Totally agree. The real bruha here was that NCSA got the award and CMU/UCSD/PSCC didn't. And worse, that they found out by web-scraping the NSB agenda rather than having the program manager call 'em and tell 'em.
Teragrid and petascale computing have always been closed games, with the price of admission being to either fund yourself (have your home institution) into a position to be recognized, or, when NSF suddenly realizes they're not looking to diversify the playing field but are always funding the same few suspects, and decrees that existing players won't get a new (small) award.
I'm less surprised that NCSA got the nod than I would have been if, say, San Diego State had won, or the University of Texas. Both are credible computing players, but neither was mentioned (and to be honest, I don't know if either was a proposed principal in a proposal, or if they simply played along).
As it was put to me once, they're not looking for the BEST in their field, but they're looking for GOOD in discipline, who can work and play well with their peers. When you put 7 folks into a volume the size of a VW van, and know that about half of them will be sick (statistically) thins get tough pretty fast. Crews work a nominal 8 hour shift but really are doing something for their entire crew-wake period. Everything's pretty scripted in the timeline. If someone's sick, someone else has to do their job in addition to their own.
If you get folks who can work well together (and that usually entails playing well, too) you see this cooperation. When you get folks who fail to do their part, whether sick or feeling good in microgravity, morale falls off and problems develop.
The astronauts assigned to the selection board have a lot of responsibility to select folks who will get along well, and these are the Office representatives who have to make what amounts o a snap decision. You occasionally see someone who's never assigned a flight, or more frequently, someone who got a flight pretty quickly in the rotation but then left the Astronaut Corps and the Agency shortly thereafter. These are the ones who didn't do well. They were either counselled to think about another job or frankly let go. Note that both of the recent targets of bad publicity were reassigned (and left the Agency) pretty quickly. Regardless of the spin NASA PAO places on these announcements, they were gone because they didn't work and play well with their peers.
Obviously you've not kept up on the latest memoranda. A Flight Readiness Review is a management and Operations team review certifying that all the paperwork and preparations have been appropriately completed for the mission. By the time you get to the FRR, you know what the result will be... you don't schedule an FRR until you know it's a foregone conclusion.
Richards would have to get past the t-21day flight physical, the t-14 day physical and the launch day medical check.
I'm not a military pilot (just a spam-can jockey) but I did take the training for NASA Flight Surgeon, and I spent 7 years at Johnson Space Center.
I'm signing on here to agree with our Pilot. The pilots and commanders are well-trained, and prepared for anything they're likely to encounter. And, taking note of a GNC puke below, I have heard the same response to the 3-3ngine abort scenario. I wonder if we worked there at the same time. Remember the Russian Shuttleski? They did an auto-land, and the cosmonauts were less than pleased at not getting to do anything but go along for the ride. Don't blame them.
While I was involved in crew training for a couple of experiments, I worked with a lot of crews. I worked with them enough that I often was invited to go "out for a beer" at the end of the workday. Most of the time, the "beer" tasted a lot like iced tea of Coca Cola, and there was just enough time to let them wind down so they could drive home and not overload their families. One particular exception stands out: He had 3 kids in Scouts and was out on-time every evening, without fail. He had to take care of his kids and did a stellar job of it.
Did anyone mention that the typical training profile for the Commander, Pilot and Mission Specialist 3 (flight crew) is typically a 13 month period from designation/selection for flight, of 13 hour days, six days a week? I agree with Dr. John Clarke. I was, and remain, much more concerned about sleep deprivation and overwork than whether they had one beer or two at the barbequeue, or a beer in crew quarters. (For what it's worth, while there are a lot of creature comforts in crew quarters, it's still pretty sparse and the variety of company is pretty slim. Maybe a beer isn't such a bad idea, after all.)
I'm not discounting the possibility that there are a couple of questionable characters. I think we've seen the result of what appears to be a change in the selection process for the worse... or, a failure of the peer process within the Astronaut Office to handle their own problems. I know that's how it used to work. And I know that it did work.
So that's the reason, when I get resumes from graduate students seeking on-campus work, in technical topics, in a well-known pure open source environment, the resumes always tell me how great they are with Microsoft products? You'd think they'd learn. I've turned down more than one (OK, more than 10) who were trying to tell me how good they are with computers by emphasizing their ability to handle Office?
> Unless the class is "how to use office 2007" and an elective, they have NO right to dictate this, remember they work for you, not the other way around.. ( even if you can get educational versions for 25 bucks )
Interestingly, two of my three kids have had to take State-mandated "computer literacy" classes, where they had to demonstrate proficiency in Excel and Word, and my daughter also took a"graphics" class where they learned to tweak images with Paint and Photoshop. Needless to say, we no longer treat as accurate any pictures she forwards our way...
I have OpenOffice installed on the computer at home, and after getting past some set-up hiccups, no one has ever had homework ("Must be written in Microsoft Word") refused, nor have they lost points. In one case where they were told to turn their work in as a PDF, my son was able to export directly... and without us buying Acrobat as he'd been told he had to do.
That said, my wife just bought Office 2007 because she got it for a steal -- and legally. She is afraid I'll ruin the middle kid's chances for good grades in his senior year because of my intransigience. Go figure.
I'm building up a new system for the 4th grader. It'll have to have a Windoze partition for some of his games, but he's gonna grow up with open source solutions as his norm, not the exception.
It's been a couple of years since we built our first multi-terabyte storate (~1.6TB for ~$5k US) and started the grand experiment with XFS. Of all we've learned, I think the biggest lesson has been that XFS has problems in this realm. We are now spinning over 30TB and likely to double soon. To SAN or not to SAN has become the question, as finding our data and metadata are both important to us, more important than slinging the files off and accidentally discovering them later.
Some form of indexing system is valuable, especially if you're looking at multiple volumes to span.
We're now looking at lustre and possibly zfs to support our solutions. For hardware, we've gone to ATA-over-Ethernet with CoRAID hardware and been pretty satisfied. It's not iSCSI but it works well and we get adequate transfer rates. We do a caching process when we're anticipating high user access periods and can predict the patterns.
I'd say, for the money, try it, benchmark it, and report back.
