I thought you might be interested in some counterexamples. I don't think Intercal or Befunge advocates have ever claimed those languages were faster than C. In fact, I don't think they claimed Intercal or Befunge was better than any other language.
Two questions for you: 1) Are you Mark Lutz? 2) Are you a Python programmer?
This isn't a flame. I love Python. I ask because the only other place I've seen this macro is in Mark Lutz' Python books (published by O'Reilly). However, I don't use C++ regularly, and wouldn't know if this was a common idiom.
I expect that these systems were designed to push users back to their vendors, hence the difficulties for anyone else. From what I read about HP, things got worse (for HP) when they were forced to support plain Windows. And heaven help the customer that calls Microsoft for help with Windows. =-)
Given what you say and what I read, it seems that the battle was between MS and HP (and other OEMs), with the user losing in either case. Seems plausible.
"So you'd use symlinks to do stuff like shortening pathnames and things of commnly used files?"
Precisely.
"But it wouldn't help with something like a web project where multiple sites all had the same data on them, changing that one file would change them all."
I think that one of us is confused. Because symlinks are really just shortcuts to the same file name (and, indirectly, file), telling emacs to edit the symlink is the same as telling it to edit the file. The same holds for hard links. So if a hard link would be useful for your above scenario, a symlink would be useful, too. Unless, of course, you wanted some protection from accidental deletion -- with a hard link, every site would have to delete their link before the linked-to file went away.
There is another consideration with hard and soft links: directory loops. You are not allowed to hard link to a directory, because that can really screw up your filesystem. I'm not sure of the details here, but since both the original directory and the hard link would simultaneously be each other's parent *and* child, one can imagine that chaos would come quickly. =-)
If I remember correctly, in very early implementations of what became the unix filesystem, there were no directories. Instead, links were used to provide heirarchical names for the mass of files on the filesystem.
Borrowing the nice, concise def'n from the ln manpage on linux, a hard link is what you've described, with the property that a file is not "deleted" until *all* hard links to it have been "deleted". A symlink (or "soft link") is simply a special file that contains a pathname to some other file.
Consider the following command sequence in a unix shell:
touch foo # creates a file, and creates a name "foo" that points to the file. ln -h foo bar # creates another name which points to the file rm foo # removes one name for the file rm bar # removes one name for the file, and since this is the last name pointing to the file,
# the file is also removed
touch foo ln -s foo bar # creates a symlink pointing to the *name* foo rm foo # removes name foo for file, and since foo is the last name pointing
# to the file, the file is removed cat bar # ERROR: "No such file or directory"
Hope that makes the difference clear (except for the crappy formatting, due to some "compression filter" slashdot complained about). I can't remember ever needing to create a hard link, but I use symlinks all the time for the sake of convenience.
Since I'm getting a lot of helpful info from this thread;-), is NTFS the default for all of MS's operating systems now, or do some still use vfat? The problem, of course, is knowing what you can count on.
If you are correct about this, that's a big relief. I've been paranoid about specifying binary when opening files in C and Python for years (and feeling guilty when I didn't), just in case someone compiled and ran my research code on a DOS/Windows machine. I generally don't care about the line terminator for the stuff I'm doing -- I was worried that my 8 bit data would lose the high bit.
I believe you are correct about a lot of the crap in DOS (and hence Windows) came from VMS, or at least from DEC. If you've ever partitioned a disk under a BSD unix, you'll remember all the (I believe the intent was) arch-independent subsections of device partitions (slices), which are given names like C:.
Did you mean that "yes, making short-cuts programatically is a pain", or "yes, NTFS 5 has symlinks"?
The text/binary mode is a hassle because you have to be careful when writing or reading files to open them appropriately. That is, everywhere else, you don't need to specify "binary" mode. However, if you write code that might be run on Windows (e.g. in Python), you need to specify binary mode explicitly in case some stupid library on someone's machine uses 7bit text mode by default. I would normally chalk this up to a problem with programmer laziness, but because 7bit files are completely unnecessary and have been for many, many years, I ascribe this to Microsoft laziness.
