...it doesn't specify a version and there is no COPYING file or mention of a license in any of the files or source code.
<IANAL>Then it has no license at all, and you are committing copyright infringement. I suggest you stop distributing your project until you've contacted the copyright holder of GladeGen and gotten him to properly license it.</IANAL>
They only simplify stuff if you don't count the complexity of the macro itself. On the balance, I don't think they're worth it except possibly in library code that Somebody Else gets to maintain. And even then I think the use of function-like macros should be minimized because a consumer of the library might not necessarily realize the function is implemented as a macro (and therefore could have unintended consequences if its arguments have side effects, e.g. FOO(arg++) expands to { (arg++) * (arg++) })
This shows your ignorance (and that of the general slashdot population). The "auto space like Word 95" issue has been addressed in the latest spec (the spec that's beeen approved). That "auto space like Word 95" behavior, and the others like it, are now marked as "deprecated" (i.e. should not use for new documents) AND are fully spec'ed.
I see links to a blog. Where are the links to the spec? Proof or STFU!
Well, the C programs I've written to date haven't needed to handle large data sets or run super-efficiently, so it hasn't been a problem. Besides, it's no worse than the case of something like LISP or Java, where everything is allocated on the heap...
...even though there was a vast outpouring of bile from the community when the draft was published... Of course, it's not really possible to write a C99 compliant compiler as the the standard mandates behavior that is sometimes either completely impossible or just completely undesirable.
What's wrong with C99? (Note: I'm curious, not argumentative.)
their Windows monopoly is crumbling down, thanks to the Vista fluke.
Every version of Windows except 95 and 2000 have been as poorly received as Vista when they first came out. It's not a fluke, and it's not evidence of impending Microsoft collapse! I wish it were, but it's not!
Wine is getting better every moment, and while ReactOS isn't exactly around the corner, in 5 years it'll be on par with WINE - with 2013's WINE (ReactOS and WINE share a lot of code).
WINE?! Don't you realize that WINE is irrelevant? Sure, maybe in 2013 WINE and/or ReactOS might be good enough to run all Win32 and MFC software. But it won't matter, because Microsoft already moved the goalposts to newer proprietary APIs that are patented to boot!
Even if OOXML becomes an ISO standard, that doesn't mean we're obligated to use it.
Of course! What makes us obligated to use it is Microsoft's monopoly. It's just that the rejection of OOXML as an ISO standard would have freed us from that monopoly, and now it won't.
Realize that in terms of CPU operations, having a language track every begining[sic] and end of every separately allocated memory space is very expensive.
Yeah, but on the other hand, the allocator itself does keep track of it. Although I agree it would be horribly inefficient to add automatic bounds checking, I think it'd be nice to have a (standardized) function called something like size_t lengthof(void *) that, when passed a pointer to any address within a dynamically-allocated buffer, would return the number of bytes from the location given to the end of the buffer or return 0 and set errno if passed a pointer to any address not within a dynamically-allocated buffer.
(Ideally, I'd also deprecate the use of sizeof() on arrays and define lengthof() to be valid on static arrays too, but that wouldn't be possible because too many people would insist on backwards compatibility.)
Personally, when I'm coding C I just don't use arrays (except string literals) at all and instead dynamically allocate everything that's not a scalar. The mechanical complexity of having to keep track of more pointers is worth the conceptual simplicity of having only one kind of thing to worry about instead of two.
I'm continually amazed that the CS majors at my major US accredited university can't program using pointers, object oriented techniques, or parallelism.
I'll bet just about every CS major can do one of those things... just not all three. And I'd add "understand algorithmic complexity" to the list, and suggest that the best you could hope for from all but the most exceptional students is two out of four.
For Heads of State, I think you'll find that the vast majority of Heads of State have MBAs or JDs (or BLs) in developed nations, or are the children of political families in lesser-developed nations, or are former warlords in even less developed nations.
Or they might even be engineers too! The guy that runs Iran has a degree in civil engineering...
Check the historic share price of the company. That tells you the confidence of people investing their money have in the company, which is diminishing and completely stationary as off recently (i.e: company is losing value in real terms).
What do you mean, "recently?" MSFT has been flat since 2000! Heck, it probably would have even been flat since mid-1998, except for the bubble (note the part where the NASDAQ diverges from the Dow, and MSFT's price before and after that interval).
Retailers are staying with XP for as long as practically possible, some PC manufacturers offer "upgrades downwards" to XP.
And they did exactly the same thing when going from Win2K->XP, and NT4->2K, and...
MS is innovating zilch
So what? Microsoft has never innovated!
I think you need to get some perspective: what you claim to be signs of impending doom for Microsoft are actually exactly how it's always been! Now, I'm no MS fanboy; I'd like to see that evil piece of shit implode just as much as you do. But I'm not going to delude myself into thinking it'll happen any time soon.
Maybe I'm missing the picture here but given the inner workings of both the XBox and the PS3, their PC-like peripherals (sans mouse), their network-ability and the mod-ability of both into Linux systems, I would argue that console gaming is dead. The only problem with that argument is that the Wii (as the only real console left) is doing pretty damned well.
Nope, here's the difference:
Consoles are locked down and run only proprietary, manufacturer-approved games, while PCs are open and free to develop for. Modchips and Linux don't count, because they are illegal or don't have access to all the hardware, respectively.
If PC -- i.e., free and open gaming dies, it'll be a sad, sad day.
Dump all the pixelvalues as arrays into a screenshot bypassing Windows, then stream together the screenshots in a video format of your choice, and you've got uncompressed, perfect digital video.
