Depends on MS's purpose in hiring them. It may be that as long as MS can keep the best talent from working for somebody else, it doesn't matter whether they actually produce anything at all.
Welcome to the wonderful world of software patents.
You should take it up with Apple, or perhaps hit up the maker of some Linux distribution and say you are willing to pay extra to Apple to get the full version of FreeType, or that you want an installation option that will let you pay Apple in return for installing the version of FreeType that does the hinting right.
For all I know IBM may have people working on massive parallelism--but small and efficient is good whether we get good at parallelism or not. At the very least you're going to want small (and energy efficient) hardware to put together that mondo Beowulf cluster, aren't you?
Well, it depends. Analysts do it continuously. Algebraists do it in groups. Topologists...I don't want to think about that. (Strong deformation retract? Ouch!)
Polluters love tourt reform since its expensive to prove that the pollants they discharge cause damage and to prove it would cost tens of millions of dollars.
<sarcasm>And we can't let trivia like proof get in the way of sticking it to the "mega corps," can we?</sarcasm>
Sure, it's probably taken me longer to write this post, than it took to find the php code I used as a basis for the search, but how much math is REALLY needed overall?
Lord Kelvin put it best (though the notebooks of Lazarus Long come darned close):
I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot express it in numbers, your knowledge is of a meager and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely, in your thoughts, advanced to the stage of science, whatever the matter may be.
Lord Kelvin
Anyone who doesn't know at least some basic math and statistics is a sucker for all the fallacies pushed by advertisers, politicians, and groups with an agenda. And once you've done your Google search for a formula to plug and chug, how do you test it if you don't know enough math to really understand it?
KDE has always included kappfinder, which searched the system for applications to add to the menu. If you not seeing them there now, then perhaps your distro is no longer running kappfinder during install. So go run it manually.
Just did which kappfinder on a box running RH 8.0; couldn't find it, and man kappfinder turned up nothing.
OK. I find now that xmms is behaving strangely, not appearing on the display and not playing the sounds. OK, time to look at what version of xmms I have, maybe dump it if I have an older version...but RH 8.0 hides the nice GNOME and KDE GUI layers around rpm from me, and the "packages" menu item it now gives one leads to a dumbed-down program that is fine for people used to the Windows "Install/Remove Program" menu item on the Control Panel, but which will not give me the information I need and that the previous GNOME and KDE programs for package management made trivial.
Well, almost. I didn't delete my.kde directory; just deleted the icons from the panel and then put their moral equivalents back. (Before doing that, they were all just some generic icon rather than something specific to what they ran, and clicking on them got a message about a file in/usr/share/applnk not existing. Looking around shows that directory to be pretty well emptied of nondirectory files, save for things that correspond to programs I installed apart from an RH install/upgrade, e.g. Opera.)
Do you know what killed OS/2? No, it wasn't IBM's marketing. It was lack of Win32 compatibility, as well as lack of hardware compability.
That lack of compatibility was urged along by the successive versions of win32s.dll, which kept breaking compatibility. IBM finally gave up, and MS stopped changing win32s.dll, after version 1.30 of win32s.dll came out, which added a call to request memory out of range of the 512 MB limit on DOS sessions under OS/2--a fairly fundamental thing to break, and a limit only recently, long after it would make much difference, eliminated. There's no real reason to be picky about the base address of memory one allocates--so it was added solely to break OS/2 compatibility. Gee...you just claimed that MS couldn't possibly do something like that. Looks like a counterexample to me.
Speaking of breaking existing software--checked MS's EULA lately? To get security fixes, you have to agree that MS has the right to automatically install OS upgrades that may prevent software from running. So nowadays, with Windows you can either live with security holes or accept "DOS isn't done until Lotus won't run" as SOP.
Compatibility is a Catch-22. Without it, people kvetch about not having it. With it, developers see no reason to bother to write for anything but Windows--and indeed, you complain about both the lack of OS/2 native software and the lack of compatibility.
That's not to say that you value the Sopranos, but I want to make it perfectly clear that ALL cultures glamorize the criminal underworld. Both portray a glamorized, clean-cut interpretation of seedy underworlds...
Got that right. All you have to do is look at the works of Damon Runyan, or to go back a long way, Francois Villon. One of Mario Pei's books starts a chapter on jargon with a rendering of a Villon poem in the corresponding twentieth-century American theives' jargon:
You yeggs that pull a real good heist
And swipe the moolah from a square, Watch out for finks that dummy up Until The Man gets in their hair.
