Re:Its Not The Size or Speed/Its Also The Granular
on
New Weather Computer
·
· Score: 1
While its nice that they have a bigger and better computer unless and until thay have better input/initialization data to feed it, I can't see how the forecasts will get any better.
Twice a day (0Z and 12 Z) the main prediction models are initialized with data from all over the world. Not only surface data, but "upper air" data as well. Upper air data come from sparsely located stations that actually have the ability to send up and record data from weather balloons.
Exactly right. On the other hand, there is recent work which may greatly increase the amount of upper-air data available: there is a program to put up a fleet of satellites to measure carefully the behavior of GPS satellite signals as they pass through the atmosphere as the edge of the earth goes "between" the GPS satellite and the receiving satellite, and use the diffraction profile observed as the GPS satellite goes "over the horizon" to derive density profiles (and potentially wind profiles, via interferometry) very densely over the entire globe. This presents an incredible opportunity to solve the atmospheric-data sparseness problem.
On the other hand, another major deficiency in today's meteorological models is getting the land-surface right (e.g., just how wet is the soil? and how much of the sun's incoming energy shows up as conducted heat, and how much serves to evaporate water?) Present models aren't very good at this; we're working on it here at NCSC (see URL http://envpro.ncsc.org/projects/dashmm but it winds up very computational -- you have to use resolutions well under 1 KM in order to get the terrain-slope/drainage effects right. And you need to do that globally! (Satellite data aren't very useful; statellite radar generally doesn't penetrate beyond about 1 cm. And there's no funding to put up long-wave-radar interferometry equipment that might do better.)
Good methodologies are designed to protect the project from idiots of all types...
To which you need to add the concept of
moral hazard: by taking a measure to avoid an undesirable outcome, you
encourage an even more undesirable outcome
(fire insurance and arson-for-profit, for example).
The moral hazard of trying to idiot-proof
the universe is to encourage the proliferation
of idiots.
The Internet is a HUGE namespace collision. Take Ajax. Suppose Boeing (or some other large, old) company holds a valid (worldwide) trademark on Ajax in rocket fuel...
Actually, I think it's Lockheed that built the Nike Ajax anti-aircraft missile...
. You miss one important possibility which has been pushed to the limits by the Software Publishers Association and the Church of Scientology: the Copyright Act permits very aggressive ex parte action by the plaintiff. According to the precedents set by the SPA and the CoS, it is entirely within the realm of legality for the plaintiff to roll up to the offender's place of business with a Federal Marshal and several eighteen-wheelers, and cart off every computer in the place for investigation of copyright violations, at the plaintiff's leisure in a warehouse of the plaintiff's choosing.
One feature I use all the time in my programming editor is multiple windows into the same edit-buffer. Doing a decent job of this is one thing I do like about MSWord, and doing it wrong (in the last version I tested) is one of the things I hate about (the mis-named) WordPerfect.
Does the current version of Applixware support this capability?
In terms of the quality of machine-code they generate, Fujitsu produces some of the best linux/x86 compilers out there. (See URL http://www.tools.fujitsu.com/) -- and at reasonable prices, too! The only sticking point from my point of view is that they don't support the OpenMP standard for shared-memory parallel programming.
If anyone from Fujitsu is listening: I'll buy your compiler suite in an instant, if you'll give me support for OpenMP!.
I do met and air quality modeling, have dual-processor RH6 desk-side boxes both at home and at the office, and
need
the parallel for development work (final production tends to be on O2000's and the like).
Question -- window manager operations
on
KDE 2.0 in Action
·
· Score: 1
I like large virtual desktops -- I currently run 2Kx1.5K@32bpp virtual, both at home and at work (because my 16M graphics cards won't support anything larger;-( ). With large virtual displays, it is really useful to be able to hang a "desks&windows" menu off one of the mouse keys, the way FVWM does with "MB2" by default.
How do I find out which WM's let me do something like this, and for that matter how do I find the documentation on how to do it? Frankly, the RTF(M4)S for FVWM has me annoyed. (I do C, Fortran, lisp, HTML, and hard-core numerical algorithms; my brain doesn't have the space left over to hold everything else, too!)
Thanks!
The Privacy Act, the RIAA, and racketeering
on
Copyright!
·
· Score: 1
Isn't this grossly illegal under the Privacy Act?
IANAL, but my understanding is that it pretty much requires written permission of the student before the school is permitted to release anything to anybody. And since they are trying to coerce the university into breaking the law on that point, doesn't this open the door for student class-action RICO suits against the RIAA? (RICO, imnho, is an unconstitutionally strong law, itself, -- but is plenty strong to completely break up the RIAA, if if it loses!).
