At the rate of.0038 m/yr, a billion years ago it was only 3800 km closer. Out of roughly 300,000 km between us and the moon, it still doesn't make much difference.
The Halting Problem doesn't really apply. Sure, you can't automatically machine-prove everything, but that goes for mathematical theorems as much as programs. It's usually fairly simple to prove that a fairly small program will end. It's harder to prove that it will produce the correct results, but it's usually possible to prove significant parts of the program that will provide a partial proof. Killing bugs in 90% of the program beats leaving them for the users, and makes finding bugs easier.
Safety-critical systems. E.g. medical equiment needs to be 'safe', and often has to prove a certain level of testing/reliability before it is legal to sell it. You can be guaranteed that the s/w producers will be liable if an X-Ray machine gives you the wrong dose
Of course! The link above is the IEEE report on the Therac 25, the only known case of human death caused by software bugs. Once in a while, the cancer machine at the hostpital would give real big doses of radation, at seemly random times. The sad, scary thing is that all the classic software responses are there - "Let's do it [the safety/sanity checking circuits in this case] in software to save a couple bucks", "It's a hardware problem", "Here's a patch for it (that doesn't fix the whole problem, but patches a few symptons)".
Consider that the intent of putting a product under the GPL might be considered to be anticompetitive
It's not illegal to be anticompetitive if you're not a monopoly.
The restrictions of the GPL could be struck down without affecting the permissions, if the court determines this is the proper remedy.
A court could order you to release your source code into the public domain no matter what license you use. But only in extreme circumstances would a court do so no matter what license you chose.
It has been interpreted to be compatible with the GPL, but that doesn't guarantee the courts will agree with this interpretation if someone using the X11 or GNU license disagrees. It is debatable whether the advertising thing is an additional and incompatible restriction.
Please make an attempt to know what you're talking about. The X11 license doesn't have an advertising clause; that's the BSD license you're thinking about.
Hell, the GPL still hasn't ever been tested in court. There are reasons to believe that releasing software under the GPL is putting it in the public domain, and it is just one test case away from being treated as such.
What type of test case is going to remove copyright protection from GPL code? No one wants that; if GPL code loses protection, so does Microsoft code.
Picking a licence causes problems, too. The most important one is licence incompatibility: choose one, and you prevent the code from being used in projects using an incompatible licence, while public domain code can be included in projects using any licence I've heard of.
Ever heard of the X11 license? The X11 license can also be used with any license ever used, and keeps warranty protection.
Because it's not in the realm of the standard - you can use C++ on systems that natively support it and don't need name mangling, because native linkers have different requirements as to maximum length and permitted characters, and because even if you do, there's still a thousand other details that need to work right that won't magically work just because the name-mangling is correct.
As a note, the IA64 ABI did specify C++ linking, so mixing C++ from different compilers should work on Linux for the IA64.
However, when you have a precompiled C++ library that used a -different- compiler from the gcc you are currently working with (such as the Sun WSPro compiler, or an older version of gcc), this is where the headaches begin.
Is true. C++ is not binary compatible across compilers. This has absolutely nothing to do with Objective C - it's true if you're mixing a precompiled C++ library with C++. So why do you blame Objective C?
Debian is working with LSB. Why would a completely new packaging system be any better for Debian than RPM?
In any case, just say it's RPM isn't quite accurate. The LSB standard is an older version of the RPM spec, with features (like triggers) removed that would be hard to support on non-RPM systems (like Debian).
If they tried to write it in Fortran and failed, it doesn't say much for their programming ability. I have one of the first compilers for Algol68, a much more complex language than Pascal, and it was written in Fortran.
When was the last time you heard an honestly spoken conservative viewpoint on the majority of those liberal college campuses?
I go to Oklahoma State University. I frequently see the chalkings of the gay/lesbian group defaced, and sometimes those of the Pagan Student Association. I've never seen any of the chalkings for the Christian groups defaced. The last real silencing on campus was when the Regents tried to ban "The Last Temptation of Christ".
I believe that conservative viewpoints get silenced too. I do not believe that it consistently happens in most colleges.
As a college student I was quite upset that part of my "Student Activity Fee" went to groups and organizations that I had absolutely no interest in supporting (Campus Leftists, Amesty International, college democrats, gay, lesbian, and bisexual alliance, et al.)
A classically liberal college (all of them, basically) doesn't just attempt to cram an vocation into your head - the goal is to develop the full person. Part of the point of the student activity fee, hence, is that the student gets an oppertunity to hear a wide variety of different viewpoints, honestly spoken.
