pushcx

Karma: 71

pushcx 1 Dec 2014 22:11 UTC
4 points
in reply to: lmm’s comment on: December 2014 Media Thread
Many of the characters have thick accents and/or a patois. If you can comfortably carry a conversation in a crowded bar you’ll be fine without them.

pushcx 31 Oct 2014 0:51 UTC
6 points
on: In the grim darkness of the far future there is only war continued by other means

So if you want to keep people occupied for a looooong time without running out of game-world, focus on PvP

Or invest in “procedural content generation”, where the game world is constantly generated or regenerated. The “roguelike” genre has made games that have been played for decades (like Rogue, Nethack, ADOM) and continues to grow (Ultima Ratio Regum, Dwarf Fortress). It’s hybridizing into other genres like action platformers (Rogue Legacy, Spelunky, Risk of Rain). Games are creating new genres by starting with PCG (FTL, Minecraft). Civilization and the Maxis Sim games are classics in large part because of content generation.

For another perspective, game designer Dan Cook has written several http://www.lostgarden.com/2010/12/steambirds-survival-goodbye-handcrafted.html blog posts on PCG leading to better-designed game systems than handcrafted content. Similarly, Jonathan Blow has argued extensively against games that extend their use of systems (eg. across all the levels of a Super Mario, Modern Warfare, or Call of Duty game, the player will see few or no changes in rules, just new sets) rather than exploring a system once thoroughly (Braid, The Witness, Portal, Polarity).

I’ll leave the comparisons to “Scientific Progress [as] the PvE of real life” for the simulationists and solipsists. But I’ve always seen the human obsession with status and gossip as a bug rather than a feature and endeavored to advance more interesting things in the world.

pushcx 19 Aug 2014 15:26 UTC
3 points
in reply to: gwern’s comment on: Harry Potter and the Methods of Rationality discussion thread, July 2014, chapter 102
Sure: there’s no indication of delivery, so you don’t even know if one of the hops in your message opened all the envelopes, took all the money, read your private note, and trashed it.

pushcx 30 Jul 2014 19:16 UTC
5 points
in reply to: Toggle’s comment on: Ethics in a Feedback Loop: A Parable
Maybe the reason the post reminds you of myth is that it’s expressing a lack of agency. It’s a common feature there; generally the world is a place where awful things happen to you just because. The poster above is in a complex system where he feels he has no control, and the “whiff of aggrieved entitelment” response touches on that exact raw nerve.

pushcx 26 Jun 2014 15:45 UTC
3 points
in reply to: pragmatist’s comment on: Identification of Force Multipliers for Success
Yes. Speeds of 100wpm are not particularly hard to reach with deliberate practice. The benefit is not the time savings of typing less, it’s the cognitive savings of spending your attention on your topic rather than the mechanics of entering text and correcting errors.
What links here?
- Antisuji's comment on Identification of Force Multipliers for Success by Nick5a1 (27 Jun 2014 6:37 UTC; 0 points)

pushcx 26 Jun 2014 15:38 UTC
0 points
in reply to: NancyLebovitz’s comment on: Identification of Force Multipliers for Success
Psychologists now classify motivation as intrinsic vs. extrinsic—are you doing something because you want to, or because someone told you to/offered you something? Importantly, for creative tasks like knowledge work, extrinsic motivators like bonuses are weaker than people’s concern for a job well done. Many studies in a variety of situtations have shown the counterinuitive result that adding bonuses to a task makes people perform worse, give up quicker, and not do it on their own initiative.

The book Drive by Daniel Pink is an excellent walk through the research.

pushcx 6 May 2014 22:02 UTC
1 point
on: Find a study partner—May 2014 Thread
Looking for a partner for open-ended study of math/cs topics like calculus, linear algebra, stats, Haskell, SICP—open to suggestions for similar topics. Ideally, we’d meet weekly for 1-2h to discuss the previous week’s study and plan for the next. Bonus points if you’re in Chicago. :)

pushcx 28 May 2013 14:23 UTC
4 points
on: The Centre for Applied Rationality: a year later from a (somewhat) outside perspective
The workshops currently cost $3,900 + travel, I don’t think it was much lower a year ago. Have your improvements recouped that cost? Has the workshop increased your income?

pushcx 2 Apr 2013 11:38 UTC
17 points
on: Welcome to Less Wrong! (5th thread, March 2013)
Hi folks, I’m Peter. I read a lot of blogs and saw enough articles on Overcoming Bias a few years ago that I was aware of Yudkowsky and some of his writing. I think I wandered from there to his personal site because I liked the writing and from there to Less Wrong, but it’s long enough ago I don’t really remember. I’ve read Yudkowsky’s Sequences and found lots of good ideas or interesting new ways to explain things (though I bounced off QM as it assumed a level of knowledge in physics I don’t have). They’re annoyingly disorganized—I realize they were originally written as an interwoven hypertext, but for long material I prefer reading linear silos, then I can feel confident I’ve read everything without getting annoyed at seeing some things over and over. Being confused by their organization when nobody else seems to be also contributes to the feeling in my last paragraph below.