...has been around for a while: have a look at http://www.norsam.com/hdrom.htm: NORSAD has been building archival storage using charged particle beam writers and optical readers for some time; they get 165GB on a disk (for computer use), and also have a low-tech analog variant where they put analog images down on silicon wafer (so that, if you have to, you can read it with just a microscope; no special hardware required. Think of the latter as the ultimate (?) evolution of microfilm. The digital HD-ROM gets write rates of 20 Mb/sec, and read rates of 60 Mb/sec, with hopes of pushing these into the Gb/sec range. Using silicon-wafer disks, both the digital HD-ROM and the analog microfiche-variant should be _very_ permanent archival media.
And real enough that they're getting some of their funding from IBM, fwiw.
a 90 Hz 24x60@300dpi (65" diagonal) quarter-cylinder screen as a backdrop for my desk (that gives you a 38" radius, leaving you a comfortable 18"-deep or so quarter-annulus for "conventional" desk-top). We had quite an extensive discussion of this sort of thing back about a year ago on the "comp.arch" news-group, with the consensus that this was a nice sweet spot for the ergonomics -- that's the resolution and scan frequency at which further increases do not significantly imprive viewability, a size that is easily viewable from your chair, and a general design that would work quite well for lots of the participants in that discussion.
For grins: the whole discussion was started as a response to some idiot who claimed a Pentium had all the CPU horsepower anyone could make real use of (just driving the screen I described requires far more horsepower than any single CPU currently available:-)
...1600x1024 max resolution. Not much less than the IBM monitor.
Let me tell you: there is a huge difference between Apple's 1600x1024 and the IBM's 2048x1536. I know... that's the virtual size I run both at home and at work (with only 16 M of video ram, the cards won't go to 2400x1700@32bpp, damn it!).
Unless you are a criminal, and quite a significant one, you have nothing to fear from the FBI.
Do you simultaneously trust Bill Clinton [Filegate] and Richard Nixon [Watergate] on that? If so, can I interest you in a fine piece of revenue-producing architecture in a great urban location? -- it's called the Brooklyn Bridge...
Actually, there's a big scandal going on in LA right now about literally thousands of illegal LAPD wiretaps used to surreptitiously gain evidence in a great number of cases.
One basic principle in any strategic endeavor: do threat analysis ("How can this be used to attack me?") in addition to intent analysis ("Who has motive to attack me?"). The threat analysis for a tappable IPV6 is decidedly hostile to a free society.
Well... there is no such "right to privacy" for individuals.
Wrong!! It happens that that bunch of lawless tyrants in Washington may violate the Constitutional provisions, but those provisions are still there and are still the fundamental law of the land, in the US, anyway. The Tenth Amendment says that the fundamental rights of the people are too numerous to be enumerated but that they must still be preserved, that the government only has those powers delegated to it, and that it may not violate those non-enumerated rights.
And by their behavior in using encryption for both their public and private correspondence, as well as by various public statements they made on the subject, it is clear that at least Washington, Jefferson (inventor of that encryption device called the "Jefferson Wheel"), Franklin, Madison, and Monroe considered communications privacy to be one of those Tenth-Amendment rights.
This is a surprise. I expected the same situation IBM went through in the eighties. The process went through the courts for about ten years in the appeals process, and I believe that IBM really didn't lose much but time.
Actually, IBM lost a tremendous amount of focus and competitiveness, due to the distraction of being in the legal spotlight.
I'm a big fan of things like the Pure Food and Drug Act. Think how many fewer "seal babies" there would have been if the Europeans were as tough as the FDA when it came to Thalidomyde.
Actually, I'm a *lot* more in favor of a Microsoft break-up than I am of the FDA. That the FDA came out of the Thalidomide thing looking good is an accident, where they *didn't* follow their own procedures, and got lucky anyway. And I consider the way they suppressed information about aspirin and heart disease, and about ulers and antibiotics to be no less than murder on the installment plan!
Don't go claiming either MS or FDA is good -- they're both thoroughly corrupt!
In the libertarian-ayn-rand-dog-eat-dog-jungle, there is no morality.
The classical libertarian definition of immoral is in terms of "theft, force, or fraud". And if you look at the facts of the Stacker, AARD, Caldera, and DoJ cases, Microsoft looks to be guilty of all three of these immoralities.
The best monitor out there right now is the 2560x2048 @ 150dpi flat-screen unit produced by Xerox's dPiX subsidiary for AWACS and medical imaging applications.