Larry Niven wrote a story, in a bar with the chirpr...s who remembered a civilization on Earth, before plants added free oxygen to the atmosphere, which was a poison to them (as with most creatures of that time.)
Another problem a lot of people have is that they are running their X Server at 75dpi, when in fact many modern displays are closer to 100-120dpi (mine's 120dpi...)
If I wanted nice large fonts, I'd run at 640x480 or 800x600. I think a lot of people use 75dpi fonts not because their display is 75dpi, but because they run high resolutions so they can get more on the screen.
How much hardware that old versions of Linux (Debian 1.1, say) supported are no longer supported? An upgrade should be an upgrade, and support everything that old stuff did.
Give yourself a universal translator and it don't matter what language you speak or use.
"You give my cousin a ball." is impossible, without context, to correctly translate to German (one of English's closest relatives, linguistically.) German has no word for cousin - Vetter, Kousine are for male, female cousin, respectively. "You" translates to either "Sie" if you aren't good friends, or "du" if you are.
As a further note, even human translators (which don't have to worry about the hard AI problem) don't do perfect translations everytime. There's dozens of translations of the Iliad to English, because the translator felt the previous one's hadn't been good enough. Translation relays are done, and the results are just like the game telephone - at the end, the message bears little relation to the start.
the same technology can and will be used to translate human languages into programming language
Why do mathematicians use mathematical notation instead of writing it in a human language? Because a human language is too imprecise, too clumsy and too hard to read for what they need done. There are and will be improvements in programming languages, but massive precision is necessary in programming. It's easy to write a word processor or a web browser; it's hard to get all the features specified exactly in a standards compliant manner.
I'm not quite sure what you're talking about, but part of the point of Free Software is that you can futz around with it. If you can't futz around with it, and redistribute it, then it's not Free Software.
They could probably still be released under the "non-free" section.
And they are. What did you think we were doing? It's Debian's fault that someone writes about it on Slashdot?
these are (from what I understand) mostly documents that predate the "Debian-approved" license they prefer.
They're documents that, legally, we can't modify and/or sell. That's a problem.
the volunteer effort of authors [...] of no consequence?
There's a lot of effort in the world that's misdirected. A lot of people do things that are redundant, or pointless, in an active attempt to do good. If you write a document that can't be used by free software people, then you haven't helped the free software community.
there are far fewer authors than Debian maintainers
I sure hope not. Are there only a few hundred free software documentation authors in the world?
I draw the "geekiness" line at pissing away your time writing silly crap like that for a computer. A computer is a tool of science and business, not a gaming machine. (s/calculator/computer/g)
Is that you meant to say? I'm guessing that a pencil is a tool of science and business, not a gaming tool, too. Doing anything serious on a TI-89 is a decent challenge, though less than the -83's or -85 that didn't come with a half meg of memory.
Not only has the rampant Tetris-playing caused my students to stop paying attention in class,
Really? Students will pay attention in math class without TI-89s? That's surprising; I wouldn't think a lot of students would pay attention no matter what you did.
This catches email with Korean subjects. Most of my Asian spam comes with that header; I'm told it's a Microsoft encoding that isn't even valid to use on the net, and I've never seen a valid email get caught by that filter.
Who in their right mind would use a word processor for equations?
The person who wanted to get something done. Come on, I've used Word about as often as I've used TeX. If I want to write something in Word, I open up Word and start working. If I want to write something in TeX, I run to the library for a book on TeX, or load up the Gentle Introduction to TeX. Why should someone who needs to stick one or two equations into a text, not more than two or three times a year, have to dig through a book every time they want to add an equation to something?
Yes, if you're doing serious equation work, some form of TeX is the tool. That doesn't make it any easier for the person who isn't doing serious equation work.
engineering students make up "most users"? Methinks not.
Engineers and engineering students make up some users. Home users are another group of users. Linguists and anthropologists and historians are another group. Buisness executives are another group. And each group has their own demands on what they want.
For math-heavy texts, one would probably better off with a tool devoted to such things
Sure, if your life is writing articles to the Journal of the AMS, use LyX. But if you want to write up some physics notes one day, type a paper for the Ancient World the next, and write a letter home, it's a lot more convientent to have one tool to learn that does all of them okay, then to learn three.
Even programmers show this habit. Why do you think most programming on Linux is done in Perl and C? Because they're present on most systems, and you already know them.
people want a tool that lets them do the things they need to do in a simple manner
And what do they need to do? There are users for each of the features in Word. Anything you consider superflous in Word, would leave some people calling Microsoft rude things if it disappeared in the next version.
