Microsoft Word Kurzweil, Ray The Singularity Is Near doc

Nano Letters
4.5 (2004): 915–19. 
20.
Chao Li et al., "Multilevel Memory Based on Molecular Devices," 
Applied Physics Letters
84.11 (March 
15,2004): 1949–51. Also see http://www.technologyreview.com/articles/rnb_051304.asp?p=1. See also 
http://nanolab.usc.edu/PDF%5CAPL84-1949.pdf. 
21.
Gary Stix, "Nano Patterning," 
Scientific American
(February 9, 2004), 
http://www.sciam.com/print_version.cfm?articleID=000170D6-C99F-l01E-861F83414B7F0000; Michael 
Kanellos, "IBM Gets Chip Circuits to Draw Themselves," CNET News.com, http://zdnet.com.com/2100-
1103-5114066.html. See also http://www.nanopolis.net/news_ind.php?type_id=3. 
22.
IBM is working on chips that automatically reconfigure as needed, such as by adding memory or accelerators. 
"In the future, the chip you have may not be the chip you bought," said Bernard Meyerson, chief technologist, 
IBM Systems and Technology Group. IBM press release, "IBM Plans Industry's First Openly Customizable 
Microprocessor," http://www.ibm.com/investor/press/mar-2004/31-03-04-1.phtml. 
23.
BBC News, " 'Nanowire' Breakthrough Hailed," April 1, 2003, 
http://news.bbc.co.uk/1/hi/sci/tech/2906621.stm. Published article is Thomas Scheibel et al., "Conducting 
Nanowires Built by Controlled Self-Assembly of Amyloid Fibers and Selective Metal Deposition," 

---

Proceedings of the National Academy of Sciences
100.8 (April 15,2003): 4527–32, published online April 2, 
2003, http://www.pnas.org/cgi/content/full/100/8/4527. 
24.
Duke University press release, "Duke Scientists 'Program' DNA Molecules to Self Assemble into Patterned 
Nanostructures," http://www.eurekalert.org/pub_releases/2003-09/du-ds092403.php, referring to Hao Yan et 
al., "DNA-Templated Self-Assembly of Protein Arrays and Highly Conductive Nanowires," 
Science
301.5641 
(September 26, 2003): 1882–84. See also http://www.phy.duke.edu/~gleb/Pdf_FILES/DNA_science.pdf. 
25.
Ibid. 
26.
Here is an example of the procedure to solve what's called the traveling-salesperson problem. We try to find an 
optimal route for a hypothetical traveler among multiple cities without having to visit a city more than once. 
Only certain city pairs are connected by routes, so finding the right path is not straightforward. 
To solve the traveling-salesperson problem, mathematician Leonard Adleman of the University of 
Southern California performed the following steps: 
1.
Generate a small strand of DNA with a unique code for each city. 
2.
Replicate each such strand (one for each city) trillions of times using PCR. 
3.
Next, put the pools of DNA (one for each city) together in a test tube. This step uses DNA'saffinity to link 
strands together. Longer strands will form automatically. Each such strand represents a possible route of 
multiple cities. The small strands representing each city link up with each other in a random fashion, so there 
is no mathematical certainty that a linked strand representing the correct answer (sequence of cities) will be 
formed. However, the number of strands is so vast that it is virtually certain that at least one strand—and 
probably millions—will be formed that represents the correct answer. 
The next steps use specially designed enzymes to eliminate the trillions of strands that represent wrong 
answers, leaving only the strands representing the correct answer: 4. Use molecules called "primers" to destroy 
those DNA strands that do not start with the start city, as well as those that do not end with the end city; then 
replicate the surviving strands, using PCR. 
4.
Use an enzyme reaction to eliminate those DNA strands that represent a travel path greater than the total 
number of cities. 
5.
Use an enzyme reaction to destroy those strands that do not include city 1. Repeat for each of the cities. 
6.
Now, each of the surviving strands represents the correct answer. Replicate these surviving strands (using 
PCR) until there are billions of such strands. 
7.
Using a technique called electrophoresis, read out the DNA sequence of these correct strands (as a group). 
The readout looks like a set of distinct lines, which specifies the correct sequence of cities. 
See L. M. Adleman, "Molecular Computation of Solutions to Combinatorial Problems," 
Science
266 
(1994): 1021–24. 
27.
Charles Choi, "DNA Computer Sets Guinness Record," http://www.upi.com/view.cfm?StoryID=20030224-
045551-7398r. See also Y. Benenson et al., "DNA Molecule Provides a Computing Machine with Both Data 
and Fuel,"