There is a behavior/personality-transformation syndrome frequent enough that it has received quite a bit of commentary within the swing-dance and ballroom-dance communities. It's worth thinking about, to some of you guys and gals (it's been observed in both sexes, though more frequently among guys and more dramatically among gals):
X is a geek. X is not the most comfortable at interactions with the opposite sex (more bluntly, X is
terrified of them). X is most probably not athletic, may be middling musical, and is very good at applying focus and intelligence at matters X finds important.
X takes a dance class, at least partly thinking, "Here is a situation in which I can interact with the opposite sex, for which the rules are simple and well-understood, and in which one of the rules is that rudeness is forbidden."
It takes a bit, but by application of intelligence, focus, and work, X finds out that he/she is actually good at dancing, and that X enjoys it (partly the intellectual challenge, partly the never-before experienced feeling of doing something physical and doing it well.
Shortly, X finds that the opposite sex (among the dancers, at least) reacts very favorably--so much so that X is probably a bit disconcerted, but, hey! X can live with that!
With the positive reinforcement, X continues to get even better; what is more, X's whole personality blossoms--becoming much more confident and outgoing.
Quite a number of the nationally recognized "dance gods" and "dance goddesses" fit this syndrome, in fact (particularly in the swing community. And I've seen it happen many times, with X's of both sexes. (There's one female engineer I know at a local university who has transformed from "pure geek" to "drop-dead gorgeous" this way -- while retaining (enhancing, actually) a really pleasant personality.)
It's worth trying. Disclaimer: I'm one of those X's myself, for that matter:-)
For the n'th time, ArsTechnica's RISC article is at least halfway clueless. The article ignores the definitive article by John Mashey, available on-line at http://www.inf.tu-dresden. de/~ag7/mashey/RISCvsCISC.html. Two major points Mashey makes (and which "Hannibal" botches) are
RISC or not is about arcitecture, not implementation
RISC is really about having an architecture whose instructions pipeline cleanly, and which responds to the demands of actual workloads.
If it's possible within an architecture to generate more than one page-fault within an instruction, then you're not on a RISC (the record seems to be a VAX 3-operand memory-to-memory indirect-indexed instruction with memory-based indexes and offsets, which can generate up to 47 (!) page-faults.
If you take the point of view that a P6 is a RISC core running an x86 interpreter, then still the user-visible architecture is not RISC. It would only be RISC if you let me program the core directly with its native micro-ops. "Hannibal" still doesn't understand this distinction between architecture and implementation.
Not even close!! Do you actually know anything about VMS ??!?
Every time Dave Cutler tried to put features into NT to give it some level of robustness, Bill Gates shot him down for not retaining backwards compatibility with Bill's two-bit (DOS) imitation of an eight-bit operating system.
Why the HELL are you specifying pixel widths ???? That kind of device-dependence should have gone away with the Dark Ages!
You're guaranteeing that it'll only render correctly on screens that look like yours. Right now, you don't know whether the user has a 72dpi screen or something with higher resolution like 110dpi (my Hitachi does that already), and in the future it's entirely reasonable that screen resolution will be MUCH higher -- the human-factors people suggest 300dpi as a reasonable compromise between hardware complexity and the eye's ability to resolve what it sees; beyond that, you still get benefits in viewability, but they're marginal).
And think: your most affluent customers are likely to be the ones with the highest-resolution hardware; do you really want to offend that part of your customer base that has the most spendable money??
The Department of the Navy serves as the country's official timekeeper, with the Master Clock facility at the Washington Naval Observatory. National Defense Authorization Act for Fiscal Years 1992 and 1993.
They have quite a number of links to their services there, including network time protocol services.
Slashdot Mirror
On the other hand, another major deficiency in today's meteorological models is getting the land-surface right (e.g., just how wet is the soil? and how much of the sun's incoming energy shows up as conducted heat, and how much serves to evaporate water?) Present models aren't very good at this; we're working on it here at NCSC (see URL http://envpro.ncsc.org/projects/dashmm but it winds up very computational -- you have to use resolutions well under 1 KM in order to get the terrain-slope/drainage effects right. And you need to do that globally! (Satellite data aren't very useful; statellite radar generally doesn't penetrate beyond about 1 cm. And there's no funding to put up long-wave-radar interferometry equipment that might do better.)
The moral hazard of trying to idiot-proof the universe is to encourage the proliferation of idiots.
