Scooter The Wonder Dog of DOOM
[Most Recent Entries]
[Calendar View]
[Friends]
Below are the 20 most recent journal entries recorded in
kyrie1618's LiveJournal:
[ << Previous 20 ]
| Tuesday, November 24th, 2009 | | 4:54 pm |
| | 4:37 pm |
| | Sunday, November 22nd, 2009 | | 6:53 pm |
But a computer of finite complexity but infinite size, on the other hand...
A computer of finite size /cannot/ produce any computable number, even if it forgets the digits output so far. It needs to remember where in the number it currently is. To produce any computable number you need more. You need an infinite amount of memory. Can imagine how to even address/index that much memory? Oh. Um. Actually, *can*...
Out here in the fields
I fight for my meals
I get my back into my living
I don't need to fight
To prove I'm right
I don't need to be forgiven
-- Baba O'Riley
| | Friday, November 20th, 2009 | | 11:37 am |
This is the best thing I've read in a while.
"Unless the illusion of continuity, the feeling that you are a single unbroken organism moving through time, is particularly strong in your mind, you must fully grasp your situation, and come to terms with the fact that you yourself cannot do much, but must rely on persuading your tribe of future selves to accomplish tasks for you." -- http://justbeast.livejournal.com/176790.html... I wish I had an internet connection. | | Wednesday, November 18th, 2009 | | 3:37 pm |
| | 3:05 pm |
One day, others may try to convince you they have forgiven you. That is more about them than you.
Am I done searching out the proper foundation for the Inferential Engine? Nothing imperfect would do - while any complete language can express any thought, they differ wildly in efficiency. Alice and Bob say one word, but that word takes twenty (or two hundred) to properly translate. My kingdom for an Oracle. Kolmogorov proved Asymptotic equivalence, and Turing proved its worthlessness. Back to the beginning: to know it for the first time. "I do not mean by this declaration to condemn those who believe otherwise; they have the same right to their belief as I have to mine. But it is necessary to the happiness of man, that he be mentally faithful to himself. Infidelity does not consist in believing, or in disbelieving; it consists in professing to believe what he does not believe." Why did we not make money in Amway? Some people did. Were they smarter? More deserving than us? Did they work harder? Did we lack commitment and enthusiasm? Were we afraid of success? It is for chickens to laugh at. Why did I ever fall for the line? "20% of the people do 80% of the work." If I was stupid then, I am stupid now. But I am catching up. When you do a perfect job and yet still fail, it is not you who failed. "you knew it was wrong. A voice you did not recognize screamed for you to stop. You saw no way out. It was the way things were. They could not be changed. You tried to convince yourself the people you were hurting deserved it. You became numb to their pain and suffering. You learned to shut out the voice speaking against it." Turing Machines are an improper foundation, impractical for application. Conway Life boards are, too. Nand networks just might be unbeatable. I had to abandon this line of research when I secured a promise of renewed funding. Back in the prediction-for-profit game! Some day I will return to Artificial Intelligence. "When a man has so far corrupted and prostituted the chastity of his mind, as to subscribe his professional belief to things he does not believe, he has prepared himself for the commission of every other crime. He takes up the trade of a priest for the sake of gain, and, in order to qualify himself for that trade, he begins with a perjury." I will return. Make the computer learn. Gradient descent backpropagation on a digital logic circuit. What was true about Amway is true in some other places. | | Tuesday, November 17th, 2009 | | 7:21 pm |
My sequence is longer than OEIS's.
I have the first 56 numbers. OEIS has the first 46. But their graphs have the first 10,000. Whatever algorithm did that, I want it. http://www.research.att.com/~njas/sequences/A0000411, 1, 2, 3, 5, 7, 11, 15, 22, 30, 42, 56, 77, 101, 135, 176, 231, 297, 385, 490, 627, 792 David W. Wilson did it right: http://www.research.att.com/~njas/sequences/b000041.txtWhy did I need this? It involved cumulant research. What was I doing? I had a practical use for it in mind. | | Thursday, November 12th, 2009 | | 2:10 pm |
| | 9:45 am |
| | 9:38 am |
| | Wednesday, November 11th, 2009 | | 10:07 pm |
| | Monday, November 9th, 2009 | | 7:43 pm |
The design is upper transcend, but the design is for the unthinking depths.