The problem with case retention on VFAT filesystems is that different system calls seem to return the filenames with different stuff capitalized. So you have to be careful. Knowing how things are truly capitalized is
1) nice for the user, so they don't wonder why suddenly the first letter is capitalized but none of the rest of their camel-case name is capitalized (think Windows Explorer)
2) important when working with case sensitive apps or moving stuff to case sensitive filesystems.
Note that I'm not actually a Windows programmer. These are problems I've helped my wife deal with, as her company became Windows-centric a few years ago. To create simple shortcuts, she had to dig into COM and/or ActiveX. The case retention problems occurred somewhere in their OS-abstraction libraries.
Truth be told, we did have the computers on a (single 1400VA) UPS. However, none of the monitors were on the UPS (in order to allow more computers to be connected). It had the same effect, but took too much effort to write the first time. =-)
My opinion is related to, but reveresed from, yours. I feel that Microsoft put so much effort into dumbing-down the users, that they've reduced the computer literacy of their users. My 74-year-old father once commented that "with DOS, I felt like I understood what was happening, and could really control the computer. With Windows, I don't feel in control and have no idea what is going on."
Furthermore, I believe that getting users interested in computers beyond email and web surfing is not only an uphill battle, but a waste of time. Instead, we should stop trying to sell general purpose machines to every person, and concentrate on meeting individual needs via computing appliances.
As you probably anticipated, I still don't agree with you completely:
1) Darwin is Unix, but OS X is not. OS X is a union of Darwin and a UI. The UI was certainly not targeted at me, and Darwin was certainly not targetted at Apple's primary audience. At any rate, my comment about Unix being made by hackers, for hackers, was primarily a historical reference to the origins of Unix (as well as GNU/Linux).
2) Why would I want to make XP look like 95? The way it *looks* is not that important to me. Heck, I thought fvwm was a good-enough window manager, since my idea of a good window manager is one which lets you put lots of terminals on one screen, and then gives you further virtual desktops to fill up with more terminals (and allows customized key-bindings to hop between the desktops). I recently saw that some Microsoft employees released "power user" software that gives a semblance of these capabilities to Windows XP, and I think this is great (even though the pager appears to be a hack using iconized windows). But why didn't Microsoft include this stuff in XP?
3) I should make it clear that calling the MS filesystems "disgraceful" is my opinion, which I believe has some merit. This is why I don't like any of the Windows filesystems (please correct me if MS has "fixed" any of these things into NTFS 5):
a) The Windows filesystems have a hard-wired connection between filesystem names and physical devices (i.e. C:\..., D:\..., etc), which causes software configuration information in the non-human-readable registry to depend on physical device configuration.
b) Using \ instead of/, since \ requires a longer reach on standard keyboards (while this is true, I'm not really serious about this)
c) The separation between devices and files. It is my opinion that the "everything is a file" philosophy employed in the Unix file systems is arguably superior to Microsoft's half-assed support of this concept ('copy con foo.txt' works, but which file represents the sound card's DSP on NTFS 5?). This philosophy goes a long ways in providing nearly-uniform access to all parts of your computer. This design decision greatly enhances the power of scripts, especially when coupled with pipes, redirection, and fine control over file handles.
d) The lack of symbolic links (does NTFS finally have these?). Windows shortcuts require unnecessary nonsense to create programmatically.
e) Windows filesystems, AFAIK, *all* make a distinction between "text" and "binary" files. It is arguable whether this was appropriate in DOS. It is clearly inappropriate now.
f) I don't know about NTFS, but VFAT is case-random. It is difficult, programatically, to correctly discover the original capitalization of a filename.
I'll stop there, because that's what was on the top of my head when I made the comment that Windows filesystems were "disgraceful".
