Sweet! The song "Carlotta Valdez", from their first album (Where Have All the Merrymakers Gone?), is one of my all-time favorite songs (its about Hitchcock's movie, Vertigo).
Uh huh... Could you, perhaps, give us the name of "somebody" or provide a reference to the study? I'd be curious as to how "somebody" got access to all these "mass murderers", how many they tested, how they recruited them, etc:
Dear Mass Murderer,
While you wait on Death Row, I wonder if you would interested in participating in a study. You will receive $15 per hour....
You switched positions, and admitted his point! If they are required to maintain diversity in their workplace, which you just claimed, and they are calling around to a diverse set of people, they have an intent to hire some of them (assuming they interview well). They aren't doing it just so that they can say "we interviewed women, blacks, and mexicans, before we hired our original choice, John WASP".
I think you need to reread the first post you responded to, and understand what it meant. He is simply saying that if someone contacts you about a job, it isn't out of political correctness (most likely), rather it is because they at least have an intent to go through the process of determining if you are someone they want to actually hire.
First you responded by implying that companies need to give the appearance of diversity, thus interviewing people they don't intend to hire. Then, you responded that they actually DO want to hire a diverse range of people, and so the interviewing is not just for show. Then you insult the poster for a position he never took, and which YOU failed to understand (based on the context of this thread, and his original post).
I've worked closely with NASA, as a contractor, for over 12 years, and one thing I've noticed (that I found shocking, frankly) is that during the Clinton years a number of regulations were lifted in order to allow civil servants (and thus me) to get more work done each day. In particular (and at the behest of Al Gore, which he loves and deserves to take credit for) certain purchasing regulations were modified. This meant that, by 1997 or so, when a decision was made, equipment could be obtained in days rather than weeks or months, and work could proceed right away when people were most enthused, motivated, etc. When I started in 1994, it could take months between the decision to buy something and obtaining it (with LOTS of paperwork and forms to fill out along the way), even very inexpensive or common items.
In the past five years, I've watched many of those changes revert, often to more regulated and restricted forms. It has become MUCH more difficult to act on decisions as of late, enough so that I've finally given up working with NASA.
I have begrudgingly come to a conclusion (anecdotally based, I admit) that since conservative philosophy says "government is inefficient", they actively work to pass regulations that make it inefficient, in order to justify their philosophy. And if the liberal philosophy is that "government has an important role to play", I have seen these legislators actively work to make more efficient government agencies (not always with success). The paradox is that the conservatives have imposed MORE regulations, while the "liberal" Clinton adminstration relaxed them. And make no mistake, the improvements I'm talking about were Driven by the Clinton administration, not the conservative Congress.
So, to the point at hand, it is no shock to me that FEMA has acted so incompetantly (and they have; I watched the director of FEMA make a complete jackass of himself, giving clueless answers to incredulous reporters about the situation after Katrina; any idiot on the street who watched TV or read a newspaper would have sounded better informed). My new hypothesis is that the current administration can't see any value in making any government agency NOT be imcompetently managed and run. To do so would imply that such agencies had some value, and the the current administration's conservatives demand a world view where they should not exist at all. Certianly not "socialist" agencies, such as FEMA.
I'm trying to remember what I was taught, before I took a number theory course in college, and I can't honestly recall (I'm now mid-thirties)... It may well be that high-school and elementary curriculum went on teaching this long after "mathematicians" had abandoned it; I can't say, although it hardly seems to be contentious enough of an issue outside of professional math circles to force new printings of textbooks.:)
Thanks for responding with your experience, btw; I do find it interesting.
I'm not sure if you are joking. Those references are supplied to support the assertion that 1 was considered prime at some point. But just because the reference was written in (say) 1984, doesn't mean that 1 was still considered prime in 1984. Just that it talked about the issue at that time. I googled around to try to find a more definitive time for when 1 came to not be considered prime, and couldn't find anything enlightening.
However, I found a reference in a paper by Eric Temple Bell, "The Queen of Mathematics", that the Fundamental Theorem of Arithmetic was proved by Ernst Zermelo in 1912, and I would have to speculate that "1 is prime" must have been going out of favor at least by then.
Interestingly, as 1 was considered prime at the time of the devising of the Goldbach Conjecture, and now is not, it seems the changeover may have made the conjecture less likely to be true (ie. possibly a very significant change). Mersenne Primes also needed a more complex definition when 1 was prime.
I would like to know more about how 2 was NOT considered prime...
Interestingly, it says that, at one time, 1 was considered prime and 2 was not. Pretty amazing, considering importance of the Fundamental Theorem of Arithmetic.
The Columbia Accident Investigation Board (CAIB) report talks about the switch to non-Freon based foam. For those interested in the details, please see CAIB report Volume VI, pages 29-30 (transcripts of interviews discussing the exact issue of the foam formulation and switchover from CFC to HCFC). Also pages 180-181.