That's been going on for a long time. Remember the "Great Folk Music Scare"? All those clean-cut college types singing about miners and migrant farmers and their working conditions.
"When was the last time you ever dug a ditch, baby?" -- Oingo Boingo, "Capitalism"
Thanks, I've read the book of Job. The Cliff Notes version: God kills and/or tortures Job and his family to settle a bar bet with Satan. When Job complains, God says, a la Chevy Chase, "I'm God, and you're not. Shut up, I can do what I want."
If an earthly father did that, he'd be in the slammer so fast your head would spin. Such a being is a vile beast.
If the Christian God exists, then yes, he must in some way want everyone going to hell to be there, because he established the initial conditions of the world, and being omniscient, knew exactly what would happen, and being omnipotent, could do exactly what he wanted.
In everyday life, foreknowledge and action add up to responsibility. If I set a shiny Ginsu knife down in front of an infant and it manages to kill itself, I'm the one hauled into prison. How much more responsibility must be borne by an omniscient, omnipotent god!
Ah...so for our own good, we are denied knowledge that would prevent us from eternal torment? I'm sorry, but any deity who thinks that's for my good is a monstrous, sadistic beast in comparison with which all the mass murderers, child abusers, and other criminals of earth pale to insignificance.
The exact wording was "In retrospect, it was a mistake to announce an agreement in principle with respect to the curriculum initiatives, a mistake for which I take responsibility." If I were in conspiracy theory mode, or even in Clintonspeak parsing mode, I'd be thinking that he wanted to hold off on publicity until it was a fait accompli.
Well, yeah, but...the conditional execution bits eat four bits out of every 32-bit instruction, which makes the constraints on immediate operands and offsets for base + offset addressing even more galling than usual on fixed-width instruction sets, for a payoff of uncertain value unless all you do is run the clever GCD code people always give as an example of how great ARM conditional execution is. Among the most perverse aspects are offsets whose ranges vary with the type of data you're loading or storing, the set of permissible immediate values (an eight-bit value shifted by an even number of bits), and an obnoxious constraint on operands of multiply instructions.
Of course, writing code for the ARM is heaven in comparison with writing x86 code, but I would not say it is all that spiffy in comparison with 68xxx or PowerPC.
Slashdot Mirror
Depends on MS's purpose in hiring them. It may be that as long as MS can keep the best talent from working for somebody else, it doesn't matter whether they actually produce anything at all.
Welcome to the wonderful world of software patents.
You should take it up with Apple, or perhaps hit up the maker of some Linux distribution and say you are willing to pay extra to Apple to get the full version of FreeType, or that you want an installation option that will let you pay Apple in return for installing the version of FreeType that does the hinting right.
The trick is, get hired by the outsourcee, not the outsourcer...
I'm not sure I'm willing to move to a Third World country...
For all I know IBM may have people working on massive parallelism--but small and efficient is good whether we get good at parallelism or not. At the very least you're going to want small (and energy efficient) hardware to put together that mondo Beowulf cluster, aren't you?
Well, it depends. Analysts do it continuously. Algebraists do it in groups. Topologists...I don't want to think about that. (Strong deformation retract? Ouch!)
Polluters love tourt reform since its expensive to prove that the pollants they discharge cause damage and to prove it would cost tens of millions of dollars.
<sarcasm>And we can't let trivia like proof get in the way of sticking it to the "mega corps," can we?</sarcasm>
OK. If one does that, what happens when one clicks on a mailto: link? Can it be configured to start one's favorite mailer?
Lord Kelvin put it best (though the notebooks of Lazarus Long come darned close):
Anyone who doesn't know at least some basic math and statistics is a sucker for all the fallacies pushed by advertisers, politicians, and groups with an agenda. And once you've done your Google search for a formula to plug and chug, how do you test it if you don't know enough math to really understand it?
Just did which kappfinder on a box running RH 8.0; couldn't find it, and man kappfinder turned up nothing.
I'll answer my own question. Yes, the new draft standard includes EQUIVALENCE.
I guess FORTRAN compiler writers need to have some fun, and if aliasing is out...
...with the sort of morbid fascination that makes people slow down to look at car wrecks:
Does the language still have EQUIVALENCE?
OK. I find now that xmms is behaving strangely, not appearing on the display and not playing the sounds. OK, time to look at what version of xmms I have, maybe dump it if I have an older version...but RH 8.0 hides the nice GNOME and KDE GUI layers around rpm from me, and the "packages" menu item it now gives one leads to a dumbed-down program that is fine for people used to the Windows "Install/Remove Program" menu item on the Control Panel, but which will not give me the information I need and that the previous GNOME and KDE programs for package management made trivial.
