How much of this is due to difference in application support: ie, *NIX being used for specific and limited robust applications (serving, number-crunching, database), while 2K boxen were used for every document written and every printjob run (lots of mix-n-match GUI)? It's those high-intensity "standard user" apps that take so much maintenance after all (somewhere like 80% of problems around here are of a "the printer won't print my doc" rather than a "the gcc compiler is broken" problem--if Linux was used for all word processing I'm sure the Linux support load would go up).
I have discovered a truly remarkable new element, but it's decay path is too long to fit in this comment.
Re:Another place for good games
on
Layoffs at WotC
·
· Score: 1
Every game of their's that I've played (maybe a dozen) has been worth at least twice what I payed for it, usually more. With a caveat is that every one is a beer&pretzels type, which are the sort I go for anyway. Neeeed Braaaiiins...
I work with high-end 3D visualization of scientific (climate) data, and we've called this "the Nintendo Effect" for at least 10 years. "The Nintendo Effect" just means that (since hardware is sold cheapo to make returns on games) we government scientists each year will be able to update to the "next-to-hottest" generation graphics system at prices affordable to gamers, e.g., $500 or less. And we do, too.
"Er, that quake engine we all just installed was tax-supported investment in the hardware industry. We'll see returns on it. Really."
The taxpayer gives the government money, and that government is buying expensive software instead of using free (Linux) or for hundreds of dollars less (office) per seat?
It is quite possible that the current setup is the result of competetive bidding. But how many times should it be said that "cost" includes support: in this case, the cost of my time (I'm a contractor, so it's billed) to make the tools they give me match those used in previous work?
And all this could be changed if software procurement was open to competitive bidding?
It is trivial to write a bid to get what one wants, especially with computers: "Must support Excel functions and macros in order to process or re-use code from other researchers." This is honest, by the way, I have (unfortunately) reached the point where OpenOffice/Excel compatibility issues would cost me over $1000 of my time to fix. Not too many more before the difference is fully paid. Tho in spare time, one tries to do enough conversion to make it eventually possible.
FWIW, in my job at a government science installation:
1. The Official Agency Word Processor is Word Perfect.
2. Much of my info is stored in Oracle Databases.
3. Everyone has Windows and Office, much of that is to read documents that others (the Public) send us. Yes, I know OpenOffice could do that. That's what I'm using.
4. For our computations, we take data from public formats (cdf), process it with legacy Fortran, and run computations on a variety of *nix, inluding Suns, and increasingly, *many* Linux boxen.
5. email is netscape.
The point being, these installations end up running, like many good companies, on a combination of legacies, IT whims, and user needs.
Execept, of course, when the next "trend from the top" comes down and, like such trends in private companies, set directives that trump the local users' needs and create another layer of mess. And this helps us, or open source, how?
This kind of stuff should be pursued the same way that the Linux "community" has pressed businesses--- with informed, local IT managers pushing open source solutions, not from the top.
Yes and no. If you're looking at an analytic (not numerical) solution to a set of time-domain-only state variables, you are correct, a negative sign is sufficient. However, in the example of the article, each state value has multiple dimensions--- the analogy that "converging" on your mother isn't an inverse of "diverging" from her is apt.
Part of this is the convergence problem. There are a large number of Life patterns that lead to a blinker or a blank screen. Starting with the blinker won't take you back to them. This doesn't invalidate running forward and matching to results. The real problem is finding the initial conditions.
Some argue that human events include irreversible processes, so perhaps the "run backwards" test is not valid.
While trying to get back up your mother might lead to loss of life and having a Complex named for you, it shouldn't be mistaken for birth with a negative sign.
Whaddaya talking about? I'm a biologist, I'm working for the govm't, and we all use fortran. Locked, indeed. My only wasted time in College was learning C.
Methods: Lately, with a long commute after work, trying to puzzle out some (engineering-type) problem, I find myself trying to do many "back of the envelope" calculations in my head. Observation: Didn't learn either on a calculator.
Results: I am absolutely terrible at long division, but do a pretty good job visualizing graphs and operations (even nonlinear) like the "shape" of two functions when divided/multiplied by each other.
Conclusions: (1) I am more visual than numerical. (2) Long division absolutely sucks as an algorithm.
I support (2)---various experts speed math tend to agree with this.