Rule 1: Avoid creepy people. Sure, they're fun and unpredictable and un-boring and interesting and frequently even amazing. Even the totally flipping insane ones carry deep insights into the human condition. 95% of the time, they're great. The trouble is, 5% of the time they are Charles Manson. That's a wee problem, now isn't it? (I kid, of course. Anything that happens less than five percent of the time must be harmless.) Median != Mean. You will never know enough. All data is sparse data. That's all that this year has taught me. What are some other good rules for interpersonal interaction? --- There's this filter. Take clay, mix with a certain proportion of coffee grounds, bake it into stone, and burn the coffee grounds away. We know, with our microscopes and medical instruments, why this filter works. It stops eggs. We might never have invented this filter without modern understanding. Yet we could have been making these filters for the last ten thousand years. We could have ended so much suffering, if only we had known how. To design something, you have to know why it works. The hands carrying out that design do not need to understand. But so many of our designs are made for use in civilization. Refrigerators only work if you have electricity, or a petroleum industry, or lots of stored bottles. But a swamp cooler works with simple wind. If the computer in my car gets a virus, that takes special skills to correct. A car that runs on solar power or ethanol can keep running even if civilization falls because the sun will always be with us and even the sumerians knew about ethanol. I want to design things that no-one understands how they work, but they are designed for and to work in our primitive 21st century. John Conway's Life is Turing-Complete. It's an information processing model capable of computing anything that can be computed. Yet we do not have computer scientists designing programs and operating systems that run on hardware implementations of Conway's Life. The very idea is silly. Discovering those algorithms would simply be too hard. --- John Conway, I think, showed me why people keep conflating quantum physics with consciousness. It's that nonsense about a "conscious choice". Point the first: all physics except for quantum is explicitly determinist. It's frequently unpredictable, but could never have been anything other than determined. 2) even quantum physics is perfectly compatible with determinism. Any hidden variable or many-world interpretation demonstrates this. These interpretations may be currently unpopular (maybe even for good reason) but they are still valid. What sets quantum physics apart from classical and relativistic physicses is that quantum allows for the possibility of nondeterminism without requiring nondeterminism. If we're to toss determinism, it has to be for another reason. 3) if we are free, and that freedom means that we are able to meaningfully choose the orientation of a polaroid filter, then Conway has convinced me that the particles going through that filter are able to meaningfully choose which way to fall. Maybe I completely misunderstood his paper. Help? 4) if we are free, where does this freedom come from? Old physics? New physics? Non physics? Since magic doesn't exist, I'll ignore the last. New physics may well be responsible, but it would be irresponsible of me to jump there without first ruling out old physics. Is it really so implausible to think maybe particles are, in some sense, free? They appear underdetermined. That's not far off, doc. 5) If I have a choice whether or not to turn right at an intersection, the many-worlds interpretation suggests that there are two real, realized futures: one in which I turn right, one in which I don't turn right. If both futures exist, then I can know that at any intersection I will in fact go every way and experience every future. If I want to turn right, one of me will turn left instead. In fact, there is one of me that has made every choice of his life in the way that he least wanted. There's another one of me that's won the lottery and retired despite the fact that I Have Never Played. This may be an interesting way to look at the universe, but it's the opposite of having a meaningful choice. Greg Egan taught me that. Larry Niven tried. 6) "Magic doesn't exist" is not an assumption but a tentative working conclusion based on the work of James Randi and Adam Savage and others like them. 7) Falsificationism (A theory can never be proven, only disproven.) is darn close but no longer convincing to me. Why? Because there are very simple theories out there. Think of a proposition that can be proven or disproven: the world is flat. "The world is flat" is a wonderful theory, making all kinds of predictions. But so is "the world is not flat." A and not-A both make predictions. A and not-A are both falsifiable. A and not-A can both be confirmed. At least, near as I can figure... I phrased them in such a way that any challenge to one helps the other. There is no third option in this case, no middle ground. If theories can be constructed in this way, then theories can in fact be proven sometimes. 