Something seems fishy with this whole "just replace one cell at a time" philosophy. Why one, why not two? Would you really destroy conscienciousness if you replaced two of your cells with machine-based dopplegangers? If not two, what about three? five? 300? 5,000,000? Is there some magic number of cells that you can replace before destroying consciousness?
I will try to respond to the better points in your comment. (Some of your comments are outside the domain of this discussion, others are completely unintelligible).
I resent your rhetoric to call education developed for hundreds of years as "archaic", and that "psychological theory" would be more correct because it is based on "latter half of the century".
Truth is the subject of philosophy. In science, psychology included, the most recent theories are best ones we have. NCTM standards are based on the best scientific research to date.
Anyway, even this guy on whose work everything is based on, claims that not everything can be learned by inventing it.
Indeed, this is discussed in the NCTM standards, and it incorporated to NCTM-based curricula. Certainly, we cannot expect that our students discover all of mathematics entirely on their own. In IMP, for instance, students first explore the concepts behind "spread" in a normal distribution, and they try various measures of spread (a couple intuitive suggestions are given to them), but eventually, when they get to standard deviation the formula is basically just given to students. However, now, students understand what standard deviation does and that its based on their simpler, more intuitive ideas of spread. There is no invention of standard deviation. Likewise, students do not invent their own terminology, and they are not expected to miraculously derive standard mathematical notation.
The real problem lies in the other parts of cooperative learning: groupwork instead of individual learning, portfolios instead of testing, counseling and Prozac instead of tracking.
These are false dichotomies. In fact, NCTM uses individual work AND groupwork, portfolios AND testing. As for counselling and Prozac -- these are medical decisions, and certainly not part of any NCTM standard.
The use of portfolios is really messed up. They are just a way for teachers to keep inflating the the grades beyond giving A's to all kids. It is also a way for lousy teachers to hide the fact that kids are becoming more and more ignorant.
Though I think your reasoning is overgeneralized and mostly unfounded, I will comment that I am fairly soft on the use of portfolios in the classroom. I think they're nice to students to summarize work, and gain a sense of accomplishment. I am skeptical of any significant value beyond this, though.
We need to verify that the kids are really learning and the new education system is just not rhetoric. The fact that teacher associations are against all standardized testing speaks for itself -- they have something to hide.
I'm not sure what official union policy is on this one. But, there are many problems with standardized testing. The scores are often misused to label a district as inadequate. Reporters do not take into account the fact that a district may have 80% students in poverty, and 40% english language learners before comparing said district to another. Also, state/federal testing requirements often come with little additional funding to prepare/administer/evaluate a test. And, most frustrating for me, tests take away time from other things. I lose more than a week of teaching to standardized testing, and the worse a student does, the more time she/he's taken out of class. Finally, many standardized tests are redundant, or unnecessarily lengthy. All these factors can do more harm than good to a student's education. I feel some testing is important, but America's on a testing frenzy as of late.
If your view of learning advocates critical thinking, why is it that only one critical thinking about this new way of learning comes from the outside of the Education community?
The education community does question the benifits and costs of NCTM standards. It's asinine to think that we blindly go about nodding our heads because someone, somewhere told us this is good. We debate and discuss within the mathematics community and within the education community. Those outside the education community are probably more cynical because they haven't read the NCTM standards, or have little experience teaching NCTM-based curricula. But, I applaude such healthy skepticism, and encourage you to continue to expore the issue. Read the standards, and observe a classroom in which they are implemented. Or, better yet, conduct a study in which student performance with NCTM and non-NCTM curricula is examined. Of course, you will need to adjust for student/teacher/environmental variations... very difficult.
Before it was public, government subsidized, American education was non-compulsory and only the wealthy were educated. Both Thomas Jefferson and Horace Mann felt that public, free education would be the "great equalizer" in society.