1. Sperm whales eat many types of smaller squid (50cm or smaller mantle length) swallowing many in big gulps. It's not at all certain what size squid they eat most.
2. Size of squid beak has nothing to do with size of prey. I've got a video of a 10cm squid biting a ~15cm long fish near the head, wrapping tentacles around, and "spinning" the fish, scales flying, while the beak strips bits of meat off, "skeletonizing" the larger fish in seconds. Food found in squid stomachs is often in little little bits but coming from fish species of larger sizes.
This process only works for established artists
(like Stephen King) who can expect a large
return for their product ahead of time.
What you're saying is that if an unknown
produces a product for $10 (like, the first
Myst), which because of its brilliance is
distributed to millions, the author gets... $10.
(A) Shout "FIRE!" and get crushed in the panic.
(B) Walk out quietly...who cares about anyone else?
(C) Tell your closest neighbor and hope that they're a fireman.
(D) Pour on gasoline so everyone will get out faster.
I bought 2 Dell laptops, one personal, one on a university service contract. Approx. same price of warantee paid for both (1 year/ship to factory).
Both have needed tech support (1 minor config, 1 hardware). The support on the contract was LIGHT YEARS better. Better staff, better response time, better service. On the private purchase, it was bad and frustrating, on the contract it was a dream (and that was the harder problem). And on the contract I could order without *any* MS products---they supply our labs with linux.
This isn't a Y2K-specific bug but just bad data-management in software. Data entry can screw up in many places, a similar bad effect may have happened if someone had misentered a birth-year of '71 as '17 (giving a mother's age in the 80s).
The key is to develop software that flags "silly" values, positive or negative. And even then you're going to miss some, so some calculation steps should be displayed in the output.
Just clumsy programming, not Y2K. Move along. (I guess we're desperate for some Y2K bugs, tho.)
Slashdot Mirror
How much of this is due to difference in application support: ie, *NIX being used for specific and limited robust applications (serving, number-crunching, database), while 2K boxen were used for every document written and every printjob run (lots of mix-n-match GUI)? It's those high-intensity "standard user" apps that take so much maintenance after all (somewhere like 80% of problems around here are of a "the printer won't print my doc" rather than a "the gcc compiler is broken" problem--if Linux was used for all word processing I'm sure the Linux support load would go up).
You mean it gets invited to all the "Big 8" parties while The Moon has to sit outside and park cars?
I have discovered a truly remarkable new element, but it's decay path is too long to fit in this comment.
Every game of their's that I've played (maybe a dozen) has been worth at least twice what I payed for it, usually more. With a caveat is that every one is a beer&pretzels type, which are the sort I go for anyway. Neeeed Braaaiiins...
Foolish, foolish geek.
What sits is but a Pointer.
Access it you must.
Well that's ok, I'm doing historical simulations and it's better to have the clock run backwards.
I work with high-end 3D visualization of scientific (climate) data, and we've called this "the Nintendo Effect" for at least 10 years. "The Nintendo Effect" just means that (since hardware is sold cheapo to make returns on games) we government scientists each year will be able to update to the "next-to-hottest" generation graphics system at prices affordable to gamers, e.g., $500 or less. And we do, too.
"Er, that quake engine we all just installed was tax-supported investment in the hardware industry. We'll see returns on it. Really."
It is quite possible that the current setup is the result of competetive bidding. But how many times should it be said that "cost" includes support: in this case, the cost of my time (I'm a contractor, so it's billed) to make the tools they give me match those used in previous work?
And all this could be changed if software procurement was open to competitive bidding?
It is trivial to write a bid to get what one wants, especially with computers: "Must support Excel functions and macros in order to process or re-use code from other researchers." This is honest, by the way, I have (unfortunately) reached the point where OpenOffice/Excel compatibility issues would cost me over $1000 of my time to fix. Not too many more before the difference is fully paid. Tho in spare time, one tries to do enough conversion to make it eventually possible.
FWIW, in my job at a government science installation:
1. The Official Agency Word Processor is Word Perfect.
2. Much of my info is stored in Oracle Databases.
3. Everyone has Windows and Office, much of that is to read documents that others (the Public) send us. Yes, I know OpenOffice could do that. That's what I'm using.
