Posted by
emmett
on from the one-small-step-for-an-accountant dept.
Spudley writes "Space.com is reporting that NASA are likely to scrap a number of planned missions, due to increasing mission costs. The cost rise is attributed to more failsafes being used, after the recent failures of a number of 'cheap' missions."
Problems with probability...
by
laborit
·
· Score: 4
It's actually kind of unfortunate that NASA has chosen this time to implement extra safety features, since it makes it more difficult to tell if they work. Statisticians are familiar with the concept of regression to the mean (or just regression): after an extraordinary period, you're most likely to have an ordinary one.
This simply reflects the fact that most of the things you do are going to have an average outcome, due to the definition of average. So if you have a string of great victories, your ordinary, expected performance will look like you're going into decline. If you have a string of failures, it will look like you're improving.
The textbook regression foulup is an experiment in which people are punished for failure and rewarded for success. Since failure naturally leads to improvement, it ends up looking like punishment helps and reward hurts.
So... NASA will implement a lot of safety features. And the missions will be more successful even if the features do nothing at all, just because they're going to have to come out of their slump sometime.
Then it will look like space missions have to be expensive to succeed, and we'll be locked into this paradigm...
- MC
--
----- Go ahead, blame me... I voted for Nader!
Re:Problems with probability...
by
wafath
·
· Score: 4
1) Repeat after me: "Lady Luck has no memory." Now write it on the chalk board 1000 times. If I am flipping a fair coin, and I get 4 heads in a row, what is the probability that the next flip will be a head? 50%. Believing anything else will get you into gambler's anonymous.
2) These are not random events. Your spacecraft getting hit by a mico-meteorite en-route is a random event. Your spacecraft digging a hole in the surface of mars because some asshole company decided to do english units is not a random event. NASA knows and is willing to live with the random. What NASA is trying to do is prevent the FUBAR's that throw away very expensive spacecraft.
W
the dispair banner ad says it all
by
omarius
·
· Score: 4
Yet another blow to the star-filled hopes of my generation; a generation that grew up on the tail end of the Space Race, who still have their copies of TIME magazine from when Viking landed on Mars. At age 10 I fully expected to get to ride on a Space Shuttle one day. At age 26 I am sad that the US seems more interested in immediacy and BS politics than expanding the role of humanity in a universe which happens to be larger than the Republican convention, no matter how it looks on TV.
They're not really in a comfortable place after all. Their funding relies on both their public image and intensive politicking (is that a word? Ah well, you know what I mean) in Congress. Every time something goes wrong with a mission it makes them look bad, no matter whose fault it really was.
This is a shame since NASA is a worthwhile endeavour and deserves a better deal than it gets from the American government. But we can do without a few minor missions in the name of getting the more important ones working - I don't think anyone would deny that money would be better spent on a Mars mission than say a Pluto one.
In the long run though it may well be that NASA fall behind other agencies and corporate interests. The public is simply not up for a huge space program with its attendant costs. NASA are trying to make space flight cheaper, but it costs money to save money in this case, and at they rate things are going, that'll be money they don't get.
Slashdot Mirror
It's actually kind of unfortunate that NASA has chosen this time to implement extra safety features, since it makes it more difficult to tell if they work. Statisticians are familiar with the concept of regression to the mean (or just regression): after an extraordinary period, you're most likely to have an ordinary one.
This simply reflects the fact that most of the things you do are going to have an average outcome, due to the definition of average. So if you have a string of great victories, your ordinary, expected performance will look like you're going into decline. If you have a string of failures, it will look like you're improving.
The textbook regression foulup is an experiment in which people are punished for failure and rewarded for success. Since failure naturally leads to improvement, it ends up looking like punishment helps and reward hurts.
So... NASA will implement a lot of safety features. And the missions will be more successful even if the features do nothing at all, just because they're going to have to come out of their slump sometime.
Then it will look like space missions have to be expensive to succeed, and we'll be locked into this paradigm...
- MC
-----
Go ahead, blame me... I voted for Nader!
-Omar
They're not really in a comfortable place after all. Their funding relies on both their public image and intensive politicking (is that a word? Ah well, you know what I mean) in Congress. Every time something goes wrong with a mission it makes them look bad, no matter whose fault it really was.
This is a shame since NASA is a worthwhile endeavour and deserves a better deal than it gets from the American government. But we can do without a few minor missions in the name of getting the more important ones working - I don't think anyone would deny that money would be better spent on a Mars mission than say a Pluto one.
In the long run though it may well be that NASA fall behind other agencies and corporate interests. The public is simply not up for a huge space program with its attendant costs. NASA are trying to make space flight cheaper, but it costs money to save money in this case, and at they rate things are going, that'll be money they don't get.