4) My accusation was that Microsoft has completely neglected the command line. That Cygnus Solutions saw fit to remedy this, as best they could, does not excuse Microsoft's utter failure to provide a reasonable command line interface by default. For me, this is important. And XEmacs isn't really the sort of command-line use I'm referring to; even if it were, it isn't supported by Microsoft or provided by default. Finally, why bother turning Windows into Unix, instead of just using Unix? Especially since you can find Free (and free, to boot) Unix implementations? Doing so makes sense for people not allowed to run Unix, but I'm not among these people.
FWIW, there are many LaTeX implementations available for Windows, and you can use TeTeX or MikTeX via cygwin. However, Microsoft decidied to focus its effort on various equation editors that can't seem to agree which font to use (for instance, curly-epsilon or set-inclusion epsilon seems to depend on which Microsoft software packages happen to be present on the computer used for display -- can't they at least be consistent within their own software?).
I wasn't really trying to support the assertion that Windows sucked for everyone. I was trying to support the assertion that Windows sucks for me. I wanted to make this point against my perception that you were asserting everyone could like Windows if they just got over their ethical hangups about Microsoft and Windows.
The Windows power users I've watched (all of which are developers or researchers, because of the company I keep) seem to make every effort to turn Windows into Unix. It seems you are suggesting that something in Windows has the possibility of causing power users to lose their minds every day. This suggests to me that Windows is ill-suited for power users. I must admit I don't understand the point of your quote from Tim O'Reilly.
Although you didn't suggest it, I'd like to emphasize that , given a choice, there is no good reason to overcome one's sense of ethics in order to use Windows.
I'll agree that a *good* and *consistent* interface is a good thing. I don't believe MS has generally done well on either of these counts. For the Start button in particular, they *have* been consistent since about 1996 when Win95 came out (consistent as long as you don't count Windows NT -- when did it finally get a Start button?). However, I don't believe the start button is *good*. Docks/app bars are *good*, and even Microsoft eventually figured that out.
It used to be that OEMs like HP would provide all sorts of user-friendly tools to help new users orient to their computer. These tools would occasionaly replace various bit of Windows functionality in some cases. An example might be a specialized "Start" button that would pop up a friendly menu tailored for the software that HP chose to install. HP might include a registration wizard that popped up the first time a person hit the Start button, etc.
Microsoft rewrote its OEM contracts to forbid such behavior, publically claiming that it hurt the integrety of the "consistent" (their word, not mine;-) Windows "look-and-feel". They claimed that this hurt users by breeding confusion. Later on, HP released statistics from their tech support department that showed users had a far harder time without HPs changes to Windows, contradicting Microsoft's claim. Furthermore, HP saw the percentage of registrations fall; I'm sure that Microsoft saw their Windows registrations rise.
More recently, Microsoft has claimed that allowing OEMs to customize Windows before shipping a machine to a customer violates their Windows copyright. In effect, I believe their argument is that the OEMs are creating an unauthorized derivative work. Ironically, it's because of Microsoft's successful defense against Apple that look-and-feel is not protected by copyright, and hence the OEMs cannot possibly be violating Microsoft's Windows copyright when they mess with the desktop icons and start button.
I think it is reasonable to conclude that the witness was trying to confuse or pursuede the judge with this statement. That is, Microsoft is trying to spread FUD in the courtroom. I'm really hoping that Judge Collen Kollar-Kottelly has learned enough computer history to be able to discard such nonsense. Failing that, I'm hoping that she is smart enough to recognize unsupported FUD and dismiss it when making her decision.
Especially true since the main two US political parties are nearly indistinguishable. Every citizen understands completely that no matter who they vote for, a politician will win and the citizen will lose.;-)
Speak for yourself. Microsoft's disgraceful filesystem and complete neglect of the command line has made Windows a painful and inefficient operating environment for me. I see no reason why I should have to accept whatever Microsoft decrees just because they have enough monopolistic power to end innovation in computing. I don't use word processors (I'm a math grad student, and use LaTeX for everything now), and I like text-based email clients.