4. For our computations, we take data from public formats (cdf), process it with legacy Fortran, and run computations on a variety of *nix, inluding Suns, and increasingly, *many* Linux boxen.
5. email is netscape.
The point being, these installations end up running, like many good companies, on a combination of legacies, IT whims, and user needs.
Execept, of course, when the next "trend from the top" comes down and, like such trends in private companies, set directives that trump the local users' needs and create another layer of mess. And this helps us, or open source, how?
This kind of stuff should be pursued the same way that the Linux "community" has pressed businesses--- with informed, local IT managers pushing open source solutions, not from the top.
Um, for what it's worth, the official supported word processor of the Department of Commerce (n-thousand employees) is WordPerfect.
Don't know about other Departments.
Sorry, forgot that the most important part is discretion.
Yes and no. If you're looking at an analytic (not numerical) solution to a set of time-domain-only state variables, you are correct, a negative sign is sufficient. However, in the example of the article, each state value has multiple dimensions--- the analogy that "converging" on your mother isn't an inverse of "diverging" from her is apt.
Part of this is the convergence problem. There are a large number of Life patterns that lead to a blinker or a blank screen. Starting with the blinker won't take you back to them. This doesn't invalidate running forward and matching to results. The real problem is finding the initial conditions.
While trying to get back up your mother might lead to loss of life and having a Complex named for you, it shouldn't be mistaken for birth with a negative sign.
So he takes the Indigo case and puts that Zip drive on top? Oh dear, the clashing, it hurts my eyes so, it is so utterly *minging*, darling.
Whaddaya talking about? I'm a biologist, I'm working for the govm't, and we all use fortran. Locked, indeed. My only wasted time in College was learning C.
Well, the sudden switch from "happy wallpapering" to "quiet menace" mood music helped...
My experiment:
Methods: Lately, with a long commute after work, trying to puzzle out some (engineering-type) problem, I find myself trying to do many "back of the envelope" calculations in my head. Observation: Didn't learn either on a calculator.
Results: I am absolutely terrible at long division, but do a pretty good job visualizing graphs and operations (even nonlinear) like the "shape" of two functions when divided/multiplied by each other.
Conclusions: (1) I am more visual than numerical. (2) Long division absolutely sucks as an algorithm.
I support (2)---various experts speed math tend to agree with this.
1. Sperm whales eat many types of smaller squid (50cm or smaller mantle length) swallowing many in big gulps. It's not at all certain what size squid they eat most.
2. Size of squid beak has nothing to do with size of prey. I've got a video of a 10cm squid biting a ~15cm long fish near the head, wrapping tentacles around, and "spinning" the fish, scales flying, while the beak strips bits of meat off, "skeletonizing" the larger fish in seconds. Food found in squid stomachs is often in little little bits but coming from fish species of larger sizes.
Like everything else on my keychain, this is only useful if it doubles as a bottle-opener.
This process only works for established artists
(like Stephen King) who can expect a large
return for their product ahead of time.
What you're saying is that if an unknown
produces a product for $10 (like, the first
Myst), which because of its brilliance is
distributed to millions, the author gets... $10.
When you see a fire in a crowded theatre, you:
(A) Shout "FIRE!" and get crushed in the panic.
(B) Walk out quietly...who cares about anyone else?
(C) Tell your closest neighbor and hope that they're a fireman.
(D) Pour on gasoline so everyone will get out faster.
I bought 2 Dell laptops, one personal, one on a university service contract. Approx. same price of warantee paid for both (1 year/ship to factory).
Both have needed tech support (1 minor config, 1 hardware). The support on the contract was LIGHT YEARS better. Better staff, better response time, better service. On the private purchase, it was bad and frustrating, on the contract it was a dream (and that was the harder problem). And on the contract I could order without *any* MS products---they supply our labs with linux.
Hey, that's a plan, bet the Taliban will be much nicer to you if you hide it in porn.
This isn't a Y2K-specific bug but just bad data-management in software. Data entry can screw up in many places, a similar bad effect may have happened if someone had misentered a birth-year of '71 as '17 (giving a mother's age in the 80s).
The key is to develop software that flags "silly" values, positive or negative. And even then you're going to miss some, so some calculation steps should be displayed in the output.
Just clumsy programming, not Y2K. Move along. (I guess we're desperate for some Y2K bugs, tho.)