What is really funny is that this pretty clearly proves that global warming was well believed (and thus fought by Niven/Pournelle) 17 years ago, and must have been for many years before that for it to get well-known enough for this spoof to be effective. Yet the idiots will continue to claim that "people used to believe we were going into the ice age just 10 years ago" or whatever, and will drag out the exact same Time magazine article every time to "prove" this. (in fact that article is pretty clearly claiming that some scientists are going against the consensus, I'm sure 20 years from now somebody will drag out a gw-denier article as "proof that scientists thought the world was getting cooler".
I very clearly remember in 1968 in elementary school (when 2001 was in the theatres, that's why I can date it well) that my liberal teachers told us that "the earth will turn into venus due to CO2 and the greenhouse effect". People have believed in global warming for a LONG time. Only "ice age" stuff I can find is from the 50's (like sci fi stories, such as Clarke's short story about the last survivor on a frozen earth thinking he is being rescued because he hears loud noises from the north, but it is actually approaching glaciers). This stuff I believe was pretty much killed off in popular literature when the measurements of Venus indicated the surface was far hotter than expected and led to the idea of the "greenhouse effect".
This in no way "falsifies" intelligent design. They will just say "oh that's micro evolution" or something. Read the posts from the ID'ers above and it is pretty obvious that this does NOTHING to "falsify" intelligent design. "God" could have created the entire earth exactly as-is last Thursday, including us and our memories. We CANNOT "falsify" that. ID'ers tend to ignore this, but they will basically claim that anything we observe is not what god did. Unless we actually create an Earth and cause life to evolve from nothing all the way to a human (and probably not some other intelligent creature) would they perhaps shut up.
Evolution is TRIVIAL to falsify. One species giving birth to an unrelated one would falsify it. A species appearing out of nothing would falsify it. Even if you assume "god" would not do something "obvious", a lack of correlation between dna and observed traits would falsify it. I believe you are confusing the lack of finding any actual evidence to show Evolution is false with "unfalsability".
Of course you are not going to understand any of this.
They will say "that's just micro evolution". You cannot win against these idiots. "macro evolution" is anything larger than the largest observed evolution. Perhaps eventually science will build a complete replica of Earth and use some sort of time warp to observe actual life arising, but if that experiment does not get all the way to intelligent life, they will still say "that's just micro evolution"
Microsoft is mentioned 4 times in the linked-to article:
With iPhone 2.0 software, iPhone does even more for your enterprise. It supports Microsoft Exchange ActiveSync, delivering push email, calendar, and contacts.
With support for Microsoft Word, Excel, and PowerPoint, PDF, JPEG, and now iWork, attachments can be viewed exactly as they were designed to.
And now that iPhone has built-in support for Microsoft Exchange ActiveSync, users get all the benefits of push email.
Support for Microsoft Exchange ActiveSync and industry-standard corporate security standards allows IT professionals to seamlessly integrate iPhone into their corporate environments.
Whether this counts as "running Microsoft software" is hard to say.
That is a quote from the article, not a Slashdot addition.
And I certainly did not read it as "bash Microsoft". I read it as "bash those people who forced Microsoft to default to standards because they made all you web developers have to do more work".
Microsoft is clearly, finally, doing the right thing. And I think that sentence is from somebody mad that their beloved Microsoft is being forced to do the right thing and is making up reasons why this is bad.
Wow! You have demonstrated an unbelievably low level of reading comprehension. Congratulations!
In case you missed it I will quote the second sentence: "Windows has more specialty applications that Linux lacks but that is only because it has been around in desktop use for longer then Linux has." Now you can argue whether that is true or false, but he directly addressed your question.
Unfortunatly the Posix subsystems exec call just pastes the argv back together with spaces and passes it to the next program (it has to, it runs through the normal Windows API at some point), so this solves nothing on Windows.