The various Private Sector "Partners" do track their forecast accuracies but mostly to see what they can improve in their forecasts. I know the Weather Channel runs its own versions of at least 2 models and does verificaiton. I expect AccuWeather does as well (well, OK, I know they do), as do WeatherBug. It's not likely they'll show their verification stats, though. As few folks really understand statistics, showing something that degrades toward a coin toss after a couple of days (not really, but I'm making a point here, OK?) isn't likely to show a competetive advantage.
I have seen one ad to broadcasters claiming one vendor can outperform by a factor of 4 on tornado predictions. In some cases, they're claiming they can predict a mesocyclonic event before the mesocyclone complex exists, no mean feat. Maybe they're working on the Storm Prediction Center's (http://spc.noaa.gov) predictive tools?
Not very accurate. In the US, a well-tuned numerical prediction is skillful for about 72 hours. At best. Anomalies happen. The forecasts I did, for example, at the time of Hurricane Katrina, were pretty decent. Those I did when Hurricane Rita came through were 'way off on track... but then, so were those from the National Hurricane Center. Model verification and tweaking are the fodder of PhD projects and dissertations.
For what it's worth, the ECMWF model tends to be good for about another 24 hrs beyond what NCEP's NAM (WRF) is good for.
Radar is but one tool in the toolchest we all use in numerical prediction. Satellite data are useful, as are in-situ observations. In fact, the US National Centers for Environmental Prediction are incorporating these data into the models (google for 3dVAR and data assimilation).
For a given isolated point (your house/apartment) and given sufficient experience (you've lived there for a couple of years watching the weather as an avocation) it's probable that you can predict for that site better than the National Weather Service (NWS) can for your area. You fail to mention how far you are from the local Weather Forecast Office (WFO).
Each WFO is responsible for updating the National Digital Forecast Database (NDFD) grids in its County Warning Area. The forecaster of the shift is responsible for writing a coherent Forecast Discussion that explains the general state of the atmosphere and the assumptions and rationale that went into that shift's forecast. She is also supposed to update the FD when there's a change to the forecast, but pop-up convective activity can cause that to be a little, well, delayed if there are severe storms that snuck in.
The concept of a bunch of small weather stations isn't new. Please look at the Oklahoma Mesonet as a shining example of a plethora of in-situ sites used to improve forecasts. Similarly, the use of consumer grade stations, with some initial attempts at quality assessment is the basis for CWOP (http://www.wxqa.com). A lot's going into how we get data into NWS and how they use it.
Note: I'm not an NWS or NOAA employee. I am familiar with mesoscale networks, their hardware and operation. I run numerical models of the atmosphere in prediction mode on a daily basis and look at the results.
Isn't this a little extreme? I mean, I'm the vict... er... product of the US education system (please attribute most of typo's tonight to the working conditions) and A) I am conversant with and routinely use the metric system; B) I've worked in the US Space Program... using metric units on a regular basis and no one flinched, or had any troubles understanding what I was saying; C) seen NASA actually send Shuttle Burn PADs up in metric units, had them entered manually, and they worked, AND we had a clue of the orbiter's attitude and delta-v from what we sent up.
That said, I've got a 22 year old daughter who's had as good a preparation in math in high school as I could have asked from public schools, who now hates it, and a son in high school who's pulling "A's" in math and science and cannot understand an inverse, or do simple algebra, and is passing honors English listening to book tapes (with the teacher's recommendation) and writing canned reviews, so I guess there might be something wrong with the system. That said, the 9 yr old is getting a bit more than the norm because he's inquisitive and still thinks Mom and Dad can possibly have a clue left. He's learning medicine from both of us, math and physics from me, biology from her, hands-on carpentry from me, (yes, I have Imperial units on most of my tapes and squares, but I'm proud of that speed square with both Imperial and SI units stamped on there!), networking from me, and general life lessons beyond the abyssmal level of "standard" imposed by Texas and blessed by the Feds.
But WAIT! We've stated we'll leave no child behind, so everyone's being tested to prove they meet minimum standards. Right? As a result, the last month of classes (or more) is dead time, as the kids have taken their standardized tests and the official curricula don't allow free-lancing that might really offer something that might not be on the test (or that might contradict an officially sanctioned fallicy), so they turn their books in with 2 weeks to go, go on "field trips" with more social benefit than academic content, have early release days, and don't get the benefit of the teachers who are still there who have the snap to teach.
Yeah, we need to strengthen the basic American education system in virtually all aspects. I'd like to see the kids learn how to write (has anyone else noticed how they're teaching kids to hold a pen or pencil today, and how it limits range of motion, making precise work almost impossible?), compose a sentence... and then a paragraph, spell without the benefit of a spell-checking program most of the time, and manage grammatically correct reports and papers. I'd like them to learn to read (online is OK!) rather than thinking they can get all their information from a buddy who didn't learn to read, either, and then be able to paraphrase the answer coherently.
And I'd like them to be able to do unit conversions better than me, but that's not too likely to happen.
Having spent a little time in the cardiovascular research and therapeutics world, I have to agree. The human fuzzy-thinking seat-of-the-pants gut-instinct doctors will... or at least SHOULD be kept around to teach the ones who think they can diagnois everything by lab, x-ray, CT and MRI, while shunning the idea of doing a decent history and physical. Sorry, guys, there are things I can discern with a decent stethoscope and a bit of clinical acumen that the labs, CTs and ultrasounds will remain equivocal on.
Over the years, I've seen my writing style evolve. I agree that the neophyte writer will tend to try to make their written communications more stilted, in an attempt to sound authoritative. We see the same thing in PhD disertations: These folks are trying to present the concept that they're the world expert in something, and they have to sound like it.
What I've found my style evolving toward is a more conversational, and more first-person presentation that still presents the facts. I've also found my style, as I've become more comfortable with teaching evolve to this, where I can speak with comfort about my experience with a particular concept in GPS, geodesy, mathematics or networking and can relate that to the student.
I believe clear organization is a key. I'd focus on the rather lost art of outlining, at least to section headings, and the idea that a paragraph really does require three sentences. I agree, generally, with the idea that the newspaper reporter's format ("Who, what, why, when, where, how?") offers a plan for clarity that would improve most engineering prose I've encountered.