I signed up because I had a silly solution to a puzzle, but I’ve otherwise hesitated to get involved. I feel I’ve skipped across the surface of LessWrong; I subscribe to a feed that only has a couple posts per week and haven’t seen anything better. I’m aware there are pages with voting, but I’m wary of the time sink of getting pulled into a community or being a filter rather than keeping up with curated content.

I’m also wary of a community so tightly focused around one guy. I have only good things to say about Yudkowsky or his writing, but a site where anyone is far and away the most active and influential writer sets off alarm bells. Despite the warning in the death spiral sequence, this community heavily revolves around him. Maybe every other time hundreds of people rally around one revelatory guy it’s bad news and it’s fine here because there are lots of arguments against things like revelation here, but things like the sequence reruns are really off-putting. It fits a well-trod antipattern; even if I can’t see anything wrong in the middle of the story I know it ends badly. (Yes, I know, I’m not.)

pushcx 3 Dec 2012 4:41 UTC
5 points
in reply to: Ritalin’s comment on: Causal Universes
I was continuing on the post’s opening thought experiment of a computed universe; I was thinking whatever program is computing the new states of the universe would do this check. Sorry for the confusion.

pushcx 1 Dec 2012 11:28 UTC
23 points
on: Causal Universes

How would you compute that in one sweep-through, without any higher-order metatime?

A way has occurred to me.

Take the basic program described in the beginning of this post, in which the universe is deterministically computed with a cached series of states of the universe. The change is to make this computation is parallel on a staggering scale, because of how Time-Turners work. I’m going to explain this like there’s only one wizard with a Time-Turner that works for one hour, but I’m pretty sure it holds up for the complex case of many Turners that can go back a varying amount of time.

A wizard (Marty McFloat, let’s say) carrying a Time-Turner constantly generates a huge number new copies of the universe that differ slightly from the ‘real’ universe in a way that has gone unobserved because of the anthropic principle. In the ‘main track’ of the universe, nothing interesting happens, a new state is computed from the previous state. Every other copy of the universe is the same except that a rough copy of the wizard has appeared.

Maybe this copy of Marty has a new life vest, or has a bruise, or just has his head turned slightly to the left and one leg tensed. There are a finite but huge number of these variations, including copies where only a single quark (or whatever the smallest computed unit of manner is) is out of place. But in every variation, this copy of Marty’s brain has an extra hour of memories in its head. (More variations!)

Let’s follow one of these, the Marty with a new vest. This is a potential future Marty. You can think of this as the appearance of a Marty who went clothes shopping and used his Time-Turner an hour from now, but it’s not: it’s a variation of the Marty wearing the Time-Turner, not a copy of part of a future computed state. Every possible Marty-variant is generated in a different parallel universe state.

The state of the universe is computed onwards. If, one hour later, Marty does not activate his Time-Turner, the universe fails its consistency check and is deleted. If Marty does activate it, the universe looks back at the Marty that was added an hour ago. If the two Martys are not bit-for-bit (at whatever the lowest scale of the computation is), the universe fails its consistency check and is deleted. If Marty is identical, the ‘younger’ one that activated the Time-Turner is deleted and the universe rolls on in a consistent, causal way.

This system has no metatime. Universe computation never has to go back and restart from a previous state, modified or modified. It just requires generating a near-infinite number of parallel branches for every state of the universe and destroying nearly all of them. (Which I guess is quite a lot of world-killing, to understate it.)

The universe is causal and can be modeled with a directed acyclic graph. Just imagine that each state of the universe is a DAG, which may include a new node with no parents (the ‘arriving’ variant wizard), and it’s not one DAG but an incredibly thick stack of DAGs, most of which are discarded. The universe never needs to compute (p8pm|9pm).

If I correctly understood the prompt (does having multiple copies of the timeline count as “higher order metatime”, or does that just mean “no rolling the clock back” as in the example?), I think this perverse solution satisfies the constraints of the question I quoted, but I’d love to hear correction.

As a variation, nothing really requires that universes be computed in parallel; as long as the computer has lots of disk space it could write out the generated parallel universes to disk and continue simulating the current universe until it fails the consistency check or ends, then pick any remaining state from disk and pick up computation from wherever that state left off. This is a in-fiction way of restating that you can trade off space for parallelism in computation, but I’m not entirely certain what “higher-order” precludes so I wanted to toss it out there as a variation.

Actually, this whole post is an example of the general principle that you can trade off space for time in programs. It just ends up looking really weird when you push this optimization technique to a ridiculous scale. As for who would simulate a universe this way, I would guess some poor sap who was overruled in the design meetings and told to “just make it work”.

(On a meta note, I’ve been wondering for a few years if anything would prompt me to create a LessWrong account and participate. I read this post from my feed reader this morning, went on to have a day that’s been very busy and meaningful for my personal life. I didn’t think about this post at any point, went to bed, and woke up three hours later with this answer fully-formed except for the one-simulation-at-a-time variation I thought of while typing it up so I can get it out of my head and go back to bed. I guess waking me up with a ridiculous answer to a puzzle I didn’t know I was chewing on will prompt it.)
What links here?
- pushcx's comment on Welcome to Less Wrong! (5th thread, March 2013) by orthonormal (2 Apr 2013 11:38 UTC; 17 points)