Or, a stupidly designed cryptographic cipher combined with an error-correcting code where again carefully corrupted packets, and requests for retransmissions, reveal information about the secret payload.
Monday, October 27, 2008
Turbo code interleaver attack
Monday, October 20, 2008
Chess armageddon
A match ends if one player leads by (say) two points, or if one player loses on time. There is guaranteed to eventually be a result.
This method seems superior to the current "white must win" Armageddon tiebreak method because it avoids the white-black asymmetry.
Another idea (by someone else) is to have geometrically decreasing amounts of time for successive games. This is elegant because the geometric series converges, so the match is guaranteed to take finite time.
Thursday, October 16, 2008
Image search
It's a nice hard AI problem, because one can not only do straight face recognition, but also use textual context to figure out names of people, locations and events, and follow the thread of those names, to find more images.
And the world will become a creepier place.
Tuesday, October 14, 2008
Porn market
How much would you need to be paid in order to post naked pictures of yourself on the internet? How much would you be willing to pay to see naked pictures of someone on the internet? There is market opportunity there for a trusted third party broker.
A model posts censored photographs of him or herself and establishes a threshold amount of money at which the uncensored photos will be released. The uncensored photos are deposited with the broker, which also collects money from "donors" interested in viewing the photos, holding the funds until the threshold is reached, at which point the model is paid, and the uncensored photos are publicly posted.
The business model of the broker is to invest the "float" between when the money is collected from the donors to when it is finally released, to the model.
Photographic challenges can be used to establish that the model is in fact the person submitting the photographs.
Donors may withdraw if the model raises the threshold. The model may lower the threshold to immediately "cash in" on the donated funds thus far.
Monday, October 13, 2008
Donut-shaped raindrops
Sunday, October 12, 2008
Saturday, October 11, 2008
2.7 Kelvin
Thursday, October 09, 2008
Turning off login drum sounds
System | Administration | Login Window | Accessibility
or /usr/sbin/gdmsetup
or edit /etc/gdm/gdm.conf-custom to include
SoundOnLogin=false
in the [greeter] section
Wednesday, October 08, 2008
Black hole moon
If the moon were replaced by a black hole of identical mass (plus the mass-energy of the one cosmic ray that triggered the transition :-) what would it look like from earth? I predict a series of concentric rings as sunlight loops a quantized number of times around the moon, then finally "reflecting" back to earth. What colors would the rings be? How bright and how big? How does the angular momentum of the moon (one revolution per month), now compressed into a black hole, affect things? Will tides on earth be affected?
What will the tranformation from moon to black hole be like? I assume it will be asymmetric, starting from the point of impact of the cosmic ray on the surface of the moon. What kind of gravitational radiation, and will we be able to see or feel it on earth? What kind of radiation, and how much, from the heat of the collapse? How long will it take?
What do "solar eclipses" look like?
Naked Singularity?
Model the worst case scenario of a large number of electrons, all spin aligned, colliding at a single point to form a black hole.
On further consideration, this probably won't work. Electrostatic repulsion is stronger than gravity. Is there a maximum charge a black hole can have? What happens if you keep dumping electrons into an existing black hole? (Practically, of course, it gets harder and harder as the black hole gets charged, and you face the problem of keeping the positively charged plasma left over away from the black hole. Maybe shoot the plasma away at high speeds to become a cosmic ray.)
Angular momentum per unit mass
Moon overhead
Insolation Force
I love Paris in the the spring time.
Graph programming language
Tuesday, October 07, 2008
Disk contention
Sunday, October 05, 2008
open-vm-tools on ubuntu hardy
This link: How to Install VMware Tools on Ubuntu Hardy 8.04 under VMware Fusion is a good start, but is out of date. Latest version is open-vm-tools-2008.09.03-114782.tar.gz . To compile it we need a newer version of uriparser than Ubuntu supplies. I downloaded uriparser-0.7.2 from sourceforge and installed it locally (NOT in /usr/bin!). Unfortunately, the configure script of open-vm-tools is screwed up such that making it see the locally installed uriparser using CUSTOM_URIPARSER_CPPFLAGS and CUSTOM_URIPARSER_LIBS did not work. When looking for the lib, it needed the -I which was not passed. Got some errors about uriFreeQueryListA. However, setting the environmental variables CFLAGS CPPFLAGS and LDFLAGS with the -I and -L paths did work.
Also, the command line you need for git is git clone git://git.opensource.vmware.com/opensource/open-vm-tools
Friday, October 03, 2008
md5 checksums of digits of pi
The digits following the decimal point, not including the "3.", with no end-of-line newline or linefeed.
| Digits | md5sum |
|---|---|
| 1 | c4ca4238a0b923820dcc509a6f75849b |
| 2 | aab3238922bcc25a6f606eb525ffdc56 |
| 5 | c92b32fbc94e2dff3e5516401d9bb463 |
| 10 | 0c70798ef9bd5bb878f53df3f85774de |
| 20 | 089b653970518aad3828d6e8630a1ae2 |
| 50 | d2ac6f2399ea4e94ec2cdfdb99c95ad9 |
| 100 | 954a383ec609b9126d4ba2ddf0bd1b37 |
| 200 | 59433bdec0a3a97f6d2785f855506ab7 |
| 500 | 5140171f33e9c149c97c8f3643f93f3e |
| 1000 | 5398eb229b2f9b4cdc42681c110707bb |
| 2000 | bb1fd8b5786fb4b3c91467e302631533 |
| 5000 | e1de0b54b87e3dcc846be7489edbdb8f |
| 10000 | 88d4eda38e94791ab2548d8463144a09 |
| 20000 | b660e34c926544e0ac174df38829ebd6 |
| 50000 | b2a057082a245a0a41817f614d95921e |
| 100000 | de18e0d8fef70b8f925cb2392c53f145 |
| 200000 | f22b756816f38eddf86ce210b3210841 |
| 500000 | 59de28f2ce1b2e4a95576e977da64c3e |
| 1000000 | e668904c195521a2a2dfef948ac54c8e |
| 2000000 | 12887cce18c4e1bffd9505c926230a90 |
| 5000000 | eb2774b75ab2f64f7d65b2d9436491d6 |
| 10000000 | 7f5f4ad06f084278f283a661ffafb379 |
| 20000000 | 13e07d09e200aa0837bf9b6c600b1818 |
| 50000000 | 8c50bb4458cbcc2c896ea50bd32da627 |
| 100000000 | dfc9f9b7080939885afb5f7bd2e04c4a |
| 200000000 | 496423ea6297b6455f7d076e2e4e2b56 |
| 500000000 | a3e9ad8d21dff2ccdc6bc96f369e3176 |
| 1000000000 | bff5d599c5aa342ab76086392cb73695 |
| 2000000000 | b2dae9b1569c3818b2ef78a0b6ad9502 |
| 5000000000 | 1d49002df7b62bc64241452d3bf75188 |
| 10000000000 | a3f0710c37a9e709edbd2de7a8925561 |
| 20000000000 | 240039d1a31f3caf6ec6cc3750b867c8 |
| 50000000000 | 67cb35130aac39f5cb8c79235b23a204 |
| 100000000000 | be5905dc2b9073a9f4c3635a1277b14a |
These checksums are in agreement with http://www.geocities.com/tsrmath/pi/picalcs.htm.
The last three entries, 20G, 50G, and 100G, were calculated from the value of pi from JA0HXV's site and not independently verified. The values less than that were verified using Ramanujan's method in Gourdon's PiFast program.
Some variations of a million digits of pi. Let DIGITS="14159...0577945815" (999,999 digits of pi after the decimal point).