I'm sure that mediveal Korean support would be considered bloat, but, as I understand it, Word is the only wordprocessor that will handle mediveal Korean characters - it's the only tool that lets Korean historians do the things they need to do in a simple manner. Should Microsoft just blow off that market? How would that help them in other markets?
complexity and price are stronger negatives than lack of features that they never use
Complexity, as in making it hard to use and learn, is a strong negative. I've never heard an end user complaint about extra features, and as for me, I like using systems that I'm comfortable can support my needs - that I probably won't be in urgent need to do something that it can't support.
I'm not sure where price comes in here. For many users, both Word and Abiword are free, legal or not. Price hasn't stopped Word yet.
> Sure. Much of the so-called "functionality" of modren commerical word processors is, for most users, nothing but bloat.
Like what? The equation editor - I know engineering students who find that very useful for school work. Full Unicode functionality - aka support for 1/5 of the world's population's native languages? Multilingual spellchecking? What?
>And if you need more, you don't want a word processor, you want a document preparation system
Most people want a simple, WYSWIG, omnipurpose tool, so that's what they use, regardless of what computer geeks think is right.
> Which commercial products? It certainly surpasses Notepad.
It's not in competition with Notepad. Notepad is a text editor, like vim, not a wordprocessor.
> When I checked several months ago it didn't measure up to WordPad, but that was several months ago.
But I don't know anyone who uses WordPad as a wordprocessor. Everyone I know uses Works or Word or Wordperfect. The only time I've ever heard of any using WordPad was to test Windows 2000's Unicode systems, not to actually write something. In other words, it doesn't measure up to a program that was so inferior to the competition that no one uses it even though it comes with Windows.
> since Sept. 11 all media outlets are rethinking what is and what isn't safe to release to the public
Is there any information the hijackers actually used such information? Or is this just raw terror? Or something more insidious - remember that the Pentagon Papers and the Nixon's tapes were matters of national security too.
So you started dishing a bunch of general complaints about the nature of Debian, and couldn't take it when someone disagreed with you, and didn't care to up the politeness level. Um, yeah, people on Debian lists complaining that Debian is doing everything wrong and they should do it my way don't tend to get much respect, just like everywhere else in the world.
For reference, the message is http://www.geocrawler.com/mail/msg.php3?msg_id=1 28 1814&list=199
Slashdot Mirror
At the rate of .0038 m/yr, a billion years ago it was only 3800 km closer. Out of roughly 300,000 km between us and the moon, it still doesn't make much difference.
The Halting Problem doesn't really apply. Sure, you can't automatically machine-prove everything, but that goes for mathematical theorems as much as programs. It's usually fairly simple to prove that a fairly small program will end. It's harder to prove that it will produce the correct results, but it's usually possible to prove significant parts of the program that will provide a partial proof. Killing bugs in 90% of the program beats leaving them for the users, and makes finding bugs easier.
Safety-critical systems. E.g. medical equiment needs to be 'safe', and often has to prove a certain level of testing/reliability before it is legal to sell it. You can be guaranteed that the s/w producers will be liable if an X-Ray machine gives you the wrong dose
Of course!
The link above is the IEEE report on the Therac 25, the only known case of human death caused by software bugs. Once in a while, the cancer machine at the hostpital would give real big doses of radation, at seemly random times. The sad, scary thing is that all the classic software responses are there - "Let's do it [the safety/sanity checking circuits in this case] in software to save a couple bucks", "It's a hardware problem", "Here's a patch for it (that doesn't fix the whole problem, but patches a few symptons)".
Consider that the intent of putting a product under the GPL might be considered to be anticompetitive
It's not illegal to be anticompetitive if you're not a monopoly.
The restrictions of the GPL could be struck down without affecting the permissions, if the court determines this is the proper remedy.
A court could order you to release your source code into the public domain no matter what license you use. But only in extreme circumstances would a court do so no matter what license you chose.
It has been interpreted to be compatible with the GPL, but that doesn't guarantee the courts will agree with this interpretation if someone using the X11 or GNU license disagrees. It is debatable whether the advertising thing is an additional and incompatible restriction.
Please make an attempt to know what you're talking about. The X11 license doesn't have an advertising clause; that's the BSD license you're thinking about.
Hell, the GPL still hasn't ever been tested in court. There are reasons to believe that releasing software under the GPL is putting it in the public domain, and it is just one test case away from being treated as such.
What type of test case is going to remove copyright protection from GPL code? No one wants that; if GPL code loses protection, so does Microsoft code.
Picking a licence causes problems, too. The most important one is licence incompatibility: choose one, and you prevent the code from being used in projects using an incompatible licence, while public domain code can be included in projects using any licence I've heard of.