But, more importantly, getting the student to know when they've reached two points, and optimized those two information transfer functions would serve to improve most of the engineering documentation I've read: Know when you've said enough to convey a minimum of information to establish the required message, and then, know when to stop talking (writing) because you're now relating drivel.
Another important concept to establish is when to start sharing your writing with colleagues, and ask them to begin the editing process to help you. I've learned two things, applicable to me specifically, about this: First, I can only do about three drafts before someone else must read it and help me determine if I'm conveying my message, or if I'm simply babbling. Second, I'm not asking my colleagues to rewrite it for me (although there are at least two who will always try). I'm writing in my style, not theirs, and I get more papers published than they do, so I must be doing something right. And, it's not THEIR paper, it's yours, so unless you're offering associated authorship (and you should only do that if they have something concrete to offer in the subject, and not simply for editorial assistance) you cannot depend on, or expect them to do anything.
Military GPS is accurate to about 1.2-1.5 meters. Civil GPS without augmentation is typically getting ~6-10 meters.... usually closer to 6 m.
Galileo starts at about 10 meters... maybe 15 meters for public data and you can achieve 1-3 cm with proprietary data.
Land surveyors and geodetic researchers get 1-2 cm horizontal and 3cm vertical using GPS. Today. Requires hardware that is designed for collecting the data on both L1 and L2 frequencies, and software that allows you to postprocess it.
1-5 cm accuracies for surveying are achieved using yet another form of a DGPS system called real-time surveying or real-time kinematic. Requires a dual-frequency system linked to and continuously updating the single-frequency rover. The rover must be designed and have the software to handle RT operations. It's *not* RTCM-SC-104 DGPS.
There's a fair bit of FUD in this whole line of commentary.
The only times I've heard of this, and been able to document the result, DoD was diddling clocks on a couple of satellites to determine what'd happen with the 1024 week rollover (an inherent problem in GPS caused by its age and limitations of its design spec). They filed the appropriate notices, which are sent via the Almananc message, and which all receivers are supposed to read and act upon. A bunch of civil receivers couldn't handle this, and reported really bad results. This wasn't a GPS-system problem, it was a ground-component problem associated with commercially-acquired hardware not following the signal spec.
There have been similar tests. My $5k receivers don't notice anything, and my old Garmin GPS-55AVD doesn't have problems. My (slightly) newer GPS-IIPlus does see errors and demands to be power-cycled but then behaves appropriately. I've seen real problems with a Magellan receiver. My OEM-board SiRF and Motorola receivers do not have problems.
GPS can provide centimeter-level accuracy. Consumer-grade receivers are not built to take advantage... collect sufficient data, nor emit the appropriate corrections, to accomplish this. For that matter, the receivers necessary for cm-level accuracy with Galileo systems won't be cheap, either. TANSTAAFL.
Galileo's business model ("free", "costly", and "Safety of Life/free") is an interesting one. With GPS, the "costly" model requires paying solely the hardware and software vendors (I use OSS for my data reduction, thank you very much). With Galileo, you pay the hardware vendors, software vendors and the EU.
To the best of my knowledge, there has been no widespread, intentional system degradation since 2 MAY 2000, when Selective Availability was turned off. I suspect there is conflict-theatre degradation, but I've no hard evidence of that. Solely inferential.
The MEO/56deg orbit of the Galileo birds will *not* provide significant changes in polar coverage over GPS. Sorry. Dirt blocks RF.
The changes in signal structure, and improvements in power budget will help getting signal into buildings. Some. Multipath will still cause problems. Higher power levels will increase the effects of multipath.
Civil code-phase GPS accuracies are on the order of 6m 2d 1-sigma, and 7.5m 1d (vertical). Carrier-phase dual-frequency determinations yield 1.2-1.5cm, 2d 2-sigma accuracies, and 3cm 1d (vertical) 2-sigma accuracies with sufficient data collected using good reference sites. Military systems using P-code or P/Y, tend to achieve L1/L2 code-phase positions on the order of 1m 2d, 2-sigma. These are based on some documentation for the military systems, and experimental data for civil receivers/systems.
The US Government has mandated the removal of SA, and identified GPS as a safety-of-life system that cannot be degraded save for dire emergencies. It wasn't degraded for 9/11 (urban legend), Afghanastan, Iraq (either time, and in fact in the 1991/1992 timeframe, SA was turned off because of the massive number of civil receivers in use by the troops in theatre). Too many folks depend on it.
Augmentation via Differential GPS (DGSP), WAAS, LAAS, or real-time kinematics accomplishes about the same degree of correction and enhancement, save DGPS is more accurate closer to the DGPS station. I usually estimate a 10x improvement as a rule of thumb when asked.
In the early announcements on Galileo in the navigation community, the program was very much a "we don't trust the US and especially not their DoD, so we're going to build our own" issue. The majority of folks involved in the discussions from the US side, and I know most of them, made suggestions that would reduce potential interference, and enhance the potential for a separate system with redundant capabilities while built using different approaches to the problem. Some of the folks involved were not particularly cooperative, but they were in the minority.
I'm still not sure this is a marked improvement over the GPS system, but having another choice is good. Personally, I think it'd have been cheaper to have invested in GLONAS and used Euros to drive the Russians to improve their system. There were some nice features about GLONAS, including the fact that SA couldn't be invoked -- by design.
There will be a short quiz at the end of the next lecture.
Well... I know a guy, in the US Army, working toward his PhD, using Unreal Tournament to help model crowd dynamics for his dissertation. The end-of-the-day test and eval sessions were a hoot!
Freedom was foreseen as primarily a US venture for launch and support, as already indicated. Ascent from French Guiana was possible (they were our friends, right?) without too much trouble in a 28.5 deg inclined orbit, but it took a fair bit of delta-V to get from Baikanor to 28.5 deg. Of course, that wasn't our problem as the Soviets were on the other side.
OK, international politics aside.
One of the real problems we saw was the US Congress, and yes, NASA management.Space Station Freedom was often a dumping ground for "retired in place" senior engineering management waiting for that magic day when they could sit at home and impede their wives instead of coming to the office and impeding engineers. That's not to say we didn't have decent, enthusiastic, qualified management but they were outnumbered... or simply out-numbed... by the incompetents.