I do agree that having the programs do the quoting and splitting is a big security hole. And I hate to admit it (because it makes it a pain to do some command lines on Unix and I have often said it's something Windows did right) but the globbing being done by the calling program is also a good security idea, some stupid browser could be convinced perhaps to run "rm" with the file named by the web page and it sent "*"
Modern operating systems (Vista, MacOS X, Linux?) use vector graphics, which means that you can never have a too high resolution, as everything can be scaled to fit. The problem is that this only works for fonts. Other graphics do not scale at all well, unless the scale is an integer (here the scale is only about 1.5). Pictures are either blurry or there is an annoying zigzag in a diagonal line, and line graphics will make things that should be evenly spaced come out irregular.
Another annoying problem on Linux and OS/X is the mysterious belief that fonts should be measured in "points" while EVERY SINGLE OTHER GRAPHIC on the screen is measured in "pixels". Basically you are forced to work with two independent coordinate systems, which makes nice graphics incredibly hard. Windows had this problem, but they "fixed" it by defining a pixel as 4/3 of a pixel always, no matter what the resolution of the screen is. Other systems have ways to specify fonts in pixels, but the problem is GUI controls and saved preferences that insist on writing points instead, making it impossible to use the pixel api.
I don't think scaled vector graphics are going to really work until we reach 200 dpi (allowing the scale to be 2) and people figure out that fonts and graphics should use the same coordinate system (probably 1 unit == the nearest integer number of pixels to 1/96 inch).
It might be a good idea to use the term "O$$" for people writing stuff that will be released under an open license, but getting paid for it. Ideally this should be people who would not bother writing it if they were not paid. This is not a small set, in fact it is probably a quarter or so of the contributions to Linux.
This would also drive crazy the astroturfers, who always go into a childish rage when they see "M$" here. I recommend that "MS" only be used for stuff Microsoft does for no income.
I love how you underestimated the most pessimistic Linux desktop usage I have ever seen (1%) by ONE HUNDRED TIMES! Now tell me why you think you know what you are talking about.
Also they completely ignore the open source software that is produced as a paid-for job where the job is *not* to produce the open source software. If you patch the Linux kernel to get your job done, then you have contributed, yet you did exactly the same work you would have even if you had not added to open source.
That is the obvious example, and probably tiny. What is HUGE (and I know personally having done exactly that) is libraries, typically LGPL, that are produced because they are needed for a commercial project, but are then released as open source, even if the commercial project is not.
MSDOS 2 was when I thought Microsoft might do the right thing. They really modified their CP/M clone to add Unix/Posix style api, when there was absolutely no business reason, it was only to make it a nicer machine to write software for. And they did it in just a few months.
Of course they did not really finish it and it has gone downhill ever since. I believe IBM insisted on a totally paranoid level of back-compatability, which is why we have backslashes in the filenames and no escape sequences in the stdio and no/dev and thus no file can be named "com" (until they finally fixed that crap about 2 decades after they should have).
But even seeing what they attempted it was unbelievably refreshing.
Anybody else have any thoughts about MSDOS 2.0?
I do agree that Windows 95 is their other high point. They really did some innovations in GUI design which you can see if you compare to contemporary X and Mac and NextStep:
1. The task bar which has an "icon" in it for a window *whether or not* the window was "open". *EVERYBODY* else only had the icon when the window was "iconized". This is a HUGE deal but everybody is so used to it now that they don't see it.
2. The start menu. Everybody else relied on searching through folders or using a shell command to start a program.
3. Removal of a graphical line between the resize window borders and the contents. I personally did this a bit earlier on Next (but not using Nextstep) and thought it looked really good, and was both very happy and also a bit mad to see my idea used by Microsoft. Compare Windows95 window borders with earlier ones and contemporary X ones and you will see how much cleaner they look.
If you have physical access to the machine and can boot a linux disk and you can read/write the main disk, then you can read it, and change anything on it. I'm sure there are millions of things you can do to be able to it to reboot as a usable Windows, or you can just stay in Linux and copy all the secret information you want off the disk.
I just can't see how this is a story. It's not a mistake on Microsoft's part. You can replace init on Linux if you want and reboot with full access.
The test might not have had 2^6 different possible combinations. If instead it was a random arrangement of 3 on and 3 off, there would only be 20 combinations. This would make the expected percentage getting it right by random chance is 5%. This is a lot closer to the measured values so I think this is what happened.