The NCTM-based curriculum is different. Some teachers and college professors believe it to be weak on mathematics because it doesn't look like the curriculum they grew up with. Traditional curriculum (teacher does a couple examples, students practice solving 30+ problems similar) has not been good enough though.
The new curriculum, based on psychology and education research from the latter half of this century, focuses on understanding in addition to the traditional acquisition of skills. It is mathematics rich with connections to other areas, and deep in content. Students start in 6th grade learning basic algebraic concepts, number theory, geometry, probability, etc. Obviously mastery of all these concepts does not happen in a single year. In fact, the curriculum spirals around the same concepts, building new understanding and making new connections with each pass so that, ideally, when students graduate their skills AND understanding will be better than that of previous generations.
Sometimes this math is called "Fuzzy Math" or the "New 'New Math'". Some educators, professionals, parents, and children feel the curriculum is weak on "real math." My concerns were similar before I started teaching the Interactive Mathematics Program (IMP).
Between 9th and 10th grade, students master basic algebra, learn the basics of the trig functions, work with standard deviation and the chi-squared measure, build and solve and maximum profit linear programming (something most math majors don't do until grad school), derive and prove the pythagorean theorem, work with exponential and logarithmic functions, do all sorts of number-theory related problems, and so much more.
Still IMP and other standards-based curricula have their problems. In my opinion, although there's plenty of problem-solving and understanding-based activities, there needs to be more traditional skill work. I supplement my lessons with such work where appropriate. Any teacher worth their stuff would do the same. Additionally, the curricula is very wordy, which is fine for middle-class suburbanites, but when you're teaching in a city where 25% of the students don't speak english as their first language, and 75% are in poverty (typically correlated with smaller vocab and weaker reading/writing skills), a wordy curriculum is just one more thing making it tough to teach/learn math.
In sum, there's a lot of hostility from the non-math-teacher world toward this new curriculum because it's so different. But, with the abismal performance of American mathematics when compared internationally, it can't be business as usual. The curriculum is already working well in the classes I've seen. And the research points to positive improvements after curriculum implementation (no large study has been completed as far as I'm aware). NCTM-based curricula is no panacea, but it's a definite improvement over the more archaic traditional curricula.
Yeah! I actually used this when teaching stats to my 11th graders. They linearly regressed the past data to predict what the record would be. Everybody in the class gasped when they saw 50.5!
Learning from books is all well and good, but I truly feel that for one to fully develop one's mathematical abilities, one must be part of an academic community, engaging in academic dialogue with living, breathing "math-people." Whether you are a Ph.D., or a middle-schooler, the BEST way to learn mathematics is by actively and routinely doing it with others.
Anyway, if you're serious about learning mathematics but scared of the cost, go to your nearest University and just sit in on the class. Listen to the lecture, ask questions, take notes, do homeworks, take tests, just don't pay. I teach mathematics at the college level... if a students showed up in my classroom who seriously wanted to learn, but didn't want to pay tuition, I would be more than supportive of his/her presence in my class. A number of my colleagues feel the same way -- learning should transcend economic boundaries. (On the other hand, though, some of my peers in our University's Physics department like the fact that tuition weeds out the middle-aged crackpots with their pseudo-scientific TOEs).
For math books freely downloadable online, dig around at http://www.math.umn.edu/~garrett
I think the point of the article was not that software shouldn't have this sort of functionality, but rather that software should not have such non-security-conscience enabled by default, allowing Users or Administrators to choose whether to turn the functionality on if they deem the benefits outweigh the costs.
Take OpenBSD as an example. The good folks at OpenBSD review their code, flush out the bugs, and turn off non-essential stuff in their default install. BUT, they also document everything well and make configuration easy enough so that if you want to turn on something risky you can. Also, they have great linux emulation and an excellent port system so that you can install tons of software, both secure and insecure. They leave it up to the admin to decide between functionality and security, but the overwhelming default is security over functionality.
I think this is essentially what Schneier envisions.