A lot was preordained, despite engineering advances. "Don't try to convince me, my mind is made up." I could go on at length about the decision to scrap the 100 khz power distribution system on Freedom in favor of the DC system. I was around when the "test" destroyed some computer hardware at MSFC that was used as justification, despite the fact that the test was protested by competent engineers with a knowledge of VAX power supply design. Were there problems with the high frequency AC distribution? Some, but not insurmountable.
SSF was also a training ground for kids right out of college. Get them in, turn 'em loose with little guidance, slap 'em around a bit until they started doing good design, then move them to Shuttle.
We had a lot of design by Aerospace Conglomerate, too. Let's get that design that Lockheed wants, because it'll make them easier to deal with at contract time. Let's use THIS design that MD wants, even if it's not what NASA wants/requires, because we think their design is going to make them do something else for us on another project.
Still, and all, most of the conglomerate designs I saw, worked with, and helped shape (and, yes, I worked for a contractor company, too, but I was doing specs and requirements, as well as working with the prototyping) would have been acceptable, even if somewhat limiting in their own ways.
The BIG problem, however, was Congress. Every three years or so, we'd get a "stop what you're doing, reassess the design, and then start over" command from the Hill. I've gotta say, we wasted a LOT of money on those exercises, and we wasted a LOT of time.
There are improvements borne of waiting time and engineering advances in ISS that would not have been, and may never have reached SSF or Alpha, but we could well have bent metal and flown hardware by 1990 if Congress had stuck to original budgets and timelines and stayed the hell out of the way. I flew prototype hardware in 1992 that was the first piece of Space Station hardware to fly, be proven and certified for on-orbit Space Station operation. I could have flown it 3 years earlier save the Challenger accident.
Final thought. We developed or promoted a lot of stuff that's now common place in the world. Speaking from the perspective of medical hardware development (I also did a bit for the medical facility in terms of GNCC and COMMS) there's a lot of stuff I see in hospitals, doctors' offices, dentists' offices and ambulances that makes me smile and think, "I worked with the prototype of that...", or, in a couple of cases, "I wrote the SBIR paperwork that made that happen".
So, yes, NASA's efforts HAVE improved life ont he planet. Really.
BPL isn't just an HF-impinging system. There are VHF, UHF and microwave signal injectors. Of these, most (although the microwave set appears to be an exception) are pretty significant emitters when energized. And pretty bad at tolerating "competing" signal impinging on the spectrum on which BPL tries to use.
Local RF emmissions, in licensed operations, near a BPL-enabled powerline are likely to cause problems if there's an overlap.
Let's address your use of satellite communications for emergency comms: As long as the load is controlled, this will probably work. What has been demonstrated time, and time again, is that local cellphone and trunked networks seem to overload pretty quickly in the local area or even region of the disaster/event if that network is used by, say, the media and local emergency managers. VHF/UHF backups using Amateur Radio have shown theur worth in those settings over and over.
Similarly, the HF networks have been useful to disseminate data from areas which have lost conventional telecommunications due to infrastructure damage or eradication. Look at large areas of Florida just last year... or Sumatra in December. The amount of data transferred out of Sumatra by Amateur Radio was pretty significant.
I've kept my ham license up. It's an interesting tool to use for RF experimentation. Did I mention the linking I've been doing with 802.11b/g hardware using the 6 channels Ham Radio happens to be primary on? 2-10 Mb/s for data and voice using conventional data methods, but in the ham commuity, is interesting. And, I've been playing with 802.16 in a similar manner. I've had a leg up on a lot of folks who don't have my RF background, but did sit through a propagation course in EE in college. I've done it, (I've studied it, too) so it's not magic or theory to me.
I'm only gonna take exception to the comment regarding the training and education of the air personalities on the Weather Channel. They *ARE* professionals and know their stuff. And they're less likely to be strong proponents of this bill than the special interest (AccuWeather) that pushed for it. They have real atmospheric scientists. Where else do you get discussion of severe weather from a PhD who's well published, and really interested, in it? WGN? I think not!
And I'm assuming you're using the http://mesonet.agron.iastate.edu/ Iowa Environmental Mesonet for your imagery? They do a good job for real-time imagery... And they're interested in data from Amateur Radio and trained spotters, too.
Learn to read FAA Forecasts? Just where do you think they come from? The FAA doesn't run models, nor do they have weather stations. Once they dissetablished their airport weather observers and ASOS was the norm, they ceased to employ anyone in observation. The few remaining briefers at the Automated Flight Service Stations have historical basis for their knowledge and the ability to interpret the data, but most are not trained meteorologists. Many's the time, in getting a "standard briefing", I've heard something along the lines of, "A warm moist southerly flow predominates. Ceilings are at or above 10,000 feet. The possibility for afternoon thunderstorms exists. VFR flight is not recommended." Problem here? Sure. There's a strong cap and the potential for those thunderstorms in miniscule. There's inadequate moisture in the region to support convection, the cloud deck is widely scattered at 10k, with a few puffy Cu, and winds are negligible. If you fly... does the term CAVU mean anything?
This bill is yet another attempt to drive NWS away from "competetion" with the privates. In my work, I've seen the privates tell us they can be more mobile and responsive to the Public needs. AWS likes to point to the Baltimore ferry accident, claiming they had an indication of the gust-front that caused the accident from their network of stations well before the NWS's more sparse network of official stations saw an indication. The result? A cooperative agreement between a private sector firm and NWS to provide NWS with that data in the case of such an event, allowing/providing NWS the data to make the appropriate warnings. Doesn't preclude AWS from offering their own warnings and value added services, but it IS a case where they can help the public rather than trying to hamstring the Weather Service.
This is *exactly* what I've been thinking of for a weather station. Add the processor to the sensor, place one on the anemometer at 10m, another on the temp sensor at 10m, but have it tightly networked to the 2m temp/humidity sensor. Presto! intelligent heat flux calculations. Tie the 10m anemometry to the 2m wind speed, voila`, 3d wind data.
Lose a sensor, no problem. The rest of the site's up. Lose a data collector? No problem. It's the same as losing a sensor.
Slashdot Mirror
Totally agree. The real bruha here was that NCSA got the award and CMU/UCSD/PSCC didn't. And worse, that they found out by web-scraping the NSB agenda rather than having the program manager call 'em and tell 'em.