Volume I also talks briefly about this issue (describing with pictures which areas of foam on the shuttle tank are what formulation) on pages 51 and 129.
In specific answer to your comment, they aren't phasing out all of the CFC-11 foam, and it is the foam they have used and still use for the hand shaped portions. CFC-11 is Freon based, according to the report.
From my reading, the switch from CFC to HCFC foam does not appear to be implicated as a safety problem, above and beyond the serious issue of foam shedding itself. I hope that those people, elsewhere in this thread, who are making such accusations, will at least read the CAIB report in full before continuing their claims.
The foam shedding was not 'dismissed' (your words)
I said 'the risk of wing impact damage was essentially dismissed' (CAIB Volume I, chapter 6, pg. 140, among other places). You are claiming I said something much broader, ie. that NASA disregarded foam shedding altogether, which I never claimed.
O-ring problem accepted as 'inevitable' (again your words)
I said "foam shedding was accepted as an inevitable consequence of launch". I was not talking about O-rings when I used that word, and I did not (again) say what you said I did.
The problem is - that statement is utterly false.
I find that you are the one making false claims, both in misquoting me, and in data you use to support your argument. You said, falsely, that "the worst burn-through incidents had been during warm weather launches.", and I posted a link to a chart that refuted this as false. The worst burn-through occured on the coldest launch (two-worst and two-coldest, if you include Challenger's last flight). Furthermore, less burn-through occured, on average, at warmer temperatures, indicating a trend. You ignored this all in your reply.
If I knew what "party line" you were referring to, I might be offended. I don't expect you (at this point, since you don't seem capable of comprehension) to understand my rebuttal. But stop making a false argument by putting words in my mouth. And if you accuse me of misrepresentations of fact, don't engage in them yourself. If you disagree with me, put me on your enemies list and be done with it.
Right. That explains why NASA had an extensive and ongoing program to reduce foam shedding... That's why we had so many images from NASA examining possible wing impacts... (The problem was the risk of *RCC* impact was downplayed.)
I said that "the risk of wing impact damage was essentially dismissed." Much of the RCC is on the wing edge, so we are in agreement. NASA having an "ongoing program to reduce foam shedding" is also exactly my point. They were aware that foam shedding caused shuttle damage, and that the shuttle was not originally designed to be hit by this foam. But, in becoming a normal event for launches, they had come to accept (I argue) an unrealistic view of how hazardous the situation was. Yes, they did try to reduce it, but nevertheless it continued to occur, and they didn't fully appreciate the extent of the damage it could cause. CAIB report, Volume I, chapter 6 discusses this in much detail, and I won't go over it here. But look at page 141, in particular, to see an example of your statement, about RCC/wing impact being downplayed.
Right. That's why NASA had an extensive and ongoing program to analyze and correct the joint problems. (The design finally accepted post Challenger was actually approved in 1984..)
Again, I argue that "an extensive and ongoing program to analyze and correct the joint problems" was moot, when clearly the problem was that the design was fundamentally flawed (burn-through was not in the spec, and it was occuring.) and needed to be corrected before further launches. They had to be redesigned anyway, at enormous cost (seven lives, a new shuttle, lost schedule, etc.). The decision to postpone launches until the problem was fixed would have been the politically tough, but ultimately correct, decision.
Of course it wasn't understood... The engineers had been telling management all along that things were fine - the existing design was low risk and the fix was in. (Which, incidentally, is the same thing the engineers said prior to Columbia's last launch.) *That* is why management was so testy with the engineers on the eve of Challenger's launch - with no change in evidence, the engineers were changing what they'd been telling managment. (Especially since the worst burn-through incidents had been during warm weather launches.)
Yes, there was a complicated interaction going on in the Challenger accident. I am not as familiar with that report, so I won't refer to it directly. But E. Tufte, in his book "Visual Explanations", has a now famous assessment of the situation, and in particular, the rather poor presentation of data to support the engineers decision to NOT recommend a launch at such a cold temperature (The only "no launch" recommendation given by Morton-Thiokol up to that point in time). I'll base my comments on the analysis from that book.
You make my point for me: "with no change in evidence, the engineers were changing what they'd been telling management". Even if I accept what you say, as truth, that engineers had been claiming that the O-rings were safe ("fixed", as you say) even at near freezing temperatures, it points out that when the engineers finally did come to management (both Morton-Thiokol and NASA) to urgently discuss a re-assessment of this opinion, their "no launch" recommedations were not given the proper merit. NASA now had to be convinced that the launch was UNSAFE, rather than accept that anomalous conditions should require launches to be proven SAFE. If management truly was "testy", as you say, then they shouldn't have been making such important launch decisions in that state of mind. People make mistakes in judgement when they are "testy" (in my experience). That accident was as much a failure of people, as it was of equipment (I think we both agree on that).
Tufte points out, in his book, that since the situation became one where the engineers has to prove t
I agree. After years of helping out my family members and friends with their computers, I've learned that "no good deed goes unpunished".
