An interface should capitalize on the abilities that most users have, but still accommodate users with less ability. If it required 10 digits, it'd lock out anyone without them. If it worked more efficiently if you had 10 digits, on the other hand (well, both hands)...
it's not easy to just move your finger by one pixel
When I used a trackpad on my desktop for a while, I found it easy to move one pixel at a time, by rolling my finger tip around (adjusting the angle of the finger with respect to the pad), without moving the tip around at all. The result is precise adjustment of the area contacting the pad.
The default setting for an equipment that can be lethal should be "Emit zero radiation". Then for each exposure, set the level of radiation you intend to use. This way, you ALWAYS KNOW the level of radiation the equipment will emit.
How about requiring them to type the radiation value as words AND numbers? So "fifty five millirems" and "55 mrem" (or whatever the units are). But then you run into the problem, like with Windows asking too many questions: users tire of the verbosity, and find ways of avoiding it. Verbosity also hides important information in the sea of noise. Maybe they'd use copy-and-paste here to speed it up.
Here, it seems a good approach would be to display the dosage in multiple ways, like a color or shape, so that the wrong setting would be more likely noticed, and look different than the right one. A doctor has in mind the dose that he intends, and would come to expect a certain color or shape. Seeing one that looks very different would be jarring.
There are some excellent technical books, but why buy books you're going to read only once when you can borrow them from the library? (it helps to live in a city with a large university library) I guess you implied as much.
No, it's true. Once I pirated a Sony music CD and got a really awful rootkit. No way a company would intentionally put that on the CD, so it must have been added by evil pirates.
With technical writing, everyone and their mother thinks they know how to write.
Tell me about it! Just an example of the latest of badly-written books, where the author can't decide how to consistently refer to himself and the reader:
This book focuses on the Java programming language and uses Java examples throughout. It is assumed that the reader has at least an intermediate understanding of Java (and a working Java system if you want to try out the examples for yourself). Example code and other support material is available at my website.
"It is assumed" refers to himself somewhat indirectly, but he later refers to "my website". He refers to the reader as "the reader" and "you", implying that there is another audience. The book is full of this kind of thing, where he switches between indirect references to himself, "I" to refer to himself, "we", "the reader", and "you".
This is closely related to writing tests first, but isn't limited to writing tests. You can make assumptions during implementations as well. By making assumptions as we work [...]
Remember one thing about LGPL. It requires that that user can replace the LGPL binary. So Qt released under LGPL must be a separate dll, no static linking.
Or you could provide the source code for your program, or perhaps the.o files (if you like angering your users).
I can just imagine it... "Sorry sir, there was fine print there, right inside that little dot; perhaps your eyesight is such that you'll need a Hitachi TM-1000 electron microscope. The price has really come down recently."
It's not extra-sensory perception unless the human us sensing things outside the normal perception channels. From the article, it sounds like this is just another input device for a computer to be controlled by a human.
And the title of the article, "Communicating person to person through the power of thought alone", is false, since this thing wouldn't work without electricity. By the same logic, I'm communicating with Slashdot readers right now by the power of thought alone, well of course with the help of food energy, muscles, a keyboard, the Internet, etc.
OK, so the problem is the government requiring companies like AT&T to even acknowledge the existence of these "audio text" businesses who take advantage of government regulations. The solution is less government regulation (allow AT&T to give these dishonest companies the finger), not more (force Google to submit to their fraud as well).
You are aware that those assumptions make your code inherently unportable, aren't you?
Yes. The point was that we already have enough types to assign to the popular integer sizes, as long as we don't assign two to the same size, as I believe Windows 64-bit does (long and int both being 32 bits).
Not only long, but assuming that int has 32 bits. I agree it may be misguided, but int is defined to be (usually the most efficient) integer data type with at least 16 bits.
Yes, int need only be 16 bits (and signed int need only support the range -32767 to +32767, i.e. sign-magnitude). Software which compiles OK but fails at run-time for 16-bit int is badly coded, since it's trivial to check the in size:
#if UINT_MAX < 0xFFFFFFFF #error "int must have at least 32 bits" #endif
I do this nowadays in my code, since supporting 16-bit int is a real pain. Implicit int occurs in lots of obscure places, and it's hard to even find platforms with 16-bit int to test on. I used to support 16-bit int, and my code was littered with casts to long, and use of long for function arguments. When compiled on Linux, these longs would then be 64 bits, which was wasteful of memory. Having switched to requiring 32 or more bits in an int, I haven't looked back nor regretted the decision.
The practical argument is never answered. But the idealogical one is simple: why should the government be able to regulate (or threaten to regulate) what a company does with its own property? Correct me if I'm wrong, but the radio spectrum is big enough for tens of cellphone networks, and each network already pays for the part of the spectrum they use. The argument would be different if it were physical infrastructure, where having ten network providers and ten sets of cables to every house would clutter everything up.
Unfortunately, Abubakr's arrangement means that the table can only be read by rotating it. That's tricky with a textbook and impossible with most computer screens.
Take a look at the table. Are there really people who can't read it without rotating it?
Try rolling the old table. It's like a square wheel! This new one is round, so it rolls very well. I'd say their "re"-invention of the wheel is a vast improvement.
BitTorrent uses a different checksum algorithm than TCP/IP, one that's probably not nearly as thorough. Plus the fact that the BitTorrent checksum would have been verified on the receiving end, likely by reading the data from memory after the entire block/file was received. Are you suggesting that it got corrupted in the short time between being verified and written to the hard drive (or read from the hard drive and sent)?