Teragrid and petascale computing have always been closed games, with the price of admission being to either fund yourself (have your home institution) into a position to be recognized, or, when NSF suddenly realizes they're not looking to diversify the playing field but are always funding the same few suspects, and decrees that existing players won't get a new (small) award.
I'm less surprised that NCSA got the nod than I would have been if, say, San Diego State had won, or the University of Texas. Both are credible computing players, but neither was mentioned (and to be honest, I don't know if either was a proposed principal in a proposal, or if they simply played along).
As it was put to me once, they're not looking for the BEST in their field, but they're looking for GOOD in discipline, who can work and play well with their peers. When you put 7 folks into a volume the size of a VW van, and know that about half of them will be sick (statistically) thins get tough pretty fast. Crews work a nominal 8 hour shift but really are doing something for their entire crew-wake period. Everything's pretty scripted in the timeline. If someone's sick, someone else has to do their job in addition to their own.
If you get folks who can work well together (and that usually entails playing well, too) you see this cooperation. When you get folks who fail to do their part, whether sick or feeling good in microgravity, morale falls off and problems develop.
The astronauts assigned to the selection board have a lot of responsibility to select folks who will get along well, and these are the Office representatives who have to make what amounts o a snap decision. You occasionally see someone who's never assigned a flight, or more frequently, someone who got a flight pretty quickly in the rotation but then left the Astronaut Corps and the Agency shortly thereafter. These are the ones who didn't do well. They were either counselled to think about another job or frankly let go. Note that both of the recent targets of bad publicity were reassigned (and left the Agency) pretty quickly. Regardless of the spin NASA PAO places on these announcements, they were gone because they didn't work and play well with their peers.
Or as we used to say (in the 1990's), "It's the best manned spacecraft money can buy... using 1970's technology."
Obviously you've not kept up on the latest memoranda. A Flight Readiness Review is a management and Operations team review certifying that all the paperwork and preparations have been appropriately completed for the mission. By the time you get to the FRR, you know what the result will be... you don't schedule an FRR until you know it's a foregone conclusion.
Richards would have to get past the t-21day flight physical, the t-14 day physical and the launch day medical check.
I'm not a military pilot (just a spam-can jockey) but I did take the training for NASA Flight Surgeon, and I spent 7 years at Johnson Space Center.
I'm signing on here to agree with our Pilot. The pilots and commanders are well-trained, and prepared for anything they're likely to encounter. And, taking note of a GNC puke below, I have heard the same response to the 3-3ngine abort scenario. I wonder if we worked there at the same time. Remember the Russian Shuttleski? They did an auto-land, and the cosmonauts were less than pleased at not getting to do anything but go along for the ride. Don't blame them.
While I was involved in crew training for a couple of experiments, I worked with a lot of crews. I worked with them enough that I often was invited to go "out for a beer" at the end of the workday. Most of the time, the "beer" tasted a lot like iced tea of Coca Cola, and there was just enough time to let them wind down so they could drive home and not overload their families. One particular exception stands out: He had 3 kids in Scouts and was out on-time every evening, without fail. He had to take care of his kids and did a stellar job of it.
Did anyone mention that the typical training profile for the Commander, Pilot and Mission Specialist 3 (flight crew) is typically a 13 month period from designation/selection for flight, of 13 hour days, six days a week? I agree with Dr. John Clarke. I was, and remain, much more concerned about sleep deprivation and overwork than whether they had one beer or two at the barbequeue, or a beer in crew quarters. (For what it's worth, while there are a lot of creature comforts in crew quarters, it's still pretty sparse and the variety of company is pretty slim. Maybe a beer isn't such a bad idea, after all.)
I'm not discounting the possibility that there are a couple of questionable characters. I think we've seen the result of what appears to be a change in the selection process for the worse... or, a failure of the peer process within the Astronaut Office to handle their own problems. I know that's how it used to work. And I know that it did work.
So that's the reason, when I get resumes from graduate students seeking on-campus work, in technical topics, in a well-known pure open source environment, the resumes always tell me how great they are with Microsoft products? You'd think they'd learn. I've turned down more than one (OK, more than 10) who were trying to tell me how good they are with computers by emphasizing their ability to handle Office?
> Unless the class is "how to use office 2007" and an elective, they have NO right to dictate this, remember they work for you, not the other way around.. ( even if you can get educational versions for 25 bucks )
Interestingly, two of my three kids have had to take State-mandated "computer literacy" classes, where they had to demonstrate proficiency in Excel and Word, and my daughter also took a"graphics" class where they learned to tweak images with Paint and Photoshop. Needless to say, we no longer treat as accurate any pictures she forwards our way...
I have OpenOffice installed on the computer at home, and after getting past some set-up hiccups, no one has ever had homework ("Must be written in Microsoft Word") refused, nor have they lost points. In one case where they were told to turn their work in as a PDF, my son was able to export directly... and without us buying Acrobat as he'd been told he had to do.
That said, my wife just bought Office 2007 because she got it for a steal -- and legally. She is afraid I'll ruin the middle kid's chances for good grades in his senior year because of my intransigience. Go figure.
I'm building up a new system for the 4th grader. It'll have to have a Windoze partition for some of his games, but he's gonna grow up with open source solutions as his norm, not the exception.
It's been a couple of years since we built our first multi-terabyte storate (~1.6TB for ~$5k US) and started the grand experiment with XFS. Of all we've learned, I think the biggest lesson has been that XFS has problems in this realm. We are now spinning over 30TB and likely to double soon. To SAN or not to SAN has become the question, as finding our data and metadata are both important to us, more important than slinging the files off and accidentally discovering them later.
Some form of indexing system is valuable, especially if you're looking at multiple volumes to span.
We're now looking at lustre and possibly zfs to support our solutions. For hardware, we've gone to ATA-over-Ethernet with CoRAID hardware and been pretty satisfied. It's not iSCSI but it works well and we get adequate transfer rates. We do a caching process when we're anticipating high user access periods and can predict the patterns.
I'd say, for the money, try it, benchmark it, and report back.
