Note it is not the owner's manual. Service manuals are often not public information. Companies sometimes sell them too. So reproduction in that context would violate copyright.
Please refrain from writing about something you know nothing about. Whether or not something is "public information" has nothing to do with copyright law. It may have something to do with the law involving trade secrets, but that's another matter.
Strange. A simple round-robin scheduler with no priorities and no possibility of interrupting a time slice or anything else fancy should be quite capable of running multiple communicating proceses, as long as they use normal blocking IO. Are you sure that there were no race conditions ?
Perhaps it was. It was a few years ago, and I don't have access to the source code anymore. It was a fairly simple program, and I was already aware of the notion of race conditions, so I doubt it was anything specifically in my code that was resulting in a race condition. On the other hand, the code was written in Python and there was some OpenGL app running in the background (and I think there was a fair amount of filesystem access as well) so there may have been problems there. However, what I do remember is that processes were being starved under heavy load in Win2K.
I don't do much work in Windows anymore, so for all I know, it may have been fixed in recent service packs. laffer1's experience just reminds me of the problems I had.
What's the "background services" mode supposed to do anyway ?
From what I've heard (and my experiences seem to confirm it), when you have the thing set to "Applications" mode, Windows gives something like 3 times as many time slices to whatever it considers to be the "foreground" application.
Guess what OSX = lower ping for WoW. Microsoft needs to overhaul there IP stack again if you ask me. I love linux or OSX for gaming simply because my ping is always better.
My guess is that it's the scheduler. At least on Windows 2000, the thing was so broken (even in optimize-for-"Background Services"-mode) that I had to re-write a multi-process application to use a sigle process with multiple threads, because producer-consumer was just too much for the OS to handle. I think the last time Linux was so bad, it was considered a bug and fixed within a few releases.
Cross-platform apps tend to be more affected, because they haven't been designed specifically to work around Microsoft's brain damage.
That better not be true... Since ANSI C says that NULL is 0.
I don't know about ANSI, but ISO/IEC 9899:1999(E) (a.k.a. "C99"), under section 7.17 "Common definitions <stddef.h>" states:
1 The following types and macros are defined in the standard header <stddef.h>. Some are also defined in other headers, as noted in their respective subclauses.
...
3 The macros are
NULL
which expands to an implementation-defined null pointer constant;
Under section 6.3.2.3 "Pointers", the "null pointer constant" is defined as follows:
3 An integer constant expression with the value 0, or such an expression cast to type void *, is called a null pointer constant. If a null pointer constant is converted to a pointer type, the resulting pointer, called a null pointer, is guaranteed to compare unequal to and object or function.
Indeed. Wavers are more useful as legal evidence that you were made aware of the risks of some activity, rather than an explicit disclaimer of liability. (At least in cases of criminal negligence. I don't know about the rest.)
Slashdot Mirror
RUFF_ILB: My name is Borg Borg.
VIKINGS: Borg, Borg, Borg, Borg, lovely Boooorg...
Really? Where? I see it neither in TFA, nor in the manual linked to in this /. comment.
Please refrain from writing about something you know nothing about. Whether or not something is "public information" has nothing to do with copyright law. It may have something to do with the law involving trade secrets, but that's another matter.
The Internet General Public Disclaimer, Version 1.
Never mind. Found it, I think (unless it's different elsewhere).
What is 711 normally for?
Perhaps it was. It was a few years ago, and I don't have access to the source code anymore. It was a fairly simple program, and I was already aware of the notion of race conditions, so I doubt it was anything specifically in my code that was resulting in a race condition. On the other hand, the code was written in Python and there was some OpenGL app running in the background (and I think there was a fair amount of filesystem access as well) so there may have been problems there. However, what I do remember is that processes were being starved under heavy load in Win2K.
I don't do much work in Windows anymore, so for all I know, it may have been fixed in recent service packs. laffer1's experience just reminds me of the problems I had.
What's the "background services" mode supposed to do anyway ?
From what I've heard (and my experiences seem to confirm it), when you have the thing set to "Applications" mode, Windows gives something like 3 times as many time slices to whatever it considers to be the "foreground" application.
/me hits Ctrl-Shift-S, and the site becomes readable
I win!
My guess is that it's the scheduler. At least on Windows 2000, the thing was so broken (even in optimize-for-"Background Services"-mode) that I had to re-write a multi-process application to use a sigle process with multiple threads, because producer-consumer was just too much for the OS to handle. I think the last time Linux was so bad, it was considered a bug and fixed within a few releases.
Cross-platform apps tend to be more affected, because they haven't been designed specifically to work around Microsoft's brain damage.
JMHO
brltty, you insensitive clod!
Priceless.
Bah. I won't believe it until I see it on Fox News!
If you've used Linspire, you'll see that there's no contradiction there.
Speaking of CDATA, how do you represent the string "]]>" inside a CDATA block?
Never mind. It looks like I just did the same thing. :-/
Why is it that when you see a reply to a message on Slashdot, you assume the person disagrees?
Bah. That should read, "any object or function."
I don't know about ANSI, but ISO/IEC 9899:1999(E) (a.k.a. "C99"), under section 7.17 "Common definitions <stddef.h>" states:
Under section 6.3.2.3 "Pointers", the "null pointer constant" is defined as follows:With some cleverness, they can still make your decisions for you, even if your communication is encrypted, if it's not also authenticated.
It probably should be:
or
Still, everyone has the freedom to hire (or convince) a programmer to fix the problem.
BSD License: RTFL
GNU GPL: RTFL
...using military-grade cryptography...
Indeed. Wavers are more useful as legal evidence that you were made aware of the risks of some activity, rather than an explicit disclaimer of liability. (At least in cases of criminal negligence. I don't know about the rest.)
Bullshit. It wasn't true when RMS wrote the GNU Manifesto, and it isn't true now.
since it would all be freely available.
Again, that is a steaming pile of non sequitur crap. Reality dictates that people are paid to work on free software. Deal with it.