8) We're made of particles and force fields. Force fields are predictable -> determined -> unfree. Particles are unpredictable, but do we really want to hang our freedom on how a particle zigs or zags? After all, complex classical systems are unpredictable, too. The distribution of prime numbers is subtle and confusing. Perhaps there is something like freedom that determined systems can have. What should we call it? 9) I have made no reference to consciousness. I've phrased this in terms of behaviors. Why? Because I doubt that freedom and consciousness have anything to do with each other. They feel linked, but then, it feels like the stars rise and set. Given how predictable conscious human behavior is before that human is conscious of a choice having been made, it seems likely that conscious choice awareness is probably an afterthought and has no power. It's probably a result of the choice, not the cause. 10) Consciousness is the one thing in the universe that we can be sure is real. Next to that freedom and choice are, the most die-hard believer will admit, possibly an illusion. We make unconscious choices all the time, a steady stream of them, that much is obvious. They are different sorts of things, consciousness and freedom. They are correlated, but which way does the causation flow? Same way is always does: from past to future. If it can be demonstrated that the choice is made before consciousness begins, then the jig is up. Maybe made choice causes awareness of the choice, but the one thing you can be sure /isn't/ true is that the choice is made by conscious processes because that would require time travel. Anyone want to take bets? --- I have ten google wave invites! Who wants? | | Wednesday, November 4th, 2009 | | 8:23 pm |
three words: massive parallel instantiation
http://en.wikipedia.org/wiki/Open_Graphics_Projecthttp://wiki.opengraphics.org/tiki-index.php
Later AGP and PCI Express variations will follow. The initial prototypes of OGD1 are available as of January 2007[1].
Full specifications will be published and open source device drivers will be released. All RTL will be released. Source code to the device drivers and BIOS will be released under the MIT and BSD licenses. The RTL (in Verilog) used for the FPGA and the RTL used for the ASIC are planned to be released under the GNU General Public License (GPL).
It will have 256 MiB of DDR RAM, be passively cooled, and follow the DDC, EDID, DPMS and VBE VESA standards. TV-out is also planned.
This is a GPU -and- an FPGA? GHDL and Icarus Verilog I get. But how does a debian/gentoo nerd burn cores to chips? | | 4:38 pm |
| | 4:04 pm |
Holy /BALLZ/!
Can I sell advice on Ebay? I want to sell advice on ebay. | | Tuesday, October 20th, 2009 | | 12:52 pm |
if i wasn't conscious
i wouldn't be angry. i wouldn't /be/ at all. i know why i be. but /how/ do i be? that's so weird | | Sunday, October 18th, 2009 | | 6:34 am |
Oh, and:
Turing machines were good. Real, real good for the 20th century. We need a new enumerable, complete computational model. I'm thinking NAND nets. Similar to neural nets, I suppose, but ever so much easier to instantiate. Concurrent, efficient, fast. So very fast. The machines will think rings around us, because they must. Got to leave the wireless hotspot now. Off to sleep, where internet doesn't go. Balls! | | 6:27 am |
Just read it.
http://www.rifters.com/crawl/?p=791 -- PRISMs, Gom Jabbars, and Consciousness
It’s not just about puzzle-solving, or reconciling conflicting inputs, either: look at all the illusions and mind tricks that depend on the brain performing exactly those kinds of operations unconsciously. Binocular rivalry, inattentional blindness, ventriloquism: incompatible views shown to each eye, buildings appearing and disappearing from our field of view, the mouth moving here but the sound coming from over there: somehow the brain puts all those conflicting inputs together and serves up the final product without any sense of conscious conflict. The speaker mouths “ga”; the sound hitting the observer’s ear is “ba”; but the observer hears “ga”, an intermediate sound, without ever being aware of a conflict in need of sophisticated and complex resolution.
...
The more sublime your awareness, the more pain you can withstand. And is anyone here not thinking of the Bene Gesserit and their gom jabbar, from Dune? Herbert, once again, had it right. His simple device reduced Voight-Kampf to its essence: testing for humanity through torture.
If Morsella is right, consciousness scales with conflict: the greater the discord between systems, the higher the level of awareness. You are never more alive, more awake, more conscious, than when in excruciating conflict with yourself.
So now I can decide how conscious I want to be. No drugs needed. I want Dumbspeech, I want it /bad/. I want to know how thinking works. Wow. | | Thursday, October 8th, 2009 | | 7:02 pm |
| | 4:54 am |
|
[ << Previous 20 ]
|
![[info]](http://l-stat.livejournal.com/img/userinfo.gif){kind=small}
kyrie1618's journal