If there is a way to change to a cheaper AOL service, tell your Dad about it. But think *3* times before taking on the responsibility of changing his email patterns (and protecting his data); unless your Dad is pleading with you for the change, it is almost never worth it to get involved. The effort will not be appreciated nearly as much as you hope, and there will be bad feelings if things don't go absolutely 100% correctly.
You say that foam shedding is "expected", and yet that misses the entire point of the post-Columbia investigation results. (And post-Challenger) for that matter.
Foam shedding was NOT expected when the shuttle was designed; the foam came in late in the game apparently. The tiles and wing structures were not designed with the intention of being struck by so much debris at such high speeds, and so initially the foam was seen as a real risk to the shuttle integrity. However, over the course of many launches, as the foam debris strikes piled up, it came to be an "expected" event, and the risk it posed was downplayed. And although you may "expect" a car with failing brakes to not crash everytime you move it, that doesn't mean you should drive it that way.
The mindset that foam was not likely to cause loss of structural integrity, was so strong for NASA shuttle managers that when the Columbia launched for the last time, they did not have a proper way to evaluate the extent of damage from foam. They had lots of data on small foam hits, and this was a BIG foam hit, and at very high speed. But since previous ones hadn't broken the shuttle, there was an "expectation" that even big foam hits probably wouldn't; they really weren't sure. The engineers obviously knew that F=ma could be a large number, even for small m, and they attempted to adapt some tools to calculate the possible damage. But the mindset that "foam is an expected event, it hasn't led to shuttle loss before" was already too well entrenched, and so the risk of wing impact damage was essentially dismissed.
This was made famously clear, when during the post-Columbia loss investigation, Scott Hubbard demanded that the foam gun tests be performed (ie. launching foam at a mock-up of the shuttle wing) in order to convince shuttle flight managers of the risk it posed. At that point, the majority of the investigation team already knew that the foam had been the cause and didn't press for those tests, but Scott knew the NASA culture, and knew they would not accept it unless proven beyond a shadow of a doubt. Against NASA's objections, the tests were performed, and we all saw the results.
It is important to remember that the exact same problem in mindset doomed the Challenger flight. The O-rings were not designed to allow any burn through of the rubber. When it started to occur, it was accepted as an inevitable consequence of launch, rather than a fatal design flaw. Because it had not burned completely through on previous flights, it was accepted that some burn through would happen, and not be catastrophic. NASA management downplayed the risk so much that even when engineers insisted that such a failure was more likely on a cold launch, their objections were not well understood. Burn-through was "expected", so how much of a risk could it be?
There is a lot to be learned about engineering, and management, by these examples. Engineering has been called the art of compromise. However, when so much compromise has been made, it becomes easy for people to not properly evaluate all the consequences. Meanwhile, management has pressures that go outside the engineering realm, and the psychology of that situation can lead to completely unrealistic assessments of risks and liabilities.
For another bit of fun, I just noticed that Google maps now has a map of the moon landing sites available, in celebration of the anniversary of the first landing.
As for exploring other systems that could be life supporting, as a NASA contractor, I hope we a) make it a bigger priority, and b) don't fuck it up by contaminating anything.
I used to think, when I was young, that if we found extraplanetary life, in my lifetime, I would be alive to see an amazing change in orthodoxy, both religious, scientific, and otherwise. Now, both religion (crudely lumping the most popular ones together) and science seem to expect such a discovery, and each has already adapted and prepared for the day when such discovery is made. I'm personally hoping what we find completely defies prediction.
Yeah, he's inheriting from a dictionary type, which is a new-style class introduced in Python 2.2.
If the code author really wants to still support Python 2.1, he would have to inherit from type UserDict, assuming he isn't making use of other 2.2 features. I suggest an email to the author indicating a change in the requirements listing, or a change back to 2.1 compatibility.
FWIW, in the introduction to "Prayers to Broken Stones", Harlan exclaims that it was he who "discovered" Dan Simmons at a writers workshop (and that when he is long forgotten for his own work, he will be remembered for that fact alone). Dan Simmons, of course, went on to write the early Hyperion Cantos books soon afterwards and win the Hugo. Dan acknowledges the discovery in that introduction (although his recounting of the events is somewhat different).
FWIW, I've seen Justice O'connor on television, nehemently denying she ever said that, or anything like that. Maybe she's a liar, maybe the news is reporting BS (no, really?). In any case, the court decision on the 2000 election was total garbage (as was the recent eminent domain one; In both cases I don't mean necessarily that the outcome was crap, but that the stated REASONS for deciding the outcome were crap)
How the heck is a Firewall necessary to keep a default Windows box secure? In other words, if a Windows firewall is there to disallow services (or protocols) from receiving connections from the outside world, then what are these services, and why are they running in the first place?
I understand that by deceiving a user, a malicious service can be started up and listen on the internet, and become a vector for infecting your machine. But that requires an act of the user. If I NEVER enable any special services on my machine, than only the default services are running, and they must somehow be allowing malware to install, right? So, why aren't these services fixed, or disabled by default?