I remember a version of this where a local band decided their name should be "Free Roast Beef". Bars would then put up a sign saying "Tonight only: Free Roast Beef".
The trains in the city were electric, not steam/coal powered.
What cars did do is replace the horse, which is really a far worse polluter. Probably much more dangerous pollution as well. Though I guess they were carbon neutral.
A lot of people have just wasted a vast amount of time contributing software to this device. They could have said this was the plan from the start and maybe those people could have concentrated on hardware drivers or interesting Windows software for it. Instead an awful lot of man years of contributed effort is wasted by this moronic decision (no, not the decision to switch to XP. The decision to, for years, lie about what direction they were going, apparently to garner publicity).
How do you change "MS-PL" to "MS-LPL"? You add an "L". And the result is a MORE restrictive license.
How do you change "GPL" to "LGPL"? You add an "L". And the result is a LESS restrictive license.
Are you seriously claiming that this is not purposly done by Microsoft to confuse things? I'm sorry, I don't believe you.
And you still didn't address the fact that you are ignoring the MS-LPL, MS-LRL, and MS-RSL, as well as several other licenses that they call "shared source".
use boost::filesystem (or whatever the analog is for your language/framework). It's really not hard.
Any interface that requires me to treat filenames differently from other string data is TOTALLY unacceptable. Try writing a real program some day, such as one that implements an interpreted language where users may want to store a filename into a string variable.
Furthermore, the Win32 api accepts BOTH forward and backward slash. I am quite able to write portable programs using it. My problem is that any communication to other programs, which is by far the area where I don't want to worry about "is this a filename?" is where it breaks. This is not really Microsoft's fault, it is clueless programmers writing applications. But Microsoft could help by insisiting the programs accept filenames in both versions.
I don't care if the _s functions are written by some standards organization. On the Linux side I really don't care what Theo says (he obviously has some problem with code that was written for BSD as his objections, and yours, are stupid). To the vast majority of people it is obvious that strlcpy is the right solution. strlcpy do the *right* thing: truncation is NOT a problem if all useful values are shorter than the truncation (such as when the buffer is compared to a list of tokens and no token is longer than the buffer, or when the OS has a limit such that the truncated value could not possibly name a file anyway). If you want to do something "properly" then use std string in C++. Also strlcpy returns the length of the buffer you need, so you can allocate it, this can result in MAJOR time savings if the majority of your work can be done in automatic arrays on the stack, with malloc only called in the rare too-long cases, no other solution provides this.
I do think everybody wants "how big is this value" to be part of the syntax. It makes no sense that it be in a header file. You might as well design a programming language where the actual conversion into machine instructions is contained in header files. That would be clever, but it would not be a "compiler".
The Win32 api certainly does accept forward slashes just fine. The problem is appliations and clueless programmers that don't think files ever contain forward slashes. Generally you cannot cut & paste or drag & drop or type in a filename with forward slashes into many applications, and operation system calls like getcwd cannot return strings with forward slashes without them crashing. What I want Microsoft to do is insist that the programs should all accept and work with these, or they don't get the "windows certified" label or something. I am really sick and tired of having to add tons of code to decide whether a piece of text is a filename or not and disabling all possibilities of quoting when it is. I am pretty certain there are plenty of programmers inside Microsoft who would like to fix this as well.
For drive letters, the current syntax would certainly continue to work. I just want an alternative syntax so that "/" can start an unambiguous filename. This would allow the disk structure to be duplicated on a Unix machine so that software can go back & forth. Best suggestion I have heard is to have "/A:/" be the same as "A:/". It would be really nice if readdir() of "/" list these.
MS is just being assholes about the C99 stuff. First of all they ignored the BSD strlcpy and strlcat, which are quite proper solutions. Their "standard" is strcpy_s which *throws an exception* when the buffer overflows. That is just ludricous, what it means is that programs will throw exceptions and cause a DOS rather than just truncating. Really what they are trying to do is force everybody to use Windows-specific calls. Adding underscores to a random set of C99 functions that are safe, especially snprintf, is the real giveaway that they just want to make it impossible to port code.