The various Private Sector "Partners" do track their forecast accuracies but mostly to see what they can improve in their forecasts. I know the Weather Channel runs its own versions of at least 2 models and does verificaiton. I expect AccuWeather does as well (well, OK, I know they do), as do WeatherBug. It's not likely they'll show their verification stats, though. As few folks really understand statistics, showing something that degrades toward a coin toss after a couple of days (not really, but I'm making a point here, OK?) isn't likely to show a competetive advantage.
I have seen one ad to broadcasters claiming one vendor can outperform by a factor of 4 on tornado predictions. In some cases, they're claiming they can predict a mesocyclonic event before the mesocyclone complex exists, no mean feat. Maybe they're working on the Storm Prediction Center's (http://spc.noaa.gov) predictive tools?
Not very accurate. In the US, a well-tuned numerical prediction is skillful for about 72 hours. At best. Anomalies happen. The forecasts I did, for example, at the time of Hurricane Katrina, were pretty decent. Those I did when Hurricane Rita came through were 'way off on track... but then, so were those from the National Hurricane Center. Model verification and tweaking are the fodder of PhD projects and dissertations.
For what it's worth, the ECMWF model tends to be good for about another 24 hrs beyond what NCEP's NAM (WRF) is good for.
Radar is but one tool in the toolchest we all use in numerical prediction. Satellite data are useful, as are in-situ observations. In fact, the US National Centers for Environmental Prediction are incorporating these data into the models (google for 3dVAR and data assimilation).
For a given isolated point (your house/apartment) and given sufficient experience (you've lived there for a couple of years watching the weather as an avocation) it's probable that you can predict for that site better than the National Weather Service (NWS) can for your area. You fail to mention how far you are from the local Weather Forecast Office (WFO).
Each WFO is responsible for updating the National Digital Forecast Database (NDFD) grids in its County Warning Area. The forecaster of the shift is responsible for writing a coherent Forecast Discussion that explains the general state of the atmosphere and the assumptions and rationale that went into that shift's forecast. She is also supposed to update the FD when there's a change to the forecast, but pop-up convective activity can cause that to be a little, well, delayed if there are severe storms that snuck in.
The concept of a bunch of small weather stations isn't new. Please look at the Oklahoma Mesonet as a shining example of a plethora of in-situ sites used to improve forecasts. Similarly, the use of consumer grade stations, with some initial attempts at quality assessment is the basis for CWOP (http://www.wxqa.com). A lot's going into how we get data into NWS and how they use it.
Note: I'm not an NWS or NOAA employee. I am familiar with mesoscale networks, their hardware and operation. I run numerical models of the atmosphere in prediction mode on a daily basis and look at the results.
Isn't this a little extreme? I mean, I'm the vict... er... product of the US education system (please attribute most of typo's tonight to the working conditions) and A) I am conversant with and routinely use the metric system; B) I've worked in the US Space Program... using metric units on a regular basis and no one flinched, or had any troubles understanding what I was saying; C) seen NASA actually send Shuttle Burn PADs up in metric units, had them entered manually, and they worked, AND we had a clue of the orbiter's attitude and delta-v from what we sent up.
That said, I've got a 22 year old daughter who's had as good a preparation in math in high school as I could have asked from public schools, who now hates it, and a son in high school who's pulling "A's" in math and science and cannot understand an inverse, or do simple algebra, and is passing honors English listening to book tapes (with the teacher's recommendation) and writing canned reviews, so I guess there might be something wrong with the system. That said, the 9 yr old is getting a bit more than the norm because he's inquisitive and still thinks Mom and Dad can possibly have a clue left. He's learning medicine from both of us, math and physics from me, biology from her, hands-on carpentry from me, (yes, I have Imperial units on most of my tapes and squares, but I'm proud of that speed square with both Imperial and SI units stamped on there!), networking from me, and general life lessons beyond the abyssmal level of "standard" imposed by Texas and blessed by the Feds.
But WAIT! We've stated we'll leave no child behind, so everyone's being tested to prove they meet minimum standards. Right? As a result, the last month of classes (or more) is dead time, as the kids have taken their standardized tests and the official curricula don't allow free-lancing that might really offer something that might not be on the test (or that might contradict an officially sanctioned fallicy), so they turn their books in with 2 weeks to go, go on "field trips" with more social benefit than academic content, have early release days, and don't get the benefit of the teachers who are still there who have the snap to teach.
Yeah, we need to strengthen the basic American education system in virtually all aspects. I'd like to see the kids learn how to write (has anyone else noticed how they're teaching kids to hold a pen or pencil today, and how it limits range of motion, making precise work almost impossible?), compose a sentence... and then a paragraph, spell without the benefit of a spell-checking program most of the time, and manage grammatically correct reports and papers. I'd like them to learn to read (online is OK!) rather than thinking they can get all their information from a buddy who didn't learn to read, either, and then be able to paraphrase the answer coherently.
And I'd like them to be able to do unit conversions better than me, but that's not too likely to happen.
Having spent a little time in the cardiovascular research and therapeutics world, I have to agree. The human fuzzy-thinking seat-of-the-pants gut-instinct doctors will... or at least SHOULD be kept around to teach the ones who think they can diagnois everything by lab, x-ray, CT and MRI, while shunning the idea of doing a decent history and physical. Sorry, guys, there are things I can discern with a decent stethoscope and a bit of clinical acumen that the labs, CTs and ultrasounds will remain equivocal on.
Er... you have a typo. It's really supposed to read,
"after all, _if_ they *HAD* a product then why should they be worried about Google?
Over the years, I've seen my writing style evolve. I agree that the neophyte writer will tend to try to make their written communications more stilted, in an attempt to sound authoritative. We see the same thing in PhD disertations: These folks are trying to present the concept that they're the world expert in something, and they have to sound like it.
What I've found my style evolving toward is a more conversational, and more first-person presentation that still presents the facts. I've also found my style, as I've become more comfortable with teaching evolve to this, where I can speak with comfort about my experience with a particular concept in GPS, geodesy, mathematics or networking and can relate that to the student.
I believe clear organization is a key. I'd focus on the rather lost art of outlining, at least to section headings, and the idea that a paragraph really does require three sentences. I agree, generally, with the idea that the newspaper reporter's format ("Who, what, why, when, where, how?") offers a plan for clarity that would improve most engineering prose I've encountered.