Finally, if these servies are necessary to the proper running of my machine, then when I use them the Windows firewall software will ask if I do not want to block that port, service, etc. Once that occurs, am I not just as unprotected as if I never used the firewall software? How does it really help?
So, that's a lot of questions, but I would appreciate an explanation. Are the attacks on windows solely due to users running malware directly, or are there vectors by which, without any user action (ie. no browsing w/ ActiveX controls, no javascript, no running malicious executables, no starting email attachments, etc) the machine can get infected anyway? If so, what are those services? It's not like a Windows machine, by default, needs to have an email/web/network disk/instant messaging service running, so why does it?
NOTE - I googled "insecure windows services" and got some info; indeed windows does have a bunch of services open to the world by default (un-f'ing believable). Can anyone say which ones are primarily allowing machines to become zombies?
A doctor friend of our family splurged on a Lisa when it first came out (1983, or so). I remember being most amazed when he pressed the big On/Off button, and all his applications saved (and closed their windows one by one) and then it powered down on its own. Clever idea that was a fair bit ahead of its time (resurrected in the 90s, I think, when costs went down).
I see your point, and won't belabor it; my very simple example was meant to illustrate how a crude version of this technique could operate, not to champion it for actual use. There are many implementation details (CPU usage, network meta-data, implementation complexity) that my example doesn't address.
If I send A and A^B to you, I could've just sent A and B instead.
Yes, but if you've sent me only A, I'm of no use to anyone who requests B from me. If you've only sent me A^B, I can be of use to others who already have either A or B.
So, it may be true that the idea isn't worth implementing, due to the realities and complexities of network dynamics. But the math proves that a computed bit (ie. A^B) is worth as least as much as either A or B by itself.
Yes, the same number of bits have to be transmitted, but with this technique, I may be able to get all the bits from more peers, more quickly.
Question: Does this mean that I'll be unable to do anything with incomplete files?
No. Incomplete files will still have the actual bits (not computed bits), but my simple two bit analogy starts to break down as an explanation of why. But let's make it slightly more complex:
I have bits 'a', 'b', and 'c'. I can compute (a XOR b XOR c) and send it. Then I send (a XOR b). Finally I send a.
The receiver gets the first two computed bits, and uses them to compute (a XOR b XOR c XOR a XOR b), from which they get bit 'c', thanks to the magic of XOR.
So, the main thing to understand is that even though bit 'c' was not transmitted, it can be recovered after only two bits are received. Following this same pattern, if we started with many more bits, we could recover 1 actual bit value after two are received, 2 bits after three are received, etc. Well, as long as we receive them in a useful order, we can.
So, using even this simple method, we can start recovering the actual file bits right away, without having to wait for all of the computed bit function values to roll in. The actual method used by the researchers is more complex than this, mainly to allow for the freedom of receiving and recovering the blocks of bits in random order. But it uses a similar idea of computing and solving linear boolean equations for the bits.
A simple example for those reading who don't understand, then some follow up comments:
Say I have bits 'a' and 'b', that other people want.
I could sent bit 'a', then bit 'b' to receiver FOO, who can pass them on to others. However, if I send bit 'a' first, and others want 'b', they have to wait.
Now, instead of transmitting to FOO bit 'a' then bit 'b', I send to FOO ('a' XOR 'b') first, then either bit 'a' or bit 'b'. I'll end up sending FOO the same amount of information (assuming the order is specified in the protocol itself).
BUT, and here's the cool part. If someone already has 'a', they can get ('a' XOR 'b') from you, and complete their set of data (bits 'a' and 'b'). Furthermore, if someone already has 'b', they also get ('a' XOR 'b') from you, and complete their set. So, by only downloading 1 bit, instead of 2, you can complete the set for others who already have one or the other bits.
Now, in practice it'll get a lot more complicated, and the method presented in the paper is not exactly like I describe, but the idea is that you can send data to help people complete their data sets, even though you yourself do not yet have the actual uncomputed data. Instead, you have a computed function of the data, which others can use immediately, and from which you can reconstruct the actual data later when you have more information.
The practical upshot is that the computed data is more valuable to other peers than the uncomputed data, as they may be able to use it to complete their data set, rather than wait for the remainder of the uncomputed data.
So, in reference to your comments, it may not be so much more practical to any one receiver; they still need to wait for all the data, in either computed or uncomputed form. But, for the network as a whole, it means that each receiver has many more options from which to download and compute each chunk, and thus make available to others. It is not hard to imagine that this can benefit the overall throughput of the network (which the authors of the paper claim).
Well, since you mention Numerical Recipes, it is obligatory to post one of the many rebuttals. Basically, the books have some okay discussions (and cover a very WIDE range of subjects) but their code is crap. I say that boldly, since I must maintain code that was originally developed using their C libraries. There have always been better alternatives, and especially these days when so much is available on the web.
