Friday, January 30, 2004

Bad letters or just scrambling journalism ?

I have read in French newspapers the study by Cambridge researchers showing that the scrambling of letters in words did not matter as long as the first and last letter be in the right place. This is definitely true in English, but I have difficulty to believe this is the case in French (replace the words in the previous link by your favorite language). Unless maybe it was just simply bad journalism ? Further research could probably benefit from this study , it would be interesting to see how much entropy is generated from changing letters in a word and how this would affect the ranking of the generated tree in Figure 1.

Thursday, January 29, 2004

Breaking the rules

As Peter Norvig was pointing out in the Making of the Gettysburg Powerpoint Presentation, sometimes it pays to break the rules. He was refering to the fact that had Abraham Lincoln made his Gettysburg address with Powerpoint (the traditional tool used to convey information these days) it may not have been as memorable. How many times has one gone through one of these powerpoint presentations which could never provide an adequate understanding of the subject at hand ?

For the Columbia, it was probably the straw that broke the camel's back. How easy was it for that presentation to NOT convey the fact that previous experiments had been performed with very small foam samples ? Hindsight is always 20/20 and as Feynman had done before (this is nicely narrated in What do you care what other people think ? ), one needed only one test to realize how bad the destruction on the wing was (such test was performed last July in San Antonio - one can see the leading edge of the wing before the test and after the test.)

Our Starnav 1 camera took pictures above that left wing for about five days. Because the camera was not designed to look for details too close to the spacecraft, we will never know if what we see in that first pictures ten hours into the flight had any relevance (for those interested, this is the photo of the beta series in this white paper.) What is for sure though is that even though we were at Mission Control, it sure did not look like to any of us that anything was amiss... Nobody broke the rule. Maybe we should have disappeared and talked more in the halls after the first coffee break.

Wednesday, January 28, 2004

Better Sleeping Technology IV

Here is another reason as to why you should have a good night sleep (A new study offers new clue to good memory.) While it is interesting to have college students trying to remember different items after a good night sleep, it probably also negate the fact that most older people ( I mean older than college students), while sleeping as long as they want, have sleep patterns of lesser quality. Somebody recently asked me what technology could also be included in this "Better sleeping technology program". Well for one, it seems pretty obvious that some of this technology include acoustic devices evaluating the resonnance of one's room, mechanisms for storing and dating noises made during the sleep, light collecting devices for evaluating its influence on waking mechanisms, cameras evaluating the flora in your mattresses (are you allergic to them, do you sneeze because of them and wake up because of it ?), cameras evaluating your movements over night, sensors evaluating temperature and humidity. Evidently, all this information would have to be recorded on a noiseless computer.

Friday, January 23, 2004

Checking bugs

Maybe this this Floating Point Simulator module could help students of computing into understanding the intrincacies of round-offs and numerical imprecision. More specifically the ones inducing disasters.
Better sleeping technology III

Just as I was talking about it, here is another study (some more details here)performed on college students on the need for a period to "sleep on it." We all knew it affected learning for the young and the health of the old. I did not realize my brain was shrinking as a result of jet-lag/sleeping pattern disturbances though. On a related note, the process of memorization is intriguing. I have observed that most people flying on the kc-135 did not remember much of their experience. This is why the french caravelle and the Weightless Wonder IV and V had LED parabola counts in them: People in the plane would not realize how many more parabolas were left. Most people flying in these birds generally do not realize they are flying 2 to 3 hours. The short term memory is deeply affected. So one has to wonder how weird it must be for astronauts to sleep. In most cases, they don't as they are too excited. Maybe some technology other than cheaper mattresses will come out of NASA's investigation of their astronauts sleep patterns especially in light of the new Moon-Mars initiative. Anyway, if anyone could point me to it, I am trying to find this study that showed that a good night sleep or a rather "boring" time between being exposed to facts and trasnfering them in the long term memory part of the brain was found to be about 6 hours....

