[...] most of the history of countries comes from the CIA Factbook, the US State Department, or [...]
There is also another explanation, other than an intended bias, for why material from U.S. federal departments and agencies would be widely used in Wikipedia: U.S. federal documents and publications are in public domain and so excerpts from them can be included in Wikipedia articles without copyright worries.
David Pogue has also authored a specific Missing Manual "Switching to the Mac." The description in the parent post would apply perfectly to it as well.
> So the machines themselves do not have any DICOM interfaces?
In this particular case, from the manufacturer's point of view, the console (that Windows workstation) is part of the MR scanner, and the part that represents the scanner towards the hospital network. Embedded control and data acquisition systems send data to the console over direct fiber-optical links using a proprietary protocol. That raw data is not clinically usable until it has been reconstructed into images on the console. After reconstruction the images can be displayed, printed, or transferred using DICOM.
Obviously some other vendors might have designed their systems differently. But I have seen some other MRI systems that in a similar way used separate reconstruction computers that appeared much like standard PC hardware. Quick reconstruction of MR images takes a lot of processing power and my hunch is that high-end off-the-shelf workstations (Windows or Unix) provide that power relatively cheap.
> The devices themselves DO NOT RUN ANY OS. The devices > are medical equipment that take pictures and send them > to a "server," basically.
"Taking pictures" is not that simple in CT and MRI, because the raw output you get is not readable like the final image. In MRI, the raw output is in frequency domain and typically a Fast Fourier Transform is needed to bring it to spatial terms. The systems I know best (Philips low-field open MRI) currently use for image reconstruction HP workstations that run Windows 2000. The same workstation is the MR scanner operator's console. Without that computer operational, no scan can be configured or images produced.
I hear that some detailed proposals for "abstract communication bootstrap procedures" have been made. One such is Dr. Hans Freudenthal's LINCOS (short for "Lingua cosmica") that he published as a book in 1960.
I have sometimes used a printed table to aid myself in memorizing a pseudorandom password, too. I did it by printing a table of random characters from the set of lower and upper case letters, numbers, and some punctuation, like this:
0 3 x C 6 m c Q 5 q u s
8 e v 7 u K T / W 8 4 1
6 j B y . 8 o r = 8 S 5
O F v L 4 g 3 4 p I W 6
c l B P E u Z 9 6 L y 5
% p U A a 9 % d 5 A H v
J e % ! C 3 b . D U 5 U
Q O S l t J Q E P r c L
P 4 g n a S 9 9 C R b 7
% 9 x E = 5 d i o l 8 G
R h Q Q A e o y x R 9 Z
R E 3 N 8 c A e I 7 0 d
and then deciding from where in that table to read a password. Obviously the password could not be in a straight row. It could be a spiral around an initial character, part of a knight's tour, alternating picks from several lines, characters at intervals based on the Fibonacci sequence, or whatever rule one could devise. This effectively replaced remembering the password by remembering a pattern. I liked to think that the number of possible permutations would probably pay back some of what I lost in randomness.
(I now fancy wallpapering my cubicle at work using sheets like this, with characters randomly colored for additional visual cues.)
What you can (and probably should) do is to regulate its use in any of these fields as that field seees fit (or not regulate at all, as the case may be). When it is to be used in medicine, regulate it as a medical technology.
Software is already regulated in some fields, such as when it is part of a medical device. See, for instance, FDA-imposed design controls on medical devices "automated with computer software" in 21 CFR 820.30. FDA has stated that "Software must be validated when it is a part of the finished device. FDA believes that this control is always needed, given the unique nature of software, to assure that software will perform as intended and will not impede safe operation by the user." (in their final rule on that "Quality System Regulation"). The regulations call for extensive documented verification and validation activities.
Slashdot Mirror
Not sure where you are going with the anti-virus, since Microsoft has never released one.
They have. There was a utility called MSAV in MS-DOS 6.
[...] most of the history of countries comes from the CIA Factbook, the US State Department, or [...]
There is also another explanation, other than an intended bias, for why material from U.S. federal departments and agencies would be widely used in Wikipedia: U.S. federal documents and publications are in public domain and so excerpts from them can be included in Wikipedia articles without copyright worries.
English skills: worse
The effect can also be the opposite for those who are not native speakers of English.
David Pogue has also authored a specific Missing Manual "Switching to the Mac." The description in the parent post would apply perfectly to it as well.
> So the machines themselves do not have any DICOM interfaces?
In this particular case, from the manufacturer's point of view, the console (that Windows workstation) is part of the MR scanner, and the part that represents the scanner towards the hospital network. Embedded control and data acquisition systems send data to the console over direct fiber-optical links using a proprietary protocol. That raw data is not clinically usable until it has been reconstructed into images on the console. After reconstruction the images can be displayed, printed, or transferred using DICOM.
Obviously some other vendors might have designed their systems differently. But I have seen some other MRI systems that in a similar way used separate reconstruction computers that appeared much like standard PC hardware. Quick reconstruction of MR images takes a lot of processing power and my hunch is that high-end off-the-shelf workstations (Windows or Unix) provide that power relatively cheap.
> The devices themselves DO NOT RUN ANY OS. The devices
> are medical equipment that take pictures and send them
> to a "server," basically.
"Taking pictures" is not that simple in CT and MRI, because the raw output you get is not readable like the final image. In MRI, the raw output is in frequency domain and typically a Fast Fourier Transform is needed to bring it to spatial terms. The systems I know best (Philips low-field open MRI) currently use for image reconstruction HP workstations that run Windows 2000. The same workstation is the MR scanner operator's console. Without that computer operational, no scan can be configured or images produced.
I hear that some detailed proposals for "abstract communication bootstrap procedures" have been made. One such is Dr. Hans Freudenthal's LINCOS (short for "Lingua cosmica") that he published as a book in 1960.
I have sometimes used a printed table to aid myself in memorizing a pseudorandom password, too. I did it by printing a table of random characters from the set of lower and upper case letters, numbers, and some punctuation, like this:
0 3 x C 6 m c Q 5 q u s8 e v 7 u K T / W 8 4 1
6 j B y . 8 o r = 8 S 5
O F v L 4 g 3 4 p I W 6
c l B P E u Z 9 6 L y 5
% p U A a 9 % d 5 A H v
J e % ! C 3 b . D U 5 U
Q O S l t J Q E P r c L
P 4 g n a S 9 9 C R b 7
% 9 x E = 5 d i o l 8 G
R h Q Q A e o y x R 9 Z
R E 3 N 8 c A e I 7 0 d
and then deciding from where in that table to read a password. Obviously the password could not be in a straight row. It could be a spiral around an initial character, part of a knight's tour, alternating picks from several lines, characters at intervals based on the Fibonacci sequence, or whatever rule one could devise. This effectively replaced remembering the password by remembering a pattern. I liked to think that the number of possible permutations would probably pay back some of what I lost in randomness.
(I now fancy wallpapering my cubicle at work using sheets like this, with characters randomly colored for additional visual cues.)
Software is already regulated in some fields, such as when it is part of a medical device. See, for instance, FDA-imposed design controls on medical devices "automated with computer software" in 21 CFR 820.30. FDA has stated that "Software must be validated when it is a part of the finished device. FDA believes that this control is always needed, given the unique nature of software, to assure that software will perform as intended and will not impede safe operation by the user." (in their final rule on that "Quality System Regulation"). The regulations call for extensive documented verification and validation activities.