Slashdot Mirror
Sweet! The song "Carlotta Valdez", from their first album (Where Have All the Merrymakers Gone?), is one of my all-time favorite songs (its about Hitchcock's movie, Vertigo).
Uh huh... Could you, perhaps, give us the name of "somebody" or provide a reference to the study? I'd be curious as to how "somebody" got access to all these "mass murderers", how many they tested, how they recruited them, etc:
You switched positions, and admitted his point! If they are required to maintain diversity in their workplace, which you just claimed, and they are calling around to a diverse set of people, they have an intent to hire some of them (assuming they interview well). They aren't doing it just so that they can say "we interviewed women, blacks, and mexicans, before we hired our original choice, John WASP".
I think you need to reread the first post you responded to, and understand what it meant. He is simply saying that if someone contacts you about a job, it isn't out of political correctness (most likely), rather it is because they at least have an intent to go through the process of determining if you are someone they want to actually hire.
First you responded by implying that companies need to give the appearance of diversity, thus interviewing people they don't intend to hire. Then, you responded that they actually DO want to hire a diverse range of people, and so the interviewing is not just for show. Then you insult the poster for a position he never took, and which YOU failed to understand (based on the context of this thread, and his original post).
I've worked closely with NASA, as a contractor, for over 12 years, and one thing I've noticed (that I found shocking, frankly) is that during the Clinton years a number of regulations were lifted in order to allow civil servants (and thus me) to get more work done each day. In particular (and at the behest of Al Gore, which he loves and deserves to take credit for) certain purchasing regulations were modified. This meant that, by 1997 or so, when a decision was made, equipment could be obtained in days rather than weeks or months, and work could proceed right away when people were most enthused, motivated, etc. When I started in 1994, it could take months between the decision to buy something and obtaining it (with LOTS of paperwork and forms to fill out along the way), even very inexpensive or common items.
In the past five years, I've watched many of those changes revert, often to more regulated and restricted forms. It has become MUCH more difficult to act on decisions as of late, enough so that I've finally given up working with NASA.
I have begrudgingly come to a conclusion (anecdotally based, I admit) that since conservative philosophy says "government is inefficient", they actively work to pass regulations that make it inefficient, in order to justify their philosophy. And if the liberal philosophy is that "government has an important role to play", I have seen these legislators actively work to make more efficient government agencies (not always with success). The paradox is that the conservatives have imposed MORE regulations, while the "liberal" Clinton adminstration relaxed them. And make no mistake, the improvements I'm talking about were Driven by the Clinton administration, not the conservative Congress.
So, to the point at hand, it is no shock to me that FEMA has acted so incompetantly (and they have; I watched the director of FEMA make a complete jackass of himself, giving clueless answers to incredulous reporters about the situation after Katrina; any idiot on the street who watched TV or read a newspaper would have sounded better informed). My new hypothesis is that the current administration can't see any value in making any government agency NOT be imcompetently managed and run. To do so would imply that such agencies had some value, and the the current administration's conservatives demand a world view where they should not exist at all. Certianly not "socialist" agencies, such as FEMA.
If he does, I'll bet he isn't sweating 14 gallons of fluid a second. Nor does he stomp on his own foot when doing the "monkeyboy" dance.
I'm trying to remember what I was taught, before I took a number theory course in college, and I can't honestly recall (I'm now mid-thirties)... It may well be that high-school and elementary curriculum went on teaching this long after "mathematicians" had abandoned it; I can't say, although it hardly seems to be contentious enough of an issue outside of professional math circles to force new printings of textbooks. :)
Thanks for responding with your experience, btw; I do find it interesting.
I'm not sure if you are joking. Those references are supplied to support the assertion that 1 was considered prime at some point. But just because the reference was written in (say) 1984, doesn't mean that 1 was still considered prime in 1984. Just that it talked about the issue at that time. I googled around to try to find a more definitive time for when 1 came to not be considered prime, and couldn't find anything enlightening.
However, I found a reference in a paper by Eric Temple Bell, "The Queen of Mathematics", that the Fundamental Theorem of Arithmetic was proved by Ernst Zermelo in 1912, and I would have to speculate that "1 is prime" must have been going out of favor at least by then.
Interestingly, as 1 was considered prime at the time of the devising of the Goldbach Conjecture, and now is not, it seems the changeover may have made the conjecture less likely to be true (ie. possibly a very significant change). Mersenne Primes also needed a more complex definition when 1 was prime.
I would like to know more about how 2 was NOT considered prime...
1... is... not... a... prime...
For info on why, see the mathworld prime number entry.
Interestingly, it says that, at one time, 1 was considered prime and 2 was not. Pretty amazing, considering importance of the Fundamental Theorem of Arithmetic.
1 is not prime, bot.
