Aspen Forum
August, 2003
Q) How many test sites are in the program?
A) There are currently 42 different companies in the test cycle with another 24 in the pipeline.
Q) Why not build your own sites?
A) Bravo! Good question. We made a statistical decision that to test in a REAL environment would produce REAL data. Meaning, there are so many variables in hyper-space....so many potential flaws in any actual satellite structure...the Powers that be demanded an Actual Real Test Eco...before they would complete the second phrase of the funding....We are not selling widgets here....The more difficult the test eco the better....including the variables of dealing with dozens of website owners...The unexpected variable is what we were searching for....
Q) Who did the art work on your site?
A) I'll take that one....The twins. I have twelve year old nieces, and they based their design upon a Project Weather site....They also received school credit...Ha..Ha..
Q) Can you speak a little bit about the origins of the project?
A) Another good question. Odd where research grants come from! In this case it was a mixture of two PHD papers, the Hilltop document pertaining to a search engine patent, and of course the aerospace industry played a major role,
Q) What was the main thesis that prompted your team?
A) In my opinion, it began with weather analysis. And of course when this work was brought to my attention in relation to partial differentials, my graduate students laughed that it was non-linear, and said we ought to ignore it....(Laughter from two graduate students)
A) If I may continue. We view traditional weather orbiting vehicles as possessing limitations...i.e. pilot control problems we often encounter at the high atmosphere edges of Space.... Don't tell NASA, but sometimes when confronted by a vexing problem those of us in Pure Research throw the chalk board away, go outside of the Box, and just take a shot.....We usually fail, except that it shows us where not to go. In this case, we were stonewalled by the lack of control of a certain sensitive weather tool at high altitude....Well, lucky for us, we went outside the box. I'm a little old for the computer X generation, however one of Professor Murrin's Grad students....made a math association between the Hilltop Paper published by a major Search Engine and the algorithms we were noodling with....Yes, noodling is a specific term....Look it up....Now, after months of discussions, arguments, staff meetings, funding proposals, and lots of academic infighting, calmer heads prevailed and agreed that this course of action was too radical....
Q) Even for the Ivy League? (Laughter from the media)
A) Harsh.
Q) What was the first step?
A) We received a small grant from MIT. They liked the math. And, we figured, no pun intended, that even if it was a bust, that the numbers were compelling...
Q) Was there any pressure?
A) You betcha'. Anytime you play on the Governments dime you have a host of eyeballs...
Q) What was the physical beginnings?
A) We literally sent up a test balloon. A remote control device, a primitive algorithm was tested and we were overwhelmed by the results. Our algorithm enabled the test vehicle to receive our commands similar to a radar device....meaning bouncing the waves off each other....And, as exciting as this was we could see that we needed an environment to road test the math.
Q) So you had to change gears?
A) Yes, in a big way. And, once again going completely outside our comfort zones we began to run tests on the World Wide Web. Now, as you might imagine the senior team members had limited experience in this realm with the exception of purchasing a few items on E-bay.... (Laughter)
Q) What attracted you to the web?
A) It was the most difficult platform we could encounter and still mine the type of data we needed. In fact, we embarked on a long period of documenting what we would be looking for, and this process opened us up to the amazing, the infinite variable tables presented in this playing field. Meaning, the more we looked at it, the hardier it became and the more we liked it. Deep Space is a hairy place....A little joke we tell, when we physicists are sitting around the watering hole. Rules that we are basing high order solutions upon can suddenly be altered. So, by trail and error we discovered the WEB to be a similarly Hairy place....
Q) You started with Incoming Links, the traditional boost to placement?
A) Yes, monkey see, monkey do, even at Harvard. However, we soon discerned that it only takes a minimum of "Boost" to fuel the direction (trajectory) of a websites climb up the chart. It is not the amount or "PR" of links which causes a site to achieve a higher rank, but the outgoing links from that site. Hilltop was the key for us. It showed us that to control the orbit of deep space vehicles we would need to write scripts which enabled us to bounce controlling "factors" back to the vehicle; weather it be a space vehicle or a website.
Q) So incoming links do matter?
A) We are not, what do they call themselves...SERPs? Web experts, or placement companies... In a nutshell, our process is sending outgoing links from a "Test Site" to Project First, where we bounce back an algorithm informing the web site of the "properties" it needs to compete in this field of natural selection. Darwin would love Google.