Comments on: Long-duration Solar-powered Flight

By: Bryan

Bryan — Sun, 18 Jul 2010 00:52:09 +0000

I was in the business and I hate it, PJ, because it won’t work and it makes getting real and important information harder by burying it in even more data than before. It’s like trying to do a criminal investigation by calling every name in the phone book and asking them what they were doing on a particular date and time – you get a lot of data, but it doesn’t help to solve the crime and it annoys a lot of people.

You get important information by limiting the scope of the investigation. The only use I ever found for a polygraph test was to eliminate people from suspicion when you have an inside job.

The no-fly list is criminally stupid. Sikhs use a very limited number of names. If you put a single Singh on the list you are affecting millions of people. Names are never used as a primary key in any serious data structure because they are rarely unique. You need a lot of information to create a unique identifier which is why the Social Security account number is so widely used – it is unique.

By: paintedjaguar

paintedjaguar — Sun, 18 Jul 2010 00:07:37 +0000

I hear you, Bryan. First you gather a haystack, then you go looking for needles. But you misunderstand me. What makes me feel somewhat paranoid is more the possibility of landing on the wrong list and becoming a specific target through some trivial error or misunderstanding. In fact, data overload, incompetence and faulty algorithms may make such an event more likely, and then you are in Kafka-land, as we've seen with the no-fly lists. Lack of results won't slow this train down either. There's too much political and monetary profit to be had from building a security state. I admit to prejudice, stemming from a couple of wrong time, wrong place experiences and then there's just the general inconvenience and unease of constant surveillance. None of this is the fault of the technology, though.

By: Bryan

Bryan — Sat, 17 Jul 2010 22:09:53 +0000

Actually, that is a very intelligent move, Jams, as it eliminates the weight and drag of landing gear, which can be quite significant.

Frankly, Jim, I would love to see a few of these over the Atlantic during hurricane season as we have lost one of out major satellite systems thanks to inaction during the Hedgemony, and there is a wealth of specific data that can only be obtained by a direct overflight. The wind speeds at various altitudes can tell a lot about a storm’s potential for development and tracking.

By: Bryan

Bryan — Sat, 17 Jul 2010 22:00:40 +0000

Mr. Duff, I don’t worry about the debt because the US has its own fiat currency and the debt is irrelevant under current economic conditions. The only people who are concerned with the national debt are those attempting to drive the world into the next, even greater depression. Something you will see under a Tory government in Britain.

If you don’t get demand up, there will be a complete collapse of the markets. Business doesn’t intend to spend, so the job is left to government. So it always was, so it will always be.

By: Jim Bales

Jim Bales — Sat, 17 Jul 2010 12:37:24 +0000

Back when I was building autonomous underwater vehicles (in the 90’s), I had a long chat with Prof. Kerry Emanuel of MIT’s atmospheric science program (one of his major contributions to the field, as I understand it, is his paper “The Theory of Hurricanes”).

He noted that prior to the 60’s/70’s, NOAA sponsored lots of meteorlogical flights from airports around the country. However, when we sttarted putting up weather satellites, the number of met flights was reduced. And, while satellites gave forecasters dense data sets on variables they otherwise could not get, the flights gave information on the vertical structure of the atmosphere that satellites could not get.

Kerry was involved in efforts in the 90’s to create “eternal” solar-powered aircraft for meteorlogical measurements (I believe this was the “Helios” aircraft, but am not certain). One hopes that the technology used in Zephyr will reach that application.

best,
Jim

By: jams O'Donnell

jams O'Donnell — Sat, 17 Jul 2010 10:10:45 +0000

That is reassuring Bryan.

As for the plane I am most impressed by the take off system!

By: Bryan

Bryan — Sat, 17 Jul 2010 05:19:04 +0000

In reply to paintedjaguar.

The thing that most people who have never been in the intelligence field don’t realize, is that, while they keep collecting more and more data, they don’t hire and train the people necessary to actually look at the data. It becomes a matter of the more data you have, the more anonymous people become.

If you have a standard deck of cards, an individual would have a 1 in 52 chance of finding any particular card. If you take 10 decks of cards and shuffle them all together, that individual still has the same odds, but may have to look through as many as 510 cards to find it. Now make a thousand decks, and while the odds stay the same, the work increases. At a million you have to hire more people and split up the cards among them, which means any particular pile may have none of the card you are looking for.

Throw in the fact that generally people aren’t sure what they are looking for, so more data just makes the job harder.

The normal answer is to put it on a computer, and let the computer find it. The problem is that search algorithms are more art than science, which is why Microsoft has never been able to even be an annoyance to Google. If you don’t have the art, money makes no difference. You don’t teach people to create the logic, you discover people who can.

How many TVs can one person watch? TSA has discovered that reading passenger screening scanners is an art that a lot of people can’t master.

Relax, it will be a long time before the government finally figures out that all of this data is a bug, not a feature.

By: paintedjaguar

paintedjaguar — Sat, 17 Jul 2010 04:35:48 +0000

As a rule I'd be cheering on any technological breakthrough, but lately it's hard to muster enthusiasm for a great new surveillance tool. O tempora! O mores!