Not all computer users look like the folks in the advertisements where everyone is smiling as they look over one-another's shoulders at Grandma via some video-conferencing tool. For me, Microsoft's operating system is mostly useless. As for new users, I've spent ages of my time helping "new computer users" figure out Windows, and have had more success teaching DOS and Unix (I am absolutely serious about this). As a "powerful user", MS Windows is the biggest handicap I could impose on myself (maybe not true for other "powerful users").
Accepting that I'm running WINDOWS isn't going to change any of this. Computer users aren't made by cookie-cutters, nor is Microsoft trying to make any particular group of computer users happy. Microsoft is trying to make their *shareholders* happy, at which they succeed by developing products that are easy to *sell*. The fact that they overcharge is nearly irrelevant, since I can't pay them enough money to care about my needs.
I prefer Linux and Unix. Made by hackers, for hackers.
If only software interoperation in the Windows world were so clean.
1) Refrigerator stores stuff and keeps it cool. The user can figure out what to do with the stuff, and there is no need for robotic arms to move stuff from the fridge to the toaster.
2) The toaster can toast bread when the bread is inserted and the proper button is pushed. If the user is an "advanced user", they are likely to succeed when toasting other things (bagels, waffles, veggie burgers) using the same basic principles.
3) The user must understand the basic principles of removing bread from the fridge, removing the twist-tie (or sometimes a quik-lok!), putting the bread in the toaster, and pushing the "toast" button. "Advanced" users will understand that replacing the twist-tie or quik-lok and placing the bread back in the fridge is important to long-term success in the kitchen.
And now that I've written this, it becomes clear that the kitchen is analagous to a filesystem, and humans are quite capable of dealing with such "abstractions" (the Unix way =-) without having the refrigerator try to put bread in the toaster for you (the misguided Windows way =-). Some people might choose to have the fridge make ice automatically, which is a harmless convenience.
I suppose that undoing the twist-tie is decompressing?;-)
"Not a big deal" depends on what else you might have on that circuit. I've spent my graduate school career in 100-year-old houses in Pittsburgh. One of them had the entire kitchen, one bedroom, and the living room on a single 15 amp fuse. That bedroom was the "office", in which I kept my computers.
Part way through a lan party, somebody was using the microwave when the refrigerator compressor kicked in. That took down 8 computers simultaneously (not to mention all the lights on that floor). That was a big deal, at least for whomever was winning when the power died.;-)
What's weird is that they wanted Nintendo to throw their brand and expertise away by cancelling the GameCube and backing the very new, very experimental, very strange Xbox. Seems idiotic to me.
Unfortunately, I agree with you.;-) I wish vocational education wasn't so stigmatized in the US. Not everyone needs, wants, or should be interested in (my personal definition of;-) a Real University Education.
Personal nitpick: Universities have *no* obligation to train people for any particular application. Universities have an obligation to teach people how to *think*. Technical/vocational training is where a person can pay to learn particular programs. This is *not* the role of a university.
I believe that people have died after having been hit by a bycicle. The shape of a bicycle is probably part of the issue. I expect that someone getting hit by a person on a segway at 14mph would be hurt pretty badly, but the shape of the segway might reduce internal injuries.
That said, I don't think I would appreciate segways sharing a sidewalk with me. I *know* I don't like bicycles sharing a sidewalk with me.
Slashdot Mirror
As a (primarily) C programmer, I saw this macro as a way to smack down those misguided C++ pedants. ;-)
-Paul Komarek
I thought you might be interested in some counterexamples. I don't think Intercal or Befunge advocates have ever claimed those languages were faster than C. In fact, I don't think they claimed Intercal or Befunge was better than any other language.
-Paul Komarek
Two questions for you:
1) Are you Mark Lutz?
2) Are you a Python programmer?
This isn't a flame. I love Python. I ask because the only other place I've seen this macro is in Mark Lutz' Python books (published by O'Reilly). However, I don't use C++ regularly, and wouldn't know if this was a common idiom.