Ever heard of the X11 license? The X11 license can also be used with any license ever used, and keeps warranty protection.
Because it's not in the realm of the standard - you can use C++ on systems that natively support it and don't need name mangling, because native linkers have different requirements as to maximum length and permitted characters, and because even if you do, there's still a thousand other details that need to work right that won't magically work just because the name-mangling is correct.
As a note, the IA64 ABI did specify C++ linking, so mixing C++ from different compilers should work on Linux for the IA64.
Huh?
However, when you have a precompiled C++ library that used a -different- compiler from the gcc you are currently working with (such as the Sun WSPro compiler, or an older version of gcc), this is where the headaches begin.
Is true. C++ is not binary compatible across compilers. This has absolutely nothing to do with Objective C - it's true if you're mixing a precompiled C++ library with C++. So why do you blame Objective C?
Debian is working with LSB. Why would a completely new packaging system be any better for Debian than RPM?
In any case, just say it's RPM isn't quite accurate. The LSB standard is an older version of the RPM spec, with features (like triggers) removed that would be hard to support on non-RPM systems (like Debian).
If they tried to write it in Fortran and failed, it doesn't say much for their programming ability. I have one of the first compilers for Algol68, a much more complex language than Pascal, and it was written in Fortran.
When was the last time you heard an honestly spoken conservative viewpoint on the majority of those liberal college campuses?
I go to Oklahoma State University. I frequently see the chalkings of the gay/lesbian group defaced, and sometimes those of the Pagan Student Association. I've never seen any of the chalkings for the Christian groups defaced. The last real silencing on campus was when the Regents tried to ban "The Last Temptation of Christ".
I believe that conservative viewpoints get silenced too. I do not believe that it consistently happens in most colleges.
As a college student I was quite upset that part of my "Student Activity Fee" went to groups and organizations that I had absolutely no interest in supporting (Campus Leftists, Amesty International, college democrats, gay, lesbian, and bisexual alliance, et al.)
A classically liberal college (all of them, basically) doesn't just attempt to cram an vocation into your head - the goal is to develop the full person. Part of the point of the student activity fee, hence, is that the student gets an oppertunity to hear a wide variety of different viewpoints, honestly spoken.
Larry Niven wrote a story, in a bar with the chirpr...s who remembered a civilization on Earth, before plants added free oxygen to the atmosphere, which was a poison to them (as with most creatures of that time.)
Another problem a lot of people have is that they are running their X Server at 75dpi, when in fact many modern displays are closer to 100-120dpi (mine's 120dpi...)
If I wanted nice large fonts, I'd run at 640x480 or 800x600. I think a lot of people use 75dpi fonts not because their display is 75dpi, but because they run high resolutions so they can get more on the screen.
How much hardware that old versions of Linux (Debian 1.1, say) supported are no longer supported? An upgrade should be an upgrade, and support everything that old stuff did.
Give yourself a universal translator and it don't matter what language you speak or use.
"You give my cousin a ball." is impossible, without context, to correctly translate to German (one of English's closest relatives, linguistically.) German has no word for cousin - Vetter, Kousine are for male, female cousin, respectively. "You" translates to either "Sie" if you aren't good friends, or "du" if you are.
As a further note, even human translators (which don't have to worry about the hard AI problem) don't do perfect translations everytime. There's dozens of translations of the Iliad to English, because the translator felt the previous one's hadn't been good enough. Translation relays are done, and the results are just like the game telephone - at the end, the message bears little relation to the start.
the same technology can and will be used to translate human languages into programming language
Why do mathematicians use mathematical notation instead of writing it in a human language? Because a human language is too imprecise, too clumsy and too hard to read for what they need done. There are and will be improvements in programming languages, but massive precision is necessary in programming. It's easy to write a word processor or a web browser; it's hard to get all the features specified exactly in a standards compliant manner.
I'm not quite sure what you're talking about, but part of the point of Free Software is that you can futz around with it. If you can't futz around with it, and redistribute it, then it's not Free Software.
They could probably still be released under the "non-free" section.
And they are. What did you think we were doing? It's Debian's fault that someone writes about it on Slashdot?
these are (from what I understand) mostly documents that predate the "Debian-approved" license they prefer.
They're documents that, legally, we can't modify and/or sell. That's a problem.
the volunteer effort of authors [...] of no consequence?
There's a lot of effort in the world that's misdirected. A lot of people do things that are redundant, or pointless, in an active attempt to do good. If you write a document that can't be used by free software people, then you haven't helped the free software community.
there are far fewer authors than Debian maintainers
I sure hope not. Are there only a few hundred free software documentation authors in the world?