. You miss one important possibility which has been pushed to the limits by the Software Publishers Association and the Church of Scientology: the Copyright Act permits very aggressive ex parte action by the plaintiff. According to the precedents set by the SPA and the CoS, it is entirely within the realm of legality for the plaintiff to roll up to the offender's place of business with a Federal Marshal and several eighteen-wheelers, and cart off every computer in the place for investigation of copyright violations, at the plaintiff's leisure in a warehouse of the plaintiff's choosing.
IANAL, either, but that's what's been happening!
Does the current version of Applixware support this capability?
I've had too damned many denial-of-service attacks disguised as Javascript pages, and I'm sick of $%(*&$! idiots who won't talk with me otherwise!
If anyone from Fujitsu is listening: I'll buy your compiler suite in an instant, if you'll give me support for OpenMP!.
I do met and air quality modeling, have dual-processor RH6 desk-side boxes both at home and at the office, and
- need
the parallel for development work (final production tends to be on O2000's and the like).How do I find out which WM's let me do something like this, and for that matter how do I find the documentation on how to do it? Frankly, the RTF(M4)S for FVWM has me annoyed. (I do C, Fortran, lisp, HTML, and hard-core numerical algorithms; my brain doesn't have the space left over to hold everything else, too!)
Thanks!
IANAL, but my understanding is that it pretty much requires written permission of the student before the school is permitted to release anything to anybody. And since they are trying to coerce the university into breaking the law on that point, doesn't this open the door for student class-action RICO suits against the RIAA? (RICO, imnho, is an unconstitutionally strong law, itself, -- but is plenty strong to completely break up the RIAA, if if it loses!).
And real enough that they're getting some of their funding from IBM, fwiw.
a 90 Hz 24x60@300dpi (65" diagonal) quarter-cylinder screen as a backdrop for my desk (that gives you a 38" radius, leaving you a comfortable 18"-deep or so quarter-annulus for "conventional" desk-top). We had quite an extensive discussion of this sort of thing back about a year ago on the "comp.arch" news-group, with the consensus that this was a nice sweet spot for the ergonomics -- that's the resolution and scan frequency at which further increases do not significantly imprive viewability, a size that is easily viewable from your chair, and a general design that would work quite well for lots of the participants in that discussion.
For grins: the whole discussion was started as a response to some idiot who claimed a Pentium had all the CPU horsepower anyone could make real use of (just driving the screen I described requires far more horsepower than any single CPU currently available :-)
One basic principle in any strategic endeavor: do threat analysis ("How can this be used to attack me?") in addition to intent analysis ("Who has motive to attack me?"). The threat analysis for a tappable IPV6 is decidedly hostile to a free society.
And by their behavior in using encryption for both their public and private correspondence, as well as by various public statements they made on the subject, it is clear that at least Washington, Jefferson (inventor of that encryption device called the "Jefferson Wheel"), Franklin, Madison, and Monroe considered communications privacy to be one of those Tenth-Amendment rights.
fwiw
Don't go claiming either MS or FDA is good -- they're both thoroughly corrupt!
I just wish I could afford one, fwiw. 25K ;-(
It's worth trying. Disclaimer: I'm one of those X's myself, for that matter :-)
- RISC or not is about arcitecture, not implementation
- RISC is really about having an architecture whose instructions pipeline cleanly, and which responds to the demands of actual workloads.
If it's possible within an architecture to generate more than one page-fault within an instruction, then you're not on a RISC (the record seems to be a VAX 3-operand memory-to-memory indirect-indexed instruction with memory-based indexes and offsets, which can generate up to 47 (!) page-faults.If you take the point of view that a P6 is a RISC core running an x86 interpreter, then still the user-visible architecture is not RISC. It would only be RISC if you let me program the core directly with its native micro-ops. "Hannibal" still doesn't understand this distinction between architecture and implementation.
Every time Dave Cutler tried to put features into NT to give it some level of robustness, Bill Gates shot him down for not retaining backwards compatibility with Bill's two-bit (DOS) imitation of an eight-bit operating system.
You're guaranteeing that it'll only render correctly on screens that look like yours. Right now, you don't know whether the user has a 72dpi screen or something with higher resolution like 110dpi (my Hitachi does that already), and in the future it's entirely reasonable that screen resolution will be MUCH higher -- the human-factors people suggest 300dpi as a reasonable compromise between hardware complexity and the eye's ability to resolve what it sees; beyond that, you still get benefits in viewability, but they're marginal).
And think: your most affluent customers are likely to be the ones with the highest-resolution hardware; do you really want to offend that part of your customer base that has the most spendable money??