The Columbia Accident Investigation Board (CAIB) report talks about the switch to non-Freon based foam. For those interested in the details, please see CAIB report Volume VI, pages 29-30 (transcripts of interviews discussing the exact issue of the foam formulation and switchover from CFC to HCFC). Also pages 180-181.
p dfp dfi b_report_volume1.pdf
Volume I also talks briefly about this issue (describing with pictures which areas of foam on the shuttle tank are what formulation) on pages 51 and 129.
In specific answer to your comment, they aren't phasing out all of the CFC-11 foam, and it is the foam they have used and still use for the hand shaped portions. CFC-11 is Freon based, according to the report.
From my reading, the switch from CFC to HCFC foam does not appear to be implicated as a safety problem, above and beyond the serious issue of foam shedding itself. I hope that those people, elsewhere in this thread, who are making such accusations, will at least read the CAIB report in full before continuing their claims.
http://caib.nasa.gov/news/report/pdf/vol6/part01.
http://caib.nasa.gov/news/report/pdf/vol6/part06.
http://caib.nasa.gov/news/report/pdf/vol1/full/ca
I said 'the risk of wing impact damage was essentially dismissed' (CAIB Volume I, chapter 6, pg. 140, among other places). You are claiming I said something much broader, ie. that NASA disregarded foam shedding altogether, which I never claimed.
I said "foam shedding was accepted as an inevitable consequence of launch". I was not talking about O-rings when I used that word, and I did not (again) say what you said I did.
I find that you are the one making false claims, both in misquoting me, and in data you use to support your argument. You said, falsely, that "the worst burn-through incidents had been during warm weather launches.", and I posted a link to a chart that refuted this as false. The worst burn-through occured on the coldest launch (two-worst and two-coldest, if you include Challenger's last flight). Furthermore, less burn-through occured, on average, at warmer temperatures, indicating a trend. You ignored this all in your reply.
Here is a different link to that chart (mid-page):
http://www.statview.com/support/techsup/faq/Tufte
If I knew what "party line" you were referring to, I might be offended. I don't expect you (at this point, since you don't seem capable of comprehension) to understand my rebuttal. But stop making a false argument by putting words in my mouth. And if you accuse me of misrepresentations of fact, don't engage in them yourself. If you disagree with me, put me on your enemies list and be done with it.
I said that "the risk of wing impact damage was essentially dismissed." Much of the RCC is on the wing edge, so we are in agreement. NASA having an "ongoing program to reduce foam shedding" is also exactly my point. They were aware that foam shedding caused shuttle damage, and that the shuttle was not originally designed to be hit by this foam. But, in becoming a normal event for launches, they had come to accept (I argue) an unrealistic view of how hazardous the situation was. Yes, they did try to reduce it, but nevertheless it continued to occur, and they didn't fully appreciate the extent of the damage it could cause. CAIB report, Volume I, chapter 6 discusses this in much detail, and I won't go over it here. But look at page 141, in particular, to see an example of your statement, about RCC/wing impact being downplayed.
Again, I argue that "an extensive and ongoing program to analyze and correct the joint problems" was moot, when clearly the problem was that the design was fundamentally flawed (burn-through was not in the spec, and it was occuring.) and needed to be corrected before further launches. They had to be redesigned anyway, at enormous cost (seven lives, a new shuttle, lost schedule, etc.). The decision to postpone launches until the problem was fixed would have been the politically tough, but ultimately correct, decision.
Yes, there was a complicated interaction going on in the Challenger accident. I am not as familiar with that report, so I won't refer to it directly. But E. Tufte, in his book "Visual Explanations", has a now famous assessment of the situation, and in particular, the rather poor presentation of data to support the engineers decision to NOT recommend a launch at such a cold temperature (The only "no launch" recommendation given by Morton-Thiokol up to that point in time). I'll base my comments on the analysis from that book.
You make my point for me: "with no change in evidence, the engineers were changing what they'd been telling management". Even if I accept
what you say, as truth, that engineers had been claiming that the O-rings were safe ("fixed", as you say) even at near freezing temperatures, it points out that when the engineers finally did come to management (both Morton-Thiokol and NASA) to urgently discuss a re-assessment of this opinion, their "no launch" recommedations were not given the proper merit. NASA now had to be convinced that the launch was UNSAFE, rather than accept that anomalous conditions should require launches to be proven SAFE. If management truly was "testy", as you say, then they shouldn't have been making such important launch decisions in that state of mind. People make mistakes in judgement when they are "testy" (in my experience). That accident was as much a failure of people, as it was of equipment (I think we both agree on that).
Tufte points out, in his book, that since the situation became one where the engineers has to prove t
I agree. After years of helping out my family members and friends with their computers, I've learned that "no good deed goes unpunished".
If there is a way to change to a cheaper AOL service, tell your Dad about it. But think *3* times before taking on the responsibility of changing his email patterns (and protecting his data); unless your Dad is pleading with you for the change, it is almost never worth it to get involved. The effort will not be appreciated nearly as much as you hope, and there will be bad feelings if things don't go absolutely 100% correctly.
Good luck.
You say that foam shedding is "expected", and yet that misses the entire point of the post-Columbia investigation results. (And post-Challenger) for that matter.