-Paul Komarek
I expect that these systems were designed to push users back to their vendors, hence the difficulties for anyone else. From what I read about HP, things got worse (for HP) when they were forced to support plain Windows. And heaven help the customer that calls Microsoft for help with Windows. =-)
Given what you say and what I read, it seems that the battle was between MS and HP (and other OEMs), with the user losing in either case. Seems plausible.
-Paul Komarek
"So you'd use symlinks to do stuff like shortening pathnames and things of commnly used files?"
Precisely.
"But it wouldn't help with something like a web project where multiple sites all had the same data on them, changing that one file would change them all."
I think that one of us is confused. Because symlinks are really just shortcuts to the same file name (and, indirectly, file), telling emacs to edit the symlink is the same as telling it to edit the file. The same holds for hard links. So if a hard link would be useful for your above scenario, a symlink would be useful, too. Unless, of course, you wanted some protection from accidental deletion -- with a hard link, every site would have to delete their link before the linked-to file went away.
There is another consideration with hard and soft links: directory loops. You are not allowed to hard link to a directory, because that can really screw up your filesystem. I'm not sure of the details here, but since both the original directory and the hard link would simultaneously be each other's parent *and* child, one can imagine that chaos would come quickly. =-)
If I remember correctly, in very early implementations of what became the unix filesystem, there were no directories. Instead, links were used to provide heirarchical names for the mass of files on the filesystem.
-Paul Komarek
Borrowing the nice, concise def'n from the ln manpage on linux, a hard link is what you've described, with the property that a file is not "deleted" until *all* hard links to it have been "deleted". A symlink (or "soft link") is simply a special file that contains a pathname to some other file.
Consider the following command sequence in a unix shell:
touch foo # creates a file, and creates a name "foo" that points to the file.
ln -h foo bar # creates another name which points to the file
rm foo # removes one name for the file
rm bar # removes one name for the file, and since this is the last name pointing to the file,
# the file is also removed
touch foo
ln -s foo bar # creates a symlink pointing to the *name* foo
rm foo # removes name foo for file, and since foo is the last name pointing
# to the file, the file is removed
cat bar # ERROR: "No such file or directory"
Hope that makes the difference clear (except for the crappy formatting, due to some "compression filter" slashdot complained about). I can't remember ever needing to create a hard link, but I use symlinks all the time for the sake of convenience.
-Paul Komarek
Does HardLink create symbolic and hard links?
;-), is NTFS the default for all of MS's operating systems now, or do some still use vfat? The problem, of course, is knowing what you can count on.
Since I'm getting a lot of helpful info from this thread
-Paul Komarek
Ah, now that is important! Okay, I'll continue my paranoia. Formerly, it was suspicious paranoia. Now, with your help, it is informed paranoia!
-Paul Komarek
If you are correct about this, that's a big relief. I've been paranoid about specifying binary when opening files in C and Python for years (and feeling guilty when I didn't), just in case someone compiled and ran my research code on a DOS/Windows machine. I generally don't care about the line terminator for the stuff I'm doing -- I was worried that my 8 bit data would lose the high bit.
-Paul Komarek
I believe you are correct about a lot of the crap in DOS (and hence Windows) came from VMS, or at least from DEC. If you've ever partitioned a disk under a BSD unix, you'll remember all the (I believe the intent was) arch-independent subsections of device partitions (slices), which are given names like C:.
Did you mean that "yes, making short-cuts programatically is a pain", or "yes, NTFS 5 has symlinks"?
The text/binary mode is a hassle because you have to be careful when writing or reading files to open them appropriately. That is, everywhere else, you don't need to specify "binary" mode. However, if you write code that might be run on Windows (e.g. in Python), you need to specify binary mode explicitly in case some stupid library on someone's machine uses 7bit text mode by default. I would normally chalk this up to a problem with programmer laziness, but because 7bit files are completely unnecessary and have been for many, many years, I ascribe this to Microsoft laziness.