The C people should realize that we want "N bits" and really don't give a damn about any other considerations. I think C should support "int:22 x" to mean an integer that holds at least 22 bits, with a further guarantee that any power of 2 greater or equal to 8 means *exactly* that many bits. This typedef stuff is nonsense. And both Windows and posix should stop declaring a new foo_t type for every integer in the world, it really does help to know that two of them are the same size and the standards should enforce this by using the same type.
The answer is a list of rather simple things, but it is not what they want to hear or expect to hear (I think they expect to hear demands for open source):
#1: Fix filenames and filesystem so they match Unix. This means you use the forward slash. Refuse to "microsoft certify" any software that will not accept a pasted or typed filename with a forward slash in it, and change all the OS api that returns filenames to return forward slashes (probably with a registry setting) and again refuse to "microsoft certify" software that fails when this setting is on. And get rid of the damn drive letters (just make "/A:/" be the same as "A:/") and support UTF-8 encoding of the filenames at all times (probably by changing the "a" version of the win32 api to be hard-coded to UTF-8).
#2: Support OpenGL, meaning that by default you get at least what Mesa provides. Supporting OpenGL 1.4 only is not acceptable.
#3: Support C99 standard functions and don't make your compiler spew a lot of bogus "warnings" that you put in there to try to encourage people to change to your windows-specific functions. Remove the underscores you stuck on lots of the functions so that portable useful code cannot be written.
And lets look at the common acronyms of the things used:
MS-PL, MS-RL I notice you did not list the MS-RSL (which was listed on the exact same faq page as your quote about OSS you cut & pasted, although they seem to have obfuscated it a bit by not listing the abbreviation so it was harder to read). You also don't list the MS-LPL and the MS-LRL.
These three are the ones being complained about (also the approximately 60 other liceneses that they also call "shared source", but to their credit they are trying to phase out and cut down to only the five listed here).
That fact that you blatently ignored the licenses that everybody is arguing about leads me to believe you are a shill.
Slashdot Mirror
What is really funny is that this pretty clearly proves that global warming was well believed (and thus fought by Niven/Pournelle) 17 years ago, and must have been for many years before that for it to get well-known enough for this spoof to be effective. Yet the idiots will continue to claim that "people used to believe we were going into the ice age just 10 years ago" or whatever, and will drag out the exact same Time magazine article every time to "prove" this. (in fact that article is pretty clearly claiming that some scientists are going against the consensus, I'm sure 20 years from now somebody will drag out a gw-denier article as "proof that scientists thought the world was getting cooler".
I very clearly remember in 1968 in elementary school (when 2001 was in the theatres, that's why I can date it well) that my liberal teachers told us that "the earth will turn into venus due to CO2 and the greenhouse effect". People have believed in global warming for a LONG time. Only "ice age" stuff I can find is from the 50's (like sci fi stories, such as Clarke's short story about the last survivor on a frozen earth thinking he is being rescued because he hears loud noises from the north, but it is actually approaching glaciers). This stuff I believe was pretty much killed off in popular literature when the measurements of Venus indicated the surface was far hotter than expected and led to the idea of the "greenhouse effect".
You are being stupid.
This in no way "falsifies" intelligent design. They will just say "oh that's micro evolution" or something. Read the posts from the ID'ers above and it is pretty obvious that this does NOTHING to "falsify" intelligent design. "God" could have created the entire earth exactly as-is last Thursday, including us and our memories. We CANNOT "falsify" that. ID'ers tend to ignore this, but they will basically claim that anything we observe is not what god did. Unless we actually create an Earth and cause life to evolve from nothing all the way to a human (and probably not some other intelligent creature) would they perhaps shut up.
Evolution is TRIVIAL to falsify. One species giving birth to an unrelated one would falsify it. A species appearing out of nothing would falsify it. Even if you assume "god" would not do something "obvious", a lack of correlation between dna and observed traits would falsify it. I believe you are confusing the lack of finding any actual evidence to show Evolution is false with "unfalsability".
Of course you are not going to understand any of this.