I am not as happy as I was this morning.
In the immortal words of AOL users, "me too."
.kde directory; just deleted the icons from the panel and then put their moral equivalents back. (Before doing that, they were all just some generic icon rather than something specific to what they ran, and clicking on them got a message about a file in /usr/share/applnk not existing. Looking around shows that directory to be pretty well emptied of nondirectory files, save for things that correspond to programs I installed apart from an RH install/upgrade, e.g. Opera.)
Well, almost. I didn't delete my
Things do seem faster. I'm a happy camper.
Do you know what killed OS/2? No, it wasn't IBM's marketing. It was lack of Win32 compatibility, as well as lack of hardware compability.
That lack of compatibility was urged along by the successive versions of win32s.dll, which kept breaking compatibility. IBM finally gave up, and MS stopped changing win32s.dll, after version 1.30 of win32s.dll came out, which added a call to request memory out of range of the 512 MB limit on DOS sessions under OS/2--a fairly fundamental thing to break, and a limit only recently, long after it would make much difference, eliminated. There's no real reason to be picky about the base address of memory one allocates--so it was added solely to break OS/2 compatibility. Gee...you just claimed that MS couldn't possibly do something like that. Looks like a counterexample to me.
Speaking of breaking existing software--checked MS's EULA lately? To get security fixes, you have to agree that MS has the right to automatically install OS upgrades that may prevent software from running. So nowadays, with Windows you can either live with security holes or accept "DOS isn't done until Lotus won't run" as SOP.
Compatibility is a Catch-22. Without it, people kvetch about not having it. With it, developers see no reason to bother to write for anything but Windows--and indeed, you complain about both the lack of OS/2 native software and the lack of compatibility.
Let us know when you win a Nebula or Hugo, OK?
Got that right. All you have to do is look at the works of Damon Runyan, or to go back a long way, Francois Villon. One of Mario Pei's books starts a chapter on jargon with a rendering of a Villon poem in the corresponding twentieth-century American theives' jargon:
That's been going on for a long time. Remember the "Great Folk Music Scare"? All those clean-cut college types singing about miners and migrant farmers and their working conditions.
"When was the last time you ever dug a ditch, baby?" -- Oingo Boingo, "Capitalism"
Thanks, I've read the book of Job. The Cliff Notes version: God kills and/or tortures Job and his family to settle a bar bet with Satan. When Job complains, God says, a la Chevy Chase, "I'm God, and you're not. Shut up, I can do what I want."
If an earthly father did that, he'd be in the slammer so fast your head would spin. Such a being is a vile beast.
If the Christian God exists, then yes, he must in some way want everyone going to hell to be there, because he established the initial conditions of the world, and being omniscient, knew exactly what would happen, and being omnipotent, could do exactly what he wanted.
In everyday life, foreknowledge and action add up to responsibility. If I set a shiny Ginsu knife down in front of an infant and it manages to kill itself, I'm the one hauled into prison. How much more responsibility must be borne by an omniscient, omnipotent god!
Ah...so for our own good, we are denied knowledge that would prevent us from eternal torment? I'm sorry, but any deity who thinks that's for my good is a monstrous, sadistic beast in comparison with which all the mass murderers, child abusers, and other criminals of earth pale to insignificance.
That's rather hard to reconcile with condemning the vast majority of us to eternal torment.
The exact wording was "In retrospect, it was a mistake to announce an agreement in principle with respect to the curriculum initiatives, a mistake for which I take responsibility." If I were in conspiracy theory mode, or even in Clintonspeak parsing mode, I'd be thinking that he wanted to hold off on publicity until it was a fait accompli.
This is news?
I would respectfully disagree. QED is a transcription of a series of lectures intended for the layman.
Well, yeah, but...the conditional execution bits eat four bits out of every 32-bit instruction, which makes the constraints on immediate operands and offsets for base + offset addressing even more galling than usual on fixed-width instruction sets, for a payoff of uncertain value unless all you do is run the clever GCD code people always give as an example of how great ARM conditional execution is. Among the most perverse aspects are offsets whose ranges vary with the type of data you're loading or storing, the set of permissible immediate values (an eight-bit value shifted by an even number of bits), and an obnoxious constraint on operands of multiply instructions.
Of course, writing code for the ARM is heaven in comparison with writing x86 code, but I would not say it is all that spiffy in comparison with 68xxx or PowerPC.