No, you don't. It's uncompressed, but not "perfect" because it still has the compression artifacts. Then, when you recompress it, it has two sets of compression artifacts. Although it's higher quality than aiming a video camera at the display, it's still more-or-less the same as the "analog hole."
To really count as "cracking," the attacker needs to get access to the decrypted but still encoded stream.
But there is a vicious catch 22 problem if the employees are willing to work only 18 months then the company is not going to invest in them just to have them higher skilled to work for a competitor. In them olden days people stayed with the company and didn't jump on any offer even if it was more appealing.
You've got the cause and effect backwards: first employers started failing to invest in their employees, and then the employees started looking out for their own best interests (by jumping on other offers) because of it.
Slashdot Mirror
<IANAL>Then it has no license at all, and you are committing copyright infringement. I suggest you stop distributing your project until you've contacted the copyright holder of GladeGen and gotten him to properly license it.</IANAL>
They only simplify stuff if you don't count the complexity of the macro itself. On the balance, I don't think they're worth it except possibly in library code that Somebody Else gets to maintain. And even then I think the use of function-like macros should be minimized because a consumer of the library might not necessarily realize the function is implemented as a macro (and therefore could have unintended consequences if its arguments have side effects, e.g. FOO(arg++) expands to { (arg++) * (arg++) })
Buddy, just be glad it's not rasterized!
I see links to a blog. Where are the links to the spec? Proof or STFU!
Well, the C programs I've written to date haven't needed to handle large data sets or run super-efficiently, so it hasn't been a problem. Besides, it's no worse than the case of something like LISP or Java, where everything is allocated on the heap...
What's wrong with C99? (Note: I'm curious, not argumentative.)
Every version of Windows except 95 and 2000 have been as poorly received as Vista when they first came out. It's not a fluke, and it's not evidence of impending Microsoft collapse! I wish it were, but it's not!
WINE?! Don't you realize that WINE is irrelevant? Sure, maybe in 2013 WINE and/or ReactOS might be good enough to run all Win32 and MFC software. But it won't matter, because Microsoft already moved the goalposts to newer proprietary APIs that are patented to boot!
Of course! What makes us obligated to use it is Microsoft's monopoly. It's just that the rejection of OOXML as an ISO standard would have freed us from that monopoly, and now it won't.
I'd say that eliminating the actual occurrence of impropriety ought to be their priority, not the mere appearance of it!
Exactly! I've said that to him myself on many an occasion... he's, so far, never replied.
And that's Twitter's duty as well. Sock puppets don't count as being "honest" or "responsible!"
Yeah, but on the other hand, the allocator itself does keep track of it. Although I agree it would be horribly inefficient to add automatic bounds checking, I think it'd be nice to have a (standardized) function called something like size_t lengthof(void *) that, when passed a pointer to any address within a dynamically-allocated buffer, would return the number of bytes from the location given to the end of the buffer or return 0 and set errno if passed a pointer to any address not within a dynamically-allocated buffer.
(Ideally, I'd also deprecate the use of sizeof() on arrays and define lengthof() to be valid on static arrays too, but that wouldn't be possible because too many people would insist on backwards compatibility.)
Personally, when I'm coding C I just don't use arrays (except string literals) at all and instead dynamically allocate everything that's not a scalar. The mechanical complexity of having to keep track of more pointers is worth the conceptual simplicity of having only one kind of thing to worry about instead of two.
I'll bet just about every CS major can do one of those things... just not all three. And I'd add "understand algorithmic complexity" to the list, and suggest that the best you could hope for from all but the most exceptional students is two out of four.
C macros hardly ever simplify stuff, even in C!
Multi-touch and parallelism are both the "next big thing," because multiple touches are touches in parallel!
; )
I don't get it. I can tell they look different, but they both look equally good. I guess #2 is a little sharper; is that the GIMP one?
Or they might even be engineers too! The guy that runs Iran has a degree in civil engineering...
What do you mean, "recently?" MSFT has been flat since 2000! Heck, it probably would have even been flat since mid-1998, except for the bubble (note the part where the NASDAQ diverges from the Dow, and MSFT's price before and after that interval).
And they did exactly the same thing when going from Win2K->XP, and NT4->2K, and...
So what? Microsoft has never innovated!
I think you need to get some perspective: what you claim to be signs of impending doom for Microsoft are actually exactly how it's always been! Now, I'm no MS fanboy; I'd like to see that evil piece of shit implode just as much as you do. But I'm not going to delude myself into thinking it'll happen any time soon.
Nope, here's the difference:
Consoles are locked down and run only proprietary, manufacturer-approved games, while PCs are open and free to develop for. Modchips and Linux don't count, because they are illegal or don't have access to all the hardware, respectively.
If PC -- i.e., free and open gaming dies, it'll be a sad, sad day.
A rhetorical question is still a question!
Sorry, this is a pet peeve of mine:
When will you learn that questions should end with a question mark?
No, you don't. It's uncompressed, but not "perfect" because it still has the compression artifacts. Then, when you recompress it, it has two sets of compression artifacts. Although it's higher quality than aiming a video camera at the display, it's still more-or-less the same as the "analog hole."
To really count as "cracking," the attacker needs to get access to the decrypted but still encoded stream.
You've got the cause and effect backwards: first employers started failing to invest in their employees, and then the employees started looking out for their own best interests (by jumping on other offers) because of it.
So put a fucking paper shredder next to the exit! It's not that fucking hard!