But, more importantly, getting the student to know when they've reached two points, and optimized those two information transfer functions would serve to improve most of the engineering documentation I've read: Know when you've said enough to convey a minimum of information to establish the required message, and then, know when to stop talking (writing) because you're now relating drivel.
Another important concept to establish is when to start sharing your writing with colleagues, and ask them to begin the editing process to help you. I've learned two things, applicable to me specifically, about this:
First, I can only do about three drafts before someone else must read it and help me determine if I'm conveying my message, or if I'm simply babbling.
Second, I'm not asking my colleagues to rewrite it for me (although there are at least two who will always try). I'm writing in my style, not theirs, and I get more papers published than they do, so I must be doing something right. And, it's not THEIR paper, it's yours, so unless you're offering associated authorship (and you should only do that if they have something concrete to offer in the subject, and not simply for editorial assistance) you cannot depend on, or expect them to do anything.
Military GPS is accurate to about 1.2-1.5 meters. Civil GPS without augmentation is typically getting ~6-10 meters.... usually closer to 6 m.
Galileo starts at about 10 meters... maybe 15 meters for public data and you can achieve 1-3 cm with proprietary data.
Land surveyors and geodetic researchers get 1-2 cm horizontal and 3cm vertical using GPS. Today. Requires hardware that is designed for collecting the data on both L1 and L2 frequencies, and software that allows you to postprocess it.
1-5 cm accuracies for surveying are achieved using yet another form of a DGPS system called real-time surveying or real-time kinematic. Requires a dual-frequency system linked to and continuously updating the single-frequency rover. The rover must be designed and have the software to handle RT operations. It's *not* RTCM-SC-104 DGPS.
There's a fair bit of FUD in this whole line of commentary.
The only times I've heard of this, and been able to document the result, DoD was diddling clocks on a couple of satellites to determine what'd happen with the 1024 week rollover (an inherent problem in GPS caused by its age and limitations of its design spec). They filed the appropriate notices, which are sent via the Almananc message, and which all receivers are supposed to read and act upon. A bunch of civil receivers couldn't handle this, and reported really bad results. This wasn't a GPS-system problem, it was a ground-component problem associated with commercially-acquired hardware not following the signal spec.
There have been similar tests. My $5k receivers don't notice anything, and my old Garmin GPS-55AVD doesn't have problems. My (slightly) newer GPS-IIPlus does see errors and demands to be power-cycled but then behaves appropriately. I've seen real problems with a Magellan receiver. My OEM-board SiRF and Motorola receivers do not have problems.
GPS can provide centimeter-level accuracy. Consumer-grade receivers are not built to take advantage... collect sufficient data, nor emit the appropriate corrections, to accomplish this. For that matter, the receivers necessary for cm-level accuracy with Galileo systems won't be cheap, either. TANSTAAFL.
Galileo's business model ("free", "costly", and "Safety of Life/free") is an interesting one. With GPS, the "costly" model requires paying solely the hardware and software vendors (I use OSS for my data reduction, thank you very much). With Galileo, you pay the hardware vendors, software vendors and the EU.
To the best of my knowledge, there has been no widespread, intentional system degradation since 2 MAY 2000, when Selective Availability was turned off. I suspect there is conflict-theatre degradation, but I've no hard evidence of that. Solely inferential.
The MEO/56deg orbit of the Galileo birds will *not* provide significant changes in polar coverage over GPS. Sorry. Dirt blocks RF.
The changes in signal structure, and improvements in power budget will help getting signal into buildings. Some. Multipath will still cause problems. Higher power levels will increase the effects of multipath.
Civil code-phase GPS accuracies are on the order of 6m 2d 1-sigma, and 7.5m 1d (vertical). Carrier-phase dual-frequency determinations yield 1.2-1.5cm, 2d 2-sigma accuracies, and 3cm 1d (vertical) 2-sigma accuracies with sufficient data collected using good reference sites. Military systems using P-code or P/Y, tend to achieve L1/L2 code-phase positions on the order of 1m 2d, 2-sigma. These are based on some documentation for the military systems, and experimental data for civil receivers/systems.
The US Government has mandated the removal of SA, and identified GPS as a safety-of-life system that cannot be degraded save for dire emergencies. It wasn't degraded for 9/11 (urban legend), Afghanastan, Iraq (either time, and in fact in the 1991/1992 timeframe, SA was turned off because of the massive number of civil receivers in use by the troops in theatre). Too many folks depend on it.
Augmentation via Differential GPS (DGSP), WAAS, LAAS, or real-time kinematics accomplishes about the same degree of correction and enhancement, save DGPS is more accurate closer to the DGPS station. I usually estimate a 10x improvement as a rule of thumb when asked.
In the early announcements on Galileo in the navigation community, the program was very much a "we don't trust the US and especially not their DoD, so we're going to build our own" issue. The majority of folks involved in the discussions from the US side, and I know most of them, made suggestions that would reduce potential interference, and enhance the potential for a separate system with redundant capabilities while built using different approaches to the problem. Some of the folks involved were not particularly cooperative, but they were in the minority.
I'm still not sure this is a marked improvement over the GPS system, but having another choice is good. Personally, I think it'd have been cheaper to have invested in GLONAS and used Euros to drive the Russians to improve their system. There were some nice features about GLONAS, including the fact that SA couldn't be invoked -- by design.
There will be a short quiz at the end of the next lecture.
Well... I know a guy, in the US Army, working toward his PhD, using Unreal Tournament to help model crowd dynamics for his dissertation. The end-of-the-day test and eval sessions were a hoot!
Freedom was foreseen as primarily a US venture for launch and support, as already indicated. Ascent from French Guiana was possible (they were our friends, right?) without too much trouble in a 28.5 deg inclined orbit, but it took a fair bit of delta-V to get from Baikanor to 28.5 deg. Of course, that wasn't our problem as the Soviets were on the other side.
OK, international politics aside.
One of the real problems we saw was the US Congress, and yes, NASA management.Space Station Freedom was often a dumping ground for "retired in place" senior engineering management waiting for that magic day when they could sit at home and impede their wives instead of coming to the office and impeding engineers. That's not to say we didn't have decent, enthusiastic, qualified management but they were outnumbered... or simply out-numbed... by the incompetents.