They will say "that's just micro evolution". You cannot win against these idiots. "macro evolution" is anything larger than the largest observed evolution. Perhaps eventually science will build a complete replica of Earth and use some sort of time warp to observe actual life arising, but if that experiment does not get all the way to intelligent life, they will still say "that's just micro evolution"
Microsoft is mentioned 4 times in the linked-to article:
With iPhone 2.0 software, iPhone does even more for your enterprise. It supports Microsoft Exchange ActiveSync, delivering push email, calendar, and contacts.
With support for Microsoft Word, Excel, and PowerPoint, PDF, JPEG, and now iWork, attachments can be viewed exactly as they were designed to.
And now that iPhone has built-in support for Microsoft Exchange ActiveSync, users get all the benefits of push email.
Support for Microsoft Exchange ActiveSync and industry-standard corporate security standards allows IT professionals to seamlessly integrate iPhone into their corporate environments.
Whether this counts as "running Microsoft software" is hard to say.
How do you run a posix program from cmd.exe? Or by using the Win32 exec call? I know it is possible.
Anyway posix->posix calls are probably in a tiny minority of exec calls. Far more calls from one to the other.
That is a quote from the article, not a Slashdot addition.
And I certainly did not read it as "bash Microsoft". I read it as "bash those people who forced Microsoft to default to standards because they made all you web developers have to do more work".
Microsoft is clearly, finally, doing the right thing. And I think that sentence is from somebody mad that their beloved Microsoft is being forced to do the right thing and is making up reasons why this is bad.
Wow! You have demonstrated an unbelievably low level of reading comprehension. Congratulations!
In case you missed it I will quote the second sentence: "Windows has more specialty applications that Linux lacks but that is only because it has been around in desktop use for longer then Linux has." Now you can argue whether that is true or false, but he directly addressed your question.
Unfortunatly the Posix subsystems exec call just pastes the argv back together with spaces and passes it to the next program (it has to, it runs through the normal Windows API at some point), so this solves nothing on Windows.
I do agree that having the programs do the quoting and splitting is a big security hole. And I hate to admit it (because it makes it a pain to do some command lines on Unix and I have often said it's something Windows did right) but the globbing being done by the calling program is also a good security idea, some stupid browser could be convinced perhaps to run "rm" with the file named by the web page and it sent "*"
Another annoying problem on Linux and OS/X is the mysterious belief that fonts should be measured in "points" while EVERY SINGLE OTHER GRAPHIC on the screen is measured in "pixels". Basically you are forced to work with two independent coordinate systems, which makes nice graphics incredibly hard. Windows had this problem, but they "fixed" it by defining a pixel as 4/3 of a pixel always, no matter what the resolution of the screen is. Other systems have ways to specify fonts in pixels, but the problem is GUI controls and saved preferences that insist on writing points instead, making it impossible to use the pixel api.
I don't think scaled vector graphics are going to really work until we reach 200 dpi (allowing the scale to be 2) and people figure out that fonts and graphics should use the same coordinate system (probably 1 unit == the nearest integer number of pixels to 1/96 inch).
It might be a good idea to use the term "O$$" for people writing stuff that will be released under an open license, but getting paid for it. Ideally this should be people who would not bother writing it if they were not paid. This is not a small set, in fact it is probably a quarter or so of the contributions to Linux.
This would also drive crazy the astroturfers, who always go into a childish rage when they see "M$" here. I recommend that "MS" only be used for stuff Microsoft does for no income.
I love how you underestimated the most pessimistic Linux desktop usage I have ever seen (1%) by ONE HUNDRED TIMES! Now tell me why you think you know what you are talking about.
Also they completely ignore the open source software that is produced as a paid-for job where the job is *not* to produce the open source software. If you patch the Linux kernel to get your job done, then you have contributed, yet you did exactly the same work you would have even if you had not added to open source.
That is the obvious example, and probably tiny. What is HUGE (and I know personally having done exactly that) is libraries, typically LGPL, that are produced because they are needed for a commercial project, but are then released as open source, even if the commercial project is not.
MSDOS 2 was when I thought Microsoft might do the right thing. They really modified their CP/M clone to add Unix/Posix style api, when there was absolutely no business reason, it was only to make it a nicer machine to write software for. And they did it in just a few months.