The problem with case retention on VFAT filesystems is that different system calls seem to return the filenames with different stuff capitalized. So you have to be careful. Knowing how things are truly capitalized is
1) nice for the user, so they don't wonder why suddenly the first letter is capitalized but none of the rest of their camel-case name is capitalized (think Windows Explorer)
2) important when working with case sensitive apps or moving stuff to case sensitive filesystems.
Note that I'm not actually a Windows programmer. These are problems I've helped my wife deal with, as her company became Windows-centric a few years ago. To create simple shortcuts, she had to dig into COM and/or ActiveX. The case retention problems occurred somewhere in their OS-abstraction libraries.
-Paul Komarek
Truth be told, we did have the computers on a (single 1400VA) UPS. However, none of the monitors were on the UPS (in order to allow more computers to be connected). It had the same effect, but took too much effort to write the first time. =-)
-Paul Komarek
Woohoo! Bring out the guillotines! ;-)
-Paul Komarek
My opinion is related to, but reveresed from, yours. I feel that Microsoft put so much effort into dumbing-down the users, that they've reduced the computer literacy of their users. My 74-year-old father once commented that "with DOS, I felt like I understood what was happening, and could really control the computer. With Windows, I don't feel in control and have no idea what is going on."
Furthermore, I believe that getting users interested in computers beyond email and web surfing is not only an uphill battle, but a waste of time. Instead, we should stop trying to sell general purpose machines to every person, and concentrate on meeting individual needs via computing appliances.
-Paul komarek
As you probably anticipated, I still don't agree with you completely:
/, since \ requires a longer reach on standard keyboards (while this is true, I'm not really serious about this)
1) Darwin is Unix, but OS X is not. OS X is a union of Darwin and a UI. The UI was certainly not targeted at me, and Darwin was certainly not targetted at Apple's primary audience. At any rate, my comment about Unix being made by hackers, for hackers, was primarily a historical reference to the origins of Unix (as well as GNU/Linux).
2) Why would I want to make XP look like 95? The way it *looks* is not that important to me. Heck, I thought fvwm was a good-enough window manager, since my idea of a good window manager is one which lets you put lots of terminals on one screen, and then gives you further virtual desktops to fill up with more terminals (and allows customized key-bindings to hop between the desktops). I recently saw that some Microsoft employees released "power user" software that gives a semblance of these capabilities to Windows XP, and I think this is great (even though the pager appears to be a hack using iconized windows). But why didn't Microsoft include this stuff in XP?
3) I should make it clear that calling the MS filesystems "disgraceful" is my opinion, which I believe has some merit. This is why I don't like any of the Windows filesystems (please correct me if MS has "fixed" any of these things into NTFS 5):
a) The Windows filesystems have a hard-wired connection between filesystem names and physical devices (i.e. C:\..., D:\..., etc), which causes software configuration information in the non-human-readable registry to depend on physical device configuration.
b) Using \ instead of
c) The separation between devices and files. It is my opinion that the "everything is a file" philosophy employed in the Unix file systems is arguably superior to Microsoft's half-assed support of this concept ('copy con foo.txt' works, but which file represents the sound card's DSP on NTFS 5?). This philosophy goes a long ways in providing nearly-uniform access to all parts of your computer. This design decision greatly enhances the power of scripts, especially when coupled with pipes, redirection, and fine control over file handles.
d) The lack of symbolic links (does NTFS finally have these?). Windows shortcuts require unnecessary nonsense to create programmatically.
e) Windows filesystems, AFAIK, *all* make a distinction between "text" and "binary" files. It is arguable whether this was appropriate in DOS. It is clearly inappropriate now.
f) I don't know about NTFS, but VFAT is case-random. It is difficult, programatically, to correctly discover the original capitalization of a filename.
I'll stop there, because that's what was on the top of my head when I made the comment that Windows filesystems were "disgraceful".