A lot was preordained, despite engineering advances. "Don't try to convince me, my mind is made up." I could go on at length about the decision to scrap the 100 khz power distribution system on Freedom in favor of the DC system. I was around when the "test" destroyed some computer hardware at MSFC that was used as justification, despite the fact that the test was protested by competent engineers with a knowledge of VAX power supply design. Were there problems with the high frequency AC distribution? Some, but not insurmountable.
SSF was also a training ground for kids right out of college. Get them in, turn 'em loose with little guidance, slap 'em around a bit until they started doing good design, then move them to Shuttle.
We had a lot of design by Aerospace Conglomerate, too. Let's get that design that Lockheed wants, because it'll make them easier to deal with at contract time. Let's use THIS design that MD wants, even if it's not what NASA wants/requires, because we think their design is going to make them do something else for us on another project.
Still, and all, most of the conglomerate designs I saw, worked with, and helped shape (and, yes, I worked for a contractor company, too, but I was doing specs and requirements, as well as working with the prototyping) would have been acceptable, even if somewhat limiting in their own ways.
The BIG problem, however, was Congress. Every three years or so, we'd get a "stop what you're doing, reassess the design, and then start over" command from the Hill. I've gotta say, we wasted a LOT of money on those exercises, and we wasted a LOT of time.
There are improvements borne of waiting time and engineering advances in ISS that would not have been, and may never have reached SSF or Alpha, but we could well have bent metal and flown hardware by 1990 if Congress had stuck to original budgets and timelines and stayed the hell out of the way. I flew prototype hardware in 1992 that was the first piece of Space Station hardware to fly, be proven and certified for on-orbit Space Station operation. I could have flown it 3 years earlier save the Challenger accident.
Final thought. We developed or promoted a lot of stuff that's now common place in the world. Speaking from the perspective of medical hardware development (I also did a bit for the medical facility in terms of GNCC and COMMS) there's a lot of stuff I see in hospitals, doctors' offices, dentists' offices and ambulances that makes me smile and think, "I worked with the prototype of that...", or, in a couple of cases, "I wrote the SBIR paperwork that made that happen".
So, yes, NASA's efforts HAVE improved life ont he planet. Really.
BPL isn't just an HF-impinging system. There are VHF, UHF and microwave signal injectors. Of these, most (although the microwave set appears to be an exception) are pretty significant emitters when energized. And pretty bad at tolerating "competing" signal impinging on the spectrum on which BPL tries to use.
Local RF emmissions, in licensed operations, near a BPL-enabled powerline are likely to cause problems if there's an overlap.
Let's address your use of satellite communications for emergency comms:
As long as the load is controlled, this will probably work. What has been demonstrated time, and time again, is that local cellphone and trunked networks seem to overload pretty quickly in the local area or even region of the disaster/event if that network is used by, say, the media and local emergency managers. VHF/UHF backups using Amateur Radio have shown theur worth in those settings over and over.
Similarly, the HF networks have been useful to disseminate data from areas which have lost conventional telecommunications due to infrastructure damage or eradication. Look at large areas of Florida just last year... or Sumatra in December. The amount of data transferred out of Sumatra by Amateur Radio was pretty significant.
I've kept my ham license up. It's an interesting tool to use for RF experimentation. Did I mention the linking I've been doing with 802.11b/g hardware using the 6 channels Ham Radio happens to be primary on? 2-10 Mb/s for data and voice using conventional data methods, but in the ham commuity, is interesting. And, I've been playing with 802.16 in a similar manner. I've had a leg up on a lot of folks who don't have my RF background, but did sit through a propagation course in EE in college. I've done it, (I've studied it, too) so it's not magic or theory to me.
In general, I agree with the sentiment herein.
I'm only gonna take exception to the comment regarding the training and education of the air personalities on the Weather Channel. They *ARE* professionals and know their stuff. And they're less likely to be strong proponents of this bill than the special interest (AccuWeather) that pushed for it. They have real atmospheric scientists. Where else do you get discussion of severe weather from a PhD who's well published, and really interested, in it? WGN? I think not!
And I'm assuming you're using the http://mesonet.agron.iastate.edu/ Iowa Environmental Mesonet for your imagery? They do a good job for real-time imagery... And they're interested in data from Amateur Radio and trained spotters, too.
Learn to read FAA Forecasts? Just where do you think they come from? The FAA doesn't run models, nor do they have weather stations. Once they dissetablished their airport weather observers and ASOS was the norm, they ceased to employ anyone in observation. The few remaining briefers at the Automated Flight Service Stations have historical basis for their knowledge and the ability to interpret the data, but most are not trained meteorologists. Many's the time, in getting a "standard briefing", I've heard something along the lines of, "A warm moist southerly flow predominates. Ceilings are at or above 10,000 feet. The possibility for afternoon thunderstorms exists. VFR flight is not recommended." Problem here? Sure. There's a strong cap and the potential for those thunderstorms in miniscule. There's inadequate moisture in the region to support convection, the cloud deck is widely scattered at 10k, with a few puffy Cu, and winds are negligible. If you fly... does the term CAVU mean anything?
This bill is yet another attempt to drive NWS away from "competetion" with the privates. In my work, I've seen the privates tell us they can be more mobile and responsive to the Public needs. AWS likes to point to the Baltimore ferry accident, claiming they had an indication of the gust-front that caused the accident from their network of stations well before the NWS's more sparse network of official stations saw an indication. The result? A cooperative agreement between a private sector firm and NWS to provide NWS with that data in the case of such an event, allowing/providing NWS the data to make the appropriate warnings. Doesn't preclude AWS from offering their own warnings and value added services, but it IS a case where they can help the public rather than trying to hamstring the Weather Service.
This is *exactly* what I've been thinking of for a weather station. Add the processor to the sensor, place one on the anemometer at 10m, another on the temp sensor at 10m, but have it tightly networked to the 2m temp/humidity sensor. Presto! intelligent heat flux calculations. Tie the 10m anemometry to the 2m wind speed, voila`, 3d wind data.
Lose a sensor, no problem. The rest of the site's up. Lose a data collector? No problem. It's the same as losing a sensor.
Wow! This is great!