Slashdot Mirror
That's amazing. I've got the same combination on my luggage!
Sounds like I don't want an OS. What is the equivalent of single? Booting to a Forth prompt?
An interface should capitalize on the abilities that most users have, but still accommodate users with less ability. If it required 10 digits, it'd lock out anyone without them. If it worked more efficiently if you had 10 digits, on the other hand (well, both hands)...
When I used a trackpad on my desktop for a while, I found it easy to move one pixel at a time, by rolling my finger tip around (adjusting the angle of the finger with respect to the pad), without moving the tip around at all. The result is precise adjustment of the area contacting the pad.
How about requiring them to type the radiation value as words AND numbers? So "fifty five millirems" and "55 mrem" (or whatever the units are). But then you run into the problem, like with Windows asking too many questions: users tire of the verbosity, and find ways of avoiding it. Verbosity also hides important information in the sea of noise. Maybe they'd use copy-and-paste here to speed it up.
Here, it seems a good approach would be to display the dosage in multiple ways, like a color or shape, so that the wrong setting would be more likely noticed, and look different than the right one. A doctor has in mind the dose that he intends, and would come to expect a certain color or shape. Seeing one that looks very different would be jarring.
So require that the calculator be verified to contain the original OS before allowing it into the test. There are many ways to do this.
There are some excellent technical books, but why buy books you're going to read only once when you can borrow them from the library? (it helps to live in a city with a large university library) I guess you implied as much.
A high-powered halogen lamp should do the trick, 200-300W. Be careful of starting fires, though.
No, it's true. Once I pirated a Sony music CD and got a really awful rootkit. No way a company would intentionally put that on the CD, so it must have been added by evil pirates.
Tell me about it! Just an example of the latest of badly-written books, where the author can't decide how to consistently refer to himself and the reader:
"It is assumed" refers to himself somewhat indirectly, but he later refers to "my website". He refers to the reader as "the reader" and "you", implying that there is another audience. The book is full of this kind of thing, where he switches between indirect references to himself, "I" to refer to himself, "we", "the reader", and "you".
Switched from referring to "you" and "we".
And we thank you for it!
Or you could provide the source code for your program, or perhaps the .o files (if you like angering your users).
I can just imagine it... "Sorry sir, there was fine print there, right inside that little dot; perhaps your eyesight is such that you'll need a Hitachi TM-1000 electron microscope. The price has really come down recently."
It's not extra-sensory perception unless the human us sensing things outside the normal perception channels. From the article, it sounds like this is just another input device for a computer to be controlled by a human.
And the title of the article, "Communicating person to person through the power of thought alone", is false, since this thing wouldn't work without electricity. By the same logic, I'm communicating with Slashdot readers right now by the power of thought alone, well of course with the help of food energy, muscles, a keyboard, the Internet, etc.
OK, so the problem is the government requiring companies like AT&T to even acknowledge the existence of these "audio text" businesses who take advantage of government regulations. The solution is less government regulation (allow AT&T to give these dishonest companies the finger), not more (force Google to submit to their fraud as well).
You have to speak their native language, with the help of your wallet.
Yes. The point was that we already have enough types to assign to the popular integer sizes, as long as we don't assign two to the same size, as I believe Windows 64-bit does (long and int both being 32 bits).
Yes, int need only be 16 bits (and signed int need only support the range -32767 to +32767, i.e. sign-magnitude). Software which compiles OK but fails at run-time for 16-bit int is badly coded, since it's trivial to check the in size:
I do this nowadays in my code, since supporting 16-bit int is a real pain. Implicit int occurs in lots of obscure places, and it's hard to even find platforms with 16-bit int to test on. I used to support 16-bit int, and my code was littered with casts to long, and use of long for function arguments. When compiled on Linux, these longs would then be 64 bits, which was wasteful of memory. Having switched to requiring 32 or more bits in an int, I haven't looked back nor regretted the decision.
We already have the types to handle it:
char: 8 bits
short: 16 bits
int: 32 bits
long: 64 bits
long long: 128 bits
Except that Microsoft keeps long at 32 bits, for buggy code that assumes its size (on Linux, long is 64 bits, as it should be).
And of course there's <stdint.h>, where you could just put int128_t.
In my 128-bit OS, the volume goes all the way to 340,282,366,920,938,463,463,374,607,431,768,211,455.
You can't tell from the photo, but that screw has a rootkit built in! If you had used a normal screw, you'd be deprived of getting screwed twice.
And I have a real C programmer's password:
012345&*M%HJOJNVFGPLkoPWHJrcp,k0cY$PO JO9 P[-97 YTJJY93528 [SIGSEGV detected]
The practical argument is never answered. But the idealogical one is simple: why should the government be able to regulate (or threaten to regulate) what a company does with its own property? Correct me if I'm wrong, but the radio spectrum is big enough for tens of cellphone networks, and each network already pays for the part of the spectrum they use. The argument would be different if it were physical infrastructure, where having ten network providers and ten sets of cables to every house would clutter everything up.
I don't get this comment from the article:
Take a look at the table. Are there really people who can't read it without rotating it?
Try rolling the old table. It's like a square wheel! This new one is round, so it rolls very well. I'd say their "re"-invention of the wheel is a vast improvement.
BitTorrent uses a different checksum algorithm than TCP/IP, one that's probably not nearly as thorough. Plus the fact that the BitTorrent checksum would have been verified on the receiving end, likely by reading the data from memory after the entire block/file was received. Are you suggesting that it got corrupted in the short time between being verified and written to the hard drive (or read from the hard drive and sent)?