4) My accusation was that Microsoft has completely neglected the command line. That Cygnus Solutions saw fit to remedy this, as best they could, does not excuse Microsoft's utter failure to provide a reasonable command line interface by default. For me, this is important. And XEmacs isn't really the sort of command-line use I'm referring to; even if it were, it isn't supported by Microsoft or provided by default. Finally, why bother turning Windows into Unix, instead of just using Unix? Especially since you can find Free (and free, to boot) Unix implementations? Doing so makes sense for people not allowed to run Unix, but I'm not among these people.
FWIW, there are many LaTeX implementations available for Windows, and you can use TeTeX or MikTeX via cygwin. However, Microsoft decidied to focus its effort on various equation editors that can't seem to agree which font to use (for instance, curly-epsilon or set-inclusion epsilon seems to depend on which Microsoft software packages happen to be present on the computer used for display -- can't they at least be consistent within their own software?).
I wasn't really trying to support the assertion that Windows sucked for everyone. I was trying to support the assertion that Windows sucks for me. I wanted to make this point against my perception that you were asserting everyone could like Windows if they just got over their ethical hangups about Microsoft and Windows.
The Windows power users I've watched (all of which are developers or researchers, because of the company I keep) seem to make every effort to turn Windows into Unix. It seems you are suggesting that something in Windows has the possibility of causing power users to lose their minds every day. This suggests to me that Windows is ill-suited for power users. I must admit I don't understand the point of your quote from Tim O'Reilly.
Although you didn't suggest it, I'd like to emphasize that , given a choice, there is no good reason to overcome one's sense of ethics in order to use Windows.
-Paul Komarek
I'll agree that a *good* and *consistent* interface is a good thing. I don't believe MS has generally done well on either of these counts. For the Start button in particular, they *have* been consistent since about 1996 when Win95 came out (consistent as long as you don't count Windows NT -- when did it finally get a Start button?). However, I don't believe the start button is *good*. Docks/app bars are *good*, and even Microsoft eventually figured that out.
-Paul Komarek
"duel booting" is what Microsoft advocates, in the assumption that they'll wipe out the competition before the signal to fire. ;-)
I expect your friend was "dual booting", as "duel-booting" would not be in his best interests.
-Paul Komarek
It used to be that OEMs like HP would provide all sorts of user-friendly tools to help new users orient to their computer. These tools would occasionaly replace various bit of Windows functionality in some cases. An example might be a specialized "Start" button that would pop up a friendly menu tailored for the software that HP chose to install. HP might include a registration wizard that popped up the first time a person hit the Start button, etc.
;-) Windows "look-and-feel". They claimed that this hurt users by breeding confusion. Later on, HP released statistics from their tech support department that showed users had a far harder time without HPs changes to Windows, contradicting Microsoft's claim. Furthermore, HP saw the percentage of registrations fall; I'm sure that Microsoft saw their Windows registrations rise.
Microsoft rewrote its OEM contracts to forbid such behavior, publically claiming that it hurt the integrety of the "consistent" (their word, not mine
More recently, Microsoft has claimed that allowing OEMs to customize Windows before shipping a machine to a customer violates their Windows copyright. In effect, I believe their argument is that the OEMs are creating an unauthorized derivative work. Ironically, it's because of Microsoft's successful defense against Apple that look-and-feel is not protected by copyright, and hence the OEMs cannot possibly be violating Microsoft's Windows copyright when they mess with the desktop icons and start button.
I think it is reasonable to conclude that the witness was trying to confuse or pursuede the judge with this statement. That is, Microsoft is trying to spread FUD in the courtroom. I'm really hoping that Judge Collen Kollar-Kottelly has learned enough computer history to be able to discard such nonsense. Failing that, I'm hoping that she is smart enough to recognize unsupported FUD and dismiss it when making her decision.