I draw the "geekiness" line at pissing away your time writing silly crap like that for a computer. A computer is a tool of science and business, not a gaming machine. (s/calculator/computer/g)
Is that you meant to say? I'm guessing that a pencil is a tool of science and business, not a gaming tool, too. Doing anything serious on a TI-89 is a decent challenge, though less than the -83's or -85 that didn't come with a half meg of memory.
Not only has the rampant Tetris-playing caused my students to stop paying attention in class,
Really? Students will pay attention in math class without TI-89s? That's surprising; I wouldn't think a lot of students would pay attention no matter what you did.
:0:
*^Subject:.*ks_c_5601*
spam
This catches email with Korean subjects. Most of my Asian spam comes with that header; I'm told it's a Microsoft encoding that isn't even valid to use on the net, and I've never seen a valid email get caught by that filter.
Who in their right mind would use a word processor for equations?
The person who wanted to get something done. Come on, I've used Word about as often as I've used TeX. If I want to write something in Word, I open up Word and start working. If I want to write something in TeX, I run to the library for a book on TeX, or load up the Gentle Introduction to TeX. Why should someone who needs to stick one or two equations into a text, not more than two or three times a year, have to dig through a book every time they want to add an equation to something?
Yes, if you're doing serious equation work, some form of TeX is the tool. That doesn't make it any easier for the person who isn't doing serious equation work.
engineering students make up "most users"? Methinks not.
Engineers and engineering students make up some users. Home users are another group of users. Linguists and anthropologists and historians are another group. Buisness executives are another group. And each group has their own demands on what they want.
For math-heavy texts, one would probably better off with a tool devoted to such things
Sure, if your life is writing articles to the Journal of the AMS, use LyX. But if you want to write up some physics notes one day, type a paper for the Ancient World the next, and write a letter home, it's a lot more convientent to have one tool to learn that does all of them okay, then to learn three.
Even programmers show this habit. Why do you think most programming on Linux is done in Perl and C? Because they're present on most systems, and you already know them.
people want a tool that lets them do the things they need to do in a simple manner
And what do they need to do? There are users for each of the features in Word. Anything you consider superflous in Word, would leave some people calling Microsoft rude things if it disappeared in the next version.
I'm sure that mediveal Korean support would be considered bloat, but, as I understand it, Word is the only wordprocessor that will handle mediveal Korean characters - it's the only tool that lets Korean historians do the things they need to do in a simple manner. Should Microsoft just blow off that market? How would that help them in other markets?
complexity and price are stronger negatives than lack of features that they never use
Complexity, as in making it hard to use and learn, is a strong negative. I've never heard an end user complaint about extra features, and as for me, I like using systems that I'm comfortable can support my needs - that I probably won't be in urgent need to do something that it can't support.
I'm not sure where price comes in here. For many users, both Word and Abiword are free, legal or not. Price hasn't stopped Word yet.
> Sure. Much of the so-called "functionality" of modren commerical word processors is, for most users, nothing but bloat.
Like what? The equation editor - I know engineering students who find that very useful for school work. Full Unicode functionality - aka support for 1/5 of the world's population's native languages? Multilingual spellchecking? What?
>And if you need more, you don't want a word processor, you want a document preparation system
Most people want a simple, WYSWIG, omnipurpose tool, so that's what they use, regardless of what computer geeks think is right.
> Which commercial products? It certainly surpasses Notepad.
It's not in competition with Notepad. Notepad is a text editor, like vim, not a wordprocessor.
> When I checked several months ago it didn't measure up to WordPad, but that was several months ago.
But I don't know anyone who uses WordPad as a wordprocessor. Everyone I know uses Works or Word or Wordperfect. The only time I've ever heard of any using WordPad was to test Windows 2000's Unicode systems, not to actually write something. In other words, it doesn't measure up to a program that was so inferior to the competition that no one uses it even though it comes with Windows.
> since Sept. 11 all media outlets are rethinking what is and what isn't safe to release to the public
Is there any information the hijackers actually used such information? Or is this just raw terror? Or something more insidious - remember that the Pentagon Papers and the Nixon's tapes were matters of national security too.
So you started dishing a bunch of general complaints about the nature of Debian, and couldn't take it when someone disagreed with you, and didn't care to up the politeness level. Um, yeah, people on Debian lists complaining that Debian is doing everything wrong and they should do it my way don't tend to get much respect, just like everywhere else in the world.
1 28 1814&list=199
For reference, the message is
http://www.geocrawler.com/mail/msg.php3?msg_id=
*