Unlike previous changes in direction, Microsoft cannot blitzkrieg itself into a Secure OS. As the Security Community states again and again: computer security takes time. It takes line-by-line code auditing, careful design decisions, and years of testing and analysis.
If Microsoft is serious about it, securing Windows is definitely a Good Thing (tm) for Microsoft, *nix, and the computing community at large.
First, as Schneier et al. pointed out, good security in a Windows OS will take a complete redesign from the ground up. There's NO WAY they can hack on patches to Windows and claim that they're serious about operating system security. Furthermore, Microsoft will NEVER let the security community examine "trade secrets" from an already released copy of Windows (if only because they'd have too many fires to put out after bugs were found). And if Microsoft is planning to doall that Schneier et al. suggest, they won't be able to neatly build on existing NT/XP code. Thus, redesign seems like their only option.
Why is a complete design of Windows a good thing for Linux/BSD?
An OS takes a lot of time to build. And if security is the focus, it will need an especially long development cycle. Linux/BSD can catch up during this time. We can add features (securely, of course) and fix our rather minor security/stability problems while MS has to start from scratch.
A more secure network OS is good for everyone. Less hacks mean less money lost to computer crime, more consumer confidence in online transactions, less work for System/Network admins, and more protection and privacy for Internet denizens.
Microsoft will need to make parts of the new OS public. Although the Open Source community may not be able to use these parts directly (many will probably remain propreitary), we can still LEARN from them, and from the Security Community's feedback about them. More information is never a bad thing.
Because MS will feel strapped for time, it will probably "steal" large parts of the existing BSD's in their new OS. Even if they do so unofficially, MS will probably need people familiar with Linux/BSD to modify BSD code for their needs. Over time, this could generate a more friendly relationship between the OpenSource community and Microsoft.
Of course, all of this is pure speculation. Still, no matter what Microsoft does, at worst we'll be unaffected by Microsoft's new move toward a secure OS. But, at best, I believe the Open Source community could gain a lot.
Although this conversation is only a few short steps from namecalling, I'd like to point out that fundamentally, the issues being discussed here are tremendously important to scientists, ethicists, and human-kind in general.
At what point does a living organism become autonomous? At what point can we consider an human cell, a full-fledged human? Is it ethical to create a human-based organism just to destroy it?
It is practically impossible to teach good programming style to students that
have had prior exposure to BASIC; as potential programmers they are mentally mutilated beyond hope of regeneration.
and this:
The use of COBOL cripples the mind; its teaching should, therefore, be
regarded as a criminal offense.
Our CS department has converted to Java as an introductory language to teach undergrads. We've found that most students find the language pleasant and fairly easy to learn (as opposed to Scheme, a language which most pupils and even some TA's dread... it made for a good "weed-out" course but not much else). Java provides a solid OO foundation for further software development, and enough C-like syntax so that if a student needs to learn C or C++, they can do so quickly. And though algorithms and data stuctures can be taught in almost any language, Java's clear and consistent syntax, and automatic memory management makes teaching the subject a lot easier than when I used C++ or C (ugh).
Still, from a more pragmatic perspective, Java sucks. We can't very well teach an OS course using Java (C is still preferred). When doing any major software engineering projects, Java is too slow to make it a very viable choice. And teaching Networking with Java is just too easy to be of any value to students (I use C again).
So Java is not universally used in all areas of CS instruction (at least in my classes). But, I find that more students learn more CS when using Java as an introductory language. And when it's time to shift to "real world" programming, it's not too difficult for them.
That being said, I would have to say that the switch from C++ to Java as an AP Computer Science Language is definitely a good move. And using Java in the college classroom is also a good thing for CS instruction (though less definitely so). However, I think its important for CS instuctors to force their students to learn "real-world" programming languages such as C, C++, and Perl. The more languages a student learns, the better they will adapt to future trends in industry.