Tuesday, January 20, 2004

We also need cheaper sleeping technology

Cable just reminded me that most people see only one thing when buying mattress and this is the price. I have to agree 120% since I still cannot fathom why it takes me a $100 to buy the most rudementary mattress. However, the sleeping technology I was mentionning is not just the mattress per se. It also includes some electronics and recording capabilities for diagnostics purposes. We are talking about much more advanced than a confortable pillow. So, in line with how disruptive technologies end up being large markets, one has to think of specific niche where one can grow the market. Affluent people, such as customers of the sharper image, could be part of that constituency but I think one should be able to find a much broader one. I have to think about it.

Friday, January 16, 2004

We need better sleeping technology

No matter how much I search, I am always amazed at the lack of intelligent solutions for sleeping better. Sure, you can always pay for a better mattress, even buy one of these NASA mattresses. But why does one need a visit to the doctor and then to the sleep clinic to figure out that you are snoring too loudly, have sleep apnea, grind your teeth, sleep walk, talk aloud, never reach deep sleep....This is all the more important that since a bad sleeping experience is at the root of many health problems or early symptoms of diseases. The surprising part is really that since sleeping is about 1/3 rd of our lives one would expect an overwhelming amount of services or products to make this a more pleasurable experience. Eating for instance, takes only a fraction of time in our lives compared to sleeping. Yet, there is an entire industry devoted to it, a multi-billion dollars one at that. Just imagine now, companies dedicated to sleeping better...

Saturday, January 10, 2004

Is the point of going into zero-g to have fun ?

Four student teams I am sponsoring, will get to ride on KC-135 this coming spring and summer. I have always wondered about the actual return on investment of this type of endeavor. Last year was my real test point in this area. Here are my initial thought after a year :
- one can identify pretty accurately the leaders in these teams. You know that you can count on them later on different projects without having to be on their tails. In my book, it pretty much defines intelligence.
- this is the first time students get to understand the issues of funding and accounting that we as researchers get to be facing on a daily basis
- there is healthy turn over from the time the teams get selected to when they fly. For some people this is the first time they face the reality that they cannot do everything well. Last year, in two of the selected teams, we had a turn over of about five people out of eight.
- all the flying experiments got to be made in the last eight days before flying. No matter how hard one tries to convince people that it will have an impact on their lives afterwards, I expect the same thing this year.
- one can get some pretty interesting movies/pictures that can eventually be a support for a funded project. I am expecting that with lessons learned from the first year, the results of the second year will become even more exciting.
- Because they have been confronted with a very different environment early on, I personally fully expect that the students that have gone through this, will eventually be the people that, when confronted with the general conformism of a large organization like NASA, will be able to stand up, speak out and say when things are unsafe or will come up with very inventive and different solutions like John Aaron did.

Friday, January 09, 2004

We need better maps II

Following up on one of my concerns. People in Morzine, a french ski resort, came up with this concept: ski nav. It looks like an HP palm of some kind, retroffited with a different logo and a GPS. At 60 euros a week, it still is too expensive for a map in Paris. Well at least, being the power drawer they are, they can always warm you up when you are lost and cold.

Still looking for that sub-10 euro/dollar solution though....I think in an urban environment, one really needs to know pointing directions, answering the question: am I heading the right direction ?

Most tourists I have seen had taken the wrong turn only once. It was enough. And they get lost in broad daylight, no starnav for you...

At long last the browser is the GUI

What a great way to do some rapid prototyping on a software GUI: Use the browser as Desktop UI. Everytime I want to go into GUI design I either rely on Matlab or Python, both of them have interesting approaches (WxPython) but generally too complex for a week end project, especially when you are not a programmer by trade.
Comparing apples and apples when you should be looking at who's eating the apple..

This " Language Performance Round-up: Benchmarking Math & File I/O " article provides the wrong impression on languages use. For all its worth, Fortran would probably do better on all these if it were given a try. Similar types of findings were found in the great language shootout (even available for win32 machines). Yet, it's been my experience that this type of benchmark is only needed at the end of a development stage. It's fine and dandy to see Java doing better than say python, but if it takes your developers three to six times longer to develop it and longer to debug it, it won't matter that Java is quicker. Your project will run into the difficulty of not being responsive to either yourself (one can get bored easily) or the people giving you work, and you may be out of business quicker than java can run this benchmark. For that matter, it just doesn't make sense to look at a math benchmark without looking at matlab or octave.