Foam shedding was NOT expected when the shuttle was designed; the foam came in late in the game apparently. The tiles and wing structures were not designed with the intention of being struck by so much debris at such high speeds, and so initially the foam was seen as a real risk to the shuttle integrity. However, over the course of many launches, as the foam debris strikes piled up, it came to be an "expected" event, and the risk it posed was downplayed. And although you may "expect" a car with failing brakes to not crash everytime you move it, that doesn't mean you should drive it that way.
The mindset that foam was not likely to cause loss of structural integrity, was so strong for NASA shuttle managers that when the Columbia launched for the last time, they did not have a proper way to evaluate the extent of damage from foam. They had lots of data on small foam hits, and this was a BIG foam hit, and at very high speed. But since previous ones hadn't broken the shuttle, there was an "expectation" that even big foam hits probably wouldn't; they really weren't sure. The engineers obviously knew that F=ma could be a large number, even for small m, and they attempted to adapt some tools to calculate the possible damage. But the mindset that "foam is an expected event, it hasn't led to shuttle loss before" was already too well entrenched, and so the risk of wing impact damage was essentially dismissed.
This was made famously clear, when during the post-Columbia loss investigation, Scott Hubbard demanded that the foam gun tests be performed (ie. launching foam at a mock-up of the shuttle wing) in order to convince shuttle flight managers of the risk it posed. At that point, the majority of the investigation team already knew that the foam had been the cause and didn't press for those tests, but Scott knew the NASA culture, and knew they would not accept it unless proven beyond a shadow of a doubt. Against NASA's objections, the tests were performed, and we all saw the results.
It is important to remember that the exact same problem in mindset doomed the Challenger flight. The O-rings were not designed to allow any burn through of the rubber. When it started to occur, it was accepted as an inevitable consequence of launch, rather than a fatal design flaw. Because it had not burned completely through on previous flights, it was accepted that some burn through would happen, and not be catastrophic. NASA management downplayed the risk so much that even when engineers insisted that such a failure was more likely on a cold launch, their objections were not well understood. Burn-through was "expected", so how much of a risk could it be?
There is a lot to be learned about engineering, and management, by these examples. Engineering has been called the art of compromise. However, when so much compromise has been made, it becomes easy for people to not properly evaluate all the consequences. Meanwhile, management has pressures that go outside the engineering realm, and the psychology of that situation can lead to completely unrealistic assessments of risks and liabilities.
For another bit of fun, I just noticed that Google maps now has a map of the moon landing sites available, in celebration of the anniversary of the first landing.
http://moon.google.com/
Try zooming in all the way...
As for exploring other systems that could be life supporting, as a NASA contractor, I hope we a) make it a bigger priority, and b) don't fuck it up by contaminating anything.
I used to think, when I was young, that if we found extraplanetary life, in my lifetime, I would be alive to see an amazing change in orthodoxy, both religious, scientific, and otherwise. Now, both religion (crudely lumping the most popular ones together) and science seem to expect such a discovery, and each has already adapted and prepared for the day when such discovery is made. I'm personally hoping what we find completely defies prediction.
Yeah, he's inheriting from a dictionary type, which is a new-style class introduced in Python 2.2.
If the code author really wants to still support Python 2.1, he would have to inherit from type UserDict, assuming he isn't making use of other 2.2 features. I suggest an email to the author indicating a change in the requirements listing, or a change back to 2.1 compatibility.
FWIW, in the introduction to "Prayers to Broken Stones", Harlan exclaims that it was he who "discovered" Dan Simmons at a writers workshop (and that when he is long forgotten for his own work, he will be remembered for that fact alone). Dan Simmons, of course, went on to write the early Hyperion Cantos books soon afterwards and win the Hugo. Dan acknowledges the discovery in that introduction (although his recounting of the events is somewhat different).
FWIW, I've seen Justice O'connor on television, nehemently denying she ever said that, or anything like that. Maybe she's a liar, maybe the news is reporting BS (no, really?). In any case, the court decision on the 2000 election was total garbage (as was the recent eminent domain one; In both cases I don't mean necessarily that the outcome was crap, but that the stated REASONS for deciding the outcome were crap)
How the heck is a Firewall necessary to keep a default Windows box secure? In other words, if a Windows firewall is there to disallow services (or protocols) from receiving connections from the outside world, then what are these services, and why are they running in the first place?
I understand that by deceiving a user, a malicious service can be started up and listen on the internet, and become a vector for infecting your machine. But that requires an act of the user. If I NEVER enable any special services on my machine, than only the default services are running, and they must somehow be allowing malware to install, right? So, why aren't these services fixed, or disabled by default?
Finally, if these servies are necessary to the proper running of my machine, then when I use them the Windows firewall software will ask if I do not want to block that port, service, etc. Once that occurs, am I not just as unprotected as if I never used the firewall software? How does it really help?