/dev and thus no file can be named "com" (until they finally fixed that crap about 2 decades after they should have).
Of course they did not really finish it and it has gone downhill ever since. I believe IBM insisted on a totally paranoid level of back-compatability, which is why we have backslashes in the filenames and no escape sequences in the stdio and no
But even seeing what they attempted it was unbelievably refreshing.
Anybody else have any thoughts about MSDOS 2.0?
I do agree that Windows 95 is their other high point. They really did some innovations in GUI design which you can see if you compare to contemporary X and Mac and NextStep:
1. The task bar which has an "icon" in it for a window *whether or not* the window was "open". *EVERYBODY* else only had the icon when the window was "iconized". This is a HUGE deal but everybody is so used to it now that they don't see it.
2. The start menu. Everybody else relied on searching through folders or using a shell command to start a program.
3. Removal of a graphical line between the resize window borders and the contents. I personally did this a bit earlier on Next (but not using Nextstep) and thought it looked really good, and was both very happy and also a bit mad to see my idea used by Microsoft. Compare Windows95 window borders with earlier ones and contemporary X ones and you will see how much cleaner they look.
Now get off my lawn.
If you have physical access to the machine and can boot a linux disk and you can read/write the main disk, then you can read it, and change anything on it. I'm sure there are millions of things you can do to be able to it to reboot as a usable Windows, or you can just stay in Linux and copy all the secret information you want off the disk.
I just can't see how this is a story. It's not a mistake on Microsoft's part. You can replace init on Linux if you want and reboot with full access.
I thought if you ate all the 72-ounce steak, your second one was free!
The test might not have had 2^6 different possible combinations. If instead it was a random arrangement of 3 on and 3 off, there would only be 20 combinations. This would make the expected percentage getting it right by random chance is 5%. This is a lot closer to the measured values so I think this is what happened.
I remember a version of this where a local band decided their name should be "Free Roast Beef". Bars would then put up a sign saying "Tonight only: Free Roast Beef".
The trains in the city were electric, not steam/coal powered.
What cars did do is replace the horse, which is really a far worse polluter. Probably much more dangerous pollution as well. Though I guess they were carbon neutral.
A lot of people have just wasted a vast amount of time contributing software to this device. They could have said this was the plan from the start and maybe those people could have concentrated on hardware drivers or interesting Windows software for it. Instead an awful lot of man years of contributed effort is wasted by this moronic decision (no, not the decision to switch to XP. The decision to, for years, lie about what direction they were going, apparently to garner publicity).
I really am sickened by this.
How do you change "MS-PL" to "MS-LPL"? You add an "L". And the result is a MORE restrictive license.
How do you change "GPL" to "LGPL"? You add an "L". And the result is a LESS restrictive license.
Are you seriously claiming that this is not purposly done by Microsoft to confuse things? I'm sorry, I don't believe you.
And you still didn't address the fact that you are ignoring the MS-LPL, MS-LRL, and MS-RSL, as well as several other licenses that they call "shared source".
use boost::filesystem (or whatever the analog is for your language/framework). It's really not hard.
Any interface that requires me to treat filenames differently from other string data is TOTALLY unacceptable. Try writing a real program some day, such as one that implements an interpreted language where users may want to store a filename into a string variable.
Furthermore, the Win32 api accepts BOTH forward and backward slash. I am quite able to write portable programs using it. My problem is that any communication to other programs, which is by far the area where I don't want to worry about "is this a filename?" is where it breaks. This is not really Microsoft's fault, it is clueless programmers writing applications. But Microsoft could help by insisiting the programs accept filenames in both versions.
I don't care if the _s functions are written by some standards organization. On the Linux side I really don't care what Theo says (he obviously has some problem with code that was written for BSD as his objections, and yours, are stupid). To the vast majority of people it is obvious that strlcpy is the right solution. strlcpy do the *right* thing: truncation is NOT a problem if all useful values are shorter than the truncation (such as when the buffer is compared to a list of tokens and no token is longer than the buffer, or when the OS has a limit such that the truncated value could not possibly name a file anyway). If you want to do something "properly" then use std string in C++. Also strlcpy returns the length of the buffer you need, so you can allocate it, this can result in MAJOR time savings if the majority of your work can be done in automatic arrays on the stack, with malloc only called in the rare too-long cases, no other solution provides this.
