquely clue in a knowing observer (or a regex pattern) to the fact that this text is somehow distinguished and worthy of scrutiny for some predetermined purpose? Motivating case: I've just encoded the very important message 11001001 as a body of English text, using a method for solving problem #1 above. I want to do something to it so that my confederate will recognise it as containing this message. One way of doing this would be to wrap it in a known initiating and terminating phrase, a bit like ASCII-armour for a PGP key block. For example, I could begin with "Once upon a time..." and end in "...and they all lived happily ever after".
3) A group of conspirators want to adopt aliases on an anonymous forum. They want to be able to reliably recognise each other by their online aliases so they can collaborate, but don't want their affiliation to be apparent to other observers. They could, ahead of time, come up with one or more highly selective rules to which their aliases must conform. The identities of all conspirators would then be known to each other.
I think it is not very hard writing something which will encode a hidden phrase using odd and even counts. (But length is key).
In response to
comment
by
on
Harry Potter and the Methods of Rationality discussion thread, March 2015, chapter 119
In response to
comment
by
on
Harry Potter and the Methods of Rationality discussion thread, March 2015, chapter 119
Using the time it takes Earth to rotate one degree gives you 86400 seconds in a day /360 degrees = 240 seconds. But the length of the day has been getting larger as the Earth slows at a rate of about 1.7 ms/century wiki
To find when one degree was equal to 234 seconds, we can find when a day was approximately 234*360 degrees = 84240 seconds, or approximately 127 million years ago. Putting the creation of the stone right in the middle of the Cretaceous Period.
Coincidentally, this also solves the issue of how the T Rex got away with such tiny arms. They had wands!
In response to
comment
by
on
Harry Potter and the Methods of Rationality discussion thread, February 2015, chapter 113
They all showed up when the Dark Mark was called, only one of them has a transfigured mask replica, and no Death Eaters are likely to be allies to Harry since Voldemort can apparently just will them into seven smoldering pieces at any time.
In response to
comment
by
on
Harry Potter and the Methods of Rationality discussion thread, February 2015, chapter 113
Yes, but who called the Dark Mark, and pointed out the transfigured mask. It could all be a ruse by LV. Constant Vigilance!
In response to
comment
by
on
Harry Potter and the Methods of Rationality discussion thread, February 2015, chapter 113
Surely following Voldemort's exact instructions and giving up his secrets would equally count as losing, without risking annoying Voldemort and getting killed or punished if your hypothesis is wrong?
In response to
comment
by
on
Harry Potter and the Methods of Rationality discussion thread, February 2015, chapter 113
Of course, that would count as losing as well. I just think he needs to explicitly acknowledge that he is losing, so that Voldemort doesn't think he is secretly plotting something else.
I'm just worried that this is all a big setup, and the 37 "Death Eaters" are really Harry's allies in disguise and Imperiused, so any attempt to get out will cause Harry to end up killing all of his friends and put him on the true path towards destroying the stars. There was enough potential foreshadowing for this to be true.
-They aren't wearing the correct battle armor, only a hastily transfigured replica.
-LV explicitly said he expected Harry's friends to show up later than they did (which could mean they were supposed to be there for this ritual).
-The two main ones I've heard people talk about seem to be Lucius Malfoy and Sirius Black, both of whom are arguably now Harry's allies.
= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
= f037147d6e6c911a85753b9abdedda8d){kind=logo}
Ankle monitors are surprisingly annoying to wear. It would be a big, long-term punishment that would probably come on top of a prison sentence.
I'm surprised no one has pushed through a cell-phone tracking app as a replacement for the ankle monitors. Sure, its not as secure, and may be left somewhere/forgotten/etc. but if you included it as a condition for parole/probation, you could probably get pretty high usage rates, with little added cost and annoyance.