-Paul Komarek
Especially true since the main two US political parties are nearly indistinguishable. Every citizen understands completely that no matter who they vote for, a politician will win and the citizen will lose. ;-)
-Paul Komarek
Speak for yourself. Microsoft's disgraceful filesystem and complete neglect of the command line has made Windows a painful and inefficient operating environment for me. I see no reason why I should have to accept whatever Microsoft decrees just because they have enough monopolistic power to end innovation in computing. I don't use word processors (I'm a math grad student, and use LaTeX for everything now), and I like text-based email clients.
Not all computer users look like the folks in the advertisements where everyone is smiling as they look over one-another's shoulders at Grandma via some video-conferencing tool. For me, Microsoft's operating system is mostly useless. As for new users, I've spent ages of my time helping "new computer users" figure out Windows, and have had more success teaching DOS and Unix (I am absolutely serious about this). As a "powerful user", MS Windows is the biggest handicap I could impose on myself (maybe not true for other "powerful users").
Accepting that I'm running WINDOWS isn't going to change any of this. Computer users aren't made by cookie-cutters, nor is Microsoft trying to make any particular group of computer users happy. Microsoft is trying to make their *shareholders* happy, at which they succeed by developing products that are easy to *sell*. The fact that they overcharge is nearly irrelevant, since I can't pay them enough money to care about my needs.
I prefer Linux and Unix. Made by hackers, for hackers.
-Paul Komarek
If only software interoperation in the Windows world were so clean.
;-)
1) Refrigerator stores stuff and keeps it cool. The user can figure out what to do with the stuff, and there is no need for robotic arms to move stuff from the fridge to the toaster.
2) The toaster can toast bread when the bread is inserted and the proper button is pushed. If the user is an "advanced user", they are likely to succeed when toasting other things (bagels, waffles, veggie burgers) using the same basic principles.
3) The user must understand the basic principles of removing bread from the fridge, removing the twist-tie (or sometimes a quik-lok!), putting the bread in the toaster, and pushing the "toast" button. "Advanced" users will understand that replacing the twist-tie or quik-lok and placing the bread back in the fridge is important to long-term success in the kitchen.
And now that I've written this, it becomes clear that the kitchen is analagous to a filesystem, and humans are quite capable of dealing with such "abstractions" (the Unix way =-) without having the refrigerator try to put bread in the toaster for you (the misguided Windows way =-). Some people might choose to have the fridge make ice automatically, which is a harmless convenience.
I suppose that undoing the twist-tie is decompressing?
-Paul Komarek
"Not a big deal" depends on what else you might have on that circuit. I've spent my graduate school career in 100-year-old houses in Pittsburgh. One of them had the entire kitchen, one bedroom, and the living room on a single 15 amp fuse. That bedroom was the "office", in which I kept my computers.
;-)
Part way through a lan party, somebody was using the microwave when the refrigerator compressor kicked in. That took down 8 computers simultaneously (not to mention all the lights on that floor). That was a big deal, at least for whomever was winning when the power died.
-Paul Komarek
What's weird is that they wanted Nintendo to throw their brand and expertise away by cancelling the GameCube and backing the very new, very experimental, very strange Xbox. Seems idiotic to me.
-Paul Komarek
Unfortunately, I agree with you. ;-) I wish vocational education wasn't so stigmatized in the US. Not everyone needs, wants, or should be interested in (my personal definition of ;-) a Real University Education.
-Paul Komarek
Personal nitpick: Universities have *no* obligation to train people for any particular application. Universities have an obligation to teach people how to *think*. Technical/vocational training is where a person can pay to learn particular programs. This is *not* the role of a university.
Wow, I *am* sensitive about this! =-)
-Paul Komarek
I believe that people have died after having been hit by a bycicle. The shape of a bicycle is probably part of the issue. I expect that someone getting hit by a person on a segway at 14mph would be hurt pretty badly, but the shape of the segway might reduce internal injuries.
That said, I don't think I would appreciate segways sharing a sidewalk with me. I *know* I don't like bicycles sharing a sidewalk with me.
-Paul Komarek