I do think everybody wants "how big is this value" to be part of the syntax. It makes no sense that it be in a header file. You might as well design a programming language where the actual conversion into machine instructions is contained in header files. That would be clever, but it would not be a "compiler".
The Win32 api certainly does accept forward slashes just fine. The problem is appliations and clueless programmers that don't think files ever contain forward slashes. Generally you cannot cut & paste or drag & drop or type in a filename with forward slashes into many applications, and operation system calls like getcwd cannot return strings with forward slashes without them crashing. What I want Microsoft to do is insist that the programs should all accept and work with these, or they don't get the "windows certified" label or something. I am really sick and tired of having to add tons of code to decide whether a piece of text is a filename or not and disabling all possibilities of quoting when it is. I am pretty certain there are plenty of programmers inside Microsoft who would like to fix this as well.
For drive letters, the current syntax would certainly continue to work. I just want an alternative syntax so that "/" can start an unambiguous filename. This would allow the disk structure to be duplicated on a Unix machine so that software can go back & forth. Best suggestion I have heard is to have "/A:/" be the same as "A:/". It would be really nice if readdir() of "/" list these.
MS is just being assholes about the C99 stuff. First of all they ignored the BSD strlcpy and strlcat, which are quite proper solutions. Their "standard" is strcpy_s which *throws an exception* when the buffer overflows. That is just ludricous, what it means is that programs will throw exceptions and cause a DOS rather than just truncating. Really what they are trying to do is force everybody to use Windows-specific calls. Adding underscores to a random set of C99 functions that are safe, especially snprintf, is the real giveaway that they just want to make it impossible to port code.
The C people should realize that we want "N bits" and really don't give a damn about any other considerations. I think C should support "int:22 x" to mean an integer that holds at least 22 bits, with a further guarantee that any power of 2 greater or equal to 8 means *exactly* that many bits. This typedef stuff is nonsense. And both Windows and posix should stop declaring a new foo_t type for every integer in the world, it really does help to know that two of them are the same size and the standards should enforce this by using the same type.
The answer is a list of rather simple things, but it is not what they want to hear or expect to hear (I think they expect to hear demands for open source):
#1: Fix filenames and filesystem so they match Unix. This means you use the forward slash. Refuse to "microsoft certify" any software that will not accept a pasted or typed filename with a forward slash in it, and change all the OS api that returns filenames to return forward slashes (probably with a registry setting) and again refuse to "microsoft certify" software that fails when this setting is on. And get rid of the damn drive letters (just make "/A:/" be the same as "A:/") and support UTF-8 encoding of the filenames at all times (probably by changing the "a" version of the win32 api to be hard-coded to UTF-8).
#2: Support OpenGL, meaning that by default you get at least what Mesa provides. Supporting OpenGL 1.4 only is not acceptable.
#3: Support C99 standard functions and don't make your compiler spew a lot of bogus "warnings" that you put in there to try to encourage people to change to your windows-specific functions. Remove the underscores you stuck on lots of the functions so that portable useful code cannot be written.
MS-PL, MS-RL I notice you did not list the MS-RSL (which was listed on the exact same faq page as your quote about OSS you cut & pasted, although they seem to have obfuscated it a bit by not listing the abbreviation so it was harder to read). You also don't list the MS-LPL and the MS-LRL.
These three are the ones being complained about (also the approximately 60 other liceneses that they also call "shared source", but to their credit they are trying to phase out and cut down to only the five listed here).
That fact that you blatently ignored the licenses that everybody is arguing about leads me to believe you are a shill.
The problem with the name "Public" is that it sounds too much like "public domain" which has it's own well-defined meaning.
One person above suggested "Free Source". That sounds pretty good and not easily confused with other terms.
Another name that seems to be catching on is "FOSS". This seems to rarely be used for anything other than code that can be reused.
I do agree with many posters above that "open source" when not capitalized really means "you can read the source".