So, that's a lot of questions, but I would appreciate an explanation. Are the attacks on windows solely due to users running malware directly, or are there vectors by which, without any user action (ie. no browsing w/ ActiveX controls, no javascript, no running malicious executables, no starting email attachments, etc) the machine can get infected anyway? If so, what are those services? It's not like a Windows machine, by default, needs to have an email/web/network disk/instant messaging service running, so why does it?
NOTE - I googled "insecure windows services" and got some info; indeed windows does have a bunch of services open to the world by default (un-f'ing believable). Can anyone say which ones are primarily allowing machines to become zombies?
http://www.ss64.com/ntsyntax/services.html
A doctor friend of our family splurged on a Lisa when it first came out (1983, or so). I remember being most amazed when he pressed the big On/Off button, and all his applications saved (and closed their windows one by one) and then it powered down on its own. Clever idea that was a fair bit ahead of its time (resurrected in the 90s, I think, when costs went down).
I see your point, and won't belabor it; my very simple example was meant to illustrate how a crude version of this technique could operate, not to champion it for actual use. There are many implementation details (CPU usage, network meta-data, implementation complexity) that my example doesn't address.
If I send A and A^B to you, I could've just sent A and B instead.
Yes, but if you've sent me only A, I'm of no use to anyone who requests B from me. If you've only sent me A^B, I can be of use to others who already have either A or B.
So, it may be true that the idea isn't worth implementing, due to the realities and complexities of network dynamics. But the math proves that a computed bit (ie. A^B) is worth as least as much as either A or B by itself.
Yes, the same number of bits have to be transmitted, but with this technique, I may be able to get all the bits from more peers, more quickly.
Question: Does this mean that I'll be unable to do anything with incomplete files?
No. Incomplete files will still have the actual bits (not computed bits), but my simple two bit analogy starts to break down as an explanation of why. But let's make it slightly more complex:
I have bits 'a', 'b', and 'c'. I can compute (a XOR b XOR c) and send it. Then I send (a XOR b). Finally I send a.
The receiver gets the first two computed bits, and uses them to compute (a XOR b XOR c XOR a XOR b), from which they get bit 'c', thanks to the magic of XOR.
So, the main thing to understand is that even though bit 'c' was not transmitted, it can be recovered after only two bits are received. Following this same pattern, if we started with many more bits, we could recover 1 actual bit value after two are received, 2 bits after three are received, etc. Well, as long as we receive them in a useful order, we can.
So, using even this simple method, we can start recovering the actual file bits right away, without having to wait for all of the computed bit function values to roll in. The actual method used by the researchers is more complex than this, mainly to allow for the freedom of receiving and recovering the blocks of bits in random order. But it uses a similar idea of computing and solving linear boolean equations for the bits.
A simple example for those reading who don't understand, then some follow up comments:
Say I have bits 'a' and 'b', that other people want.
I could sent bit 'a', then bit 'b' to receiver FOO, who can pass them on to others. However, if I send bit 'a' first, and others want 'b', they have to wait.
Now, instead of transmitting to FOO bit 'a' then bit 'b', I send to FOO ('a' XOR 'b') first, then either bit 'a' or bit 'b'. I'll end up sending FOO the same amount of information (assuming the order is specified in the protocol itself).
BUT, and here's the cool part. If someone already has 'a', they can get ('a' XOR 'b') from you, and complete their set of data (bits 'a' and 'b'). Furthermore, if someone already has 'b', they also get ('a' XOR 'b') from you, and complete their set. So, by only downloading 1 bit, instead of 2, you can complete the set for others who already have one or the other bits.
Now, in practice it'll get a lot more complicated, and the method presented in the paper is not exactly like I describe, but the idea is that you can send data to help people complete their data sets, even though you yourself do not yet have the actual uncomputed data. Instead, you have a computed function of the data, which others can use immediately, and from which you can reconstruct the actual data later when you have more information.
The practical upshot is that the computed data is more valuable to other peers than the uncomputed data, as they may be able to use it to complete their data set, rather than wait for the remainder of the uncomputed data.
So, in reference to your comments, it may not be so much more practical to any one receiver; they still need to wait for all the data, in either computed or uncomputed form. But, for the network as a whole, it means that each receiver has many more options from which to download and compute each chunk, and thus make available to others. It is not hard to imagine that this can benefit the overall throughput of the network (which the authors of the paper claim).
Well, since you mention Numerical Recipes, it is obligatory to post one of the many rebuttals. Basically, the books have some okay discussions (and cover a very WIDE range of subjects) but their code is crap. I say that boldly, since I must maintain code that was originally developed using their C libraries. There have always been better alternatives, and especially these days when so much is available on the web.
r y p osition
http://en.wikipedia.org/wiki/Numerical_Recipes
http://en.wikipedia.org/wiki/GNU_Scientific_Libra
http://en.wikipedia.org/wiki/Linear_least_squares
http://en.wikipedia.org/wiki/Singular_value_decom
http://www.netlib.org/