It's ironical. Like rain on a rainy day. Like a free Unix for which one pays. It's the good advice that Windows just didn't take. Who'd 've thought it figured.
Re:I see they fixed lots of security issues...
on
OpenBSD 2.9 Released
·
· Score: 1
These Buffer Overflows were local holes (i.e. you had to have to have a foothold on the box to take advantage of it).
For the past four years, OpenBSD has had no remote holes in the default install.
What Alex says is right on the nose (i.e. mod that comment up!).
I too switched from Linux to OpenBSD at home almost 3 years ago, and have been using it for various projects at work for the past 2 years. This is what I've discovered:
Major security problems are few and far between (and usually not in the default install). I sleep a lot easier at night knowing that there's little chance my machines will be cracked.
Documentation is your best friend. Almost everything you need to know is in the FAQ or the man pages.
There is a general aura of pissiness from the community. You're expected to thoroughly research every problem before asking for their help. Like I said, the documentation is your best friend, not fellow OpenBSDers. This certainly encouraged me to be very self-sufficient, but I spent many an hour scouring documentation looking for help with undocumented problems.
The ports collection has gotten some 40,000% better since I first started using OpenBSD. Still, it's up to YOU to keep the ports current and patched.
Theo's attitude wears on me. Granted I usually agree with his decisions for the OS, I still feel that he's a powderkeg ready to blow, and I'll be left with an OS that I no longer love. As the above comment mentioned, Theo regards OpenBSD as his toy. His loyalties lie with himself, not with his userbase. That is, I'd classify him as a dictator, but I'm not sure he's so benevolent...
In conclusion, I'd say trying OpenBSD is something every geek should do. But, admittedly, my loyalty to the OS (Theo) is waning, and I'm beginning to think FreeBSD might be a better choice.
Whatever the case, I'm sure Theo doesn't give a good god-damn...
Hang on, so Theo et al. in all their line by line code auditing never noticed that IPF didn't fall under the BSD license??!
How many times has OpenBSD code been audited?!?
How long has IPF been a part of OpenBSD??!
In the upcoming volume IV of Knuth's TAOCP, Seminumerical Searches for Hoaring Triplets, he conjectures the following about the Theory of Programming Languages:
Computers are ultimately a tool for their users. As such, modern programming languages should allow the users of these inherently unwieldy tools means of effective and efficient control. In volume 1 of TAOCP, we proved that all computer uses fall into one of two categories: searching for pr0n and viewing pr0n. Thus, any modern computer language should have as its core requirement the shortest semantic expressions for searching and viewing pr0n. (As usual, I will cheerfully pay $2.56 to the first finder of any pr0n that I have not already found).
Along these lines, I pose the following questions:
How effective is python at finding me pr0n?
What is the shortest python program that you know of that will allow me to veiw such pr0n?
I don't think anyone is arguing Linux's popularity among Unixens (Unixers? Eunuchs?). The point is whether it is some sort of de facto standard like Apache has become in webservers, TCP/IP in networking protocols, and Perl in scripting languages (yes, that was a red herring).
If you ask me (which you didn't, but you should have!) I really don't think that this is the case. As much as Linux has contributed in technology to other Unix variants, it has stolen. HP-Unix, Tru64 (OSF/1), AIX, *BSD, and Solaris are still THRIVING in both corporate and academic circles. Indeed, the systems programming classes taught in our Computer Science department are Solaris-biased, not Linux. The Unix of choice for work in our Physics department is NOT Linux, but Tru64.
Yes, I agree that Linux is popular. And it is generally becoming true that where you find Unix, you will find some Linux. But if Linux is a de facto standard, then it is only the case among slashdotters. It is no more the de facto standard of Unix, than RedHat is the de facto standard of Linux distributions.
Just for shits and giggles here's the 10 commandments run through Babelfish a couple times (English ->Span ->French->German->English).
Now imagine what 4000 years of the same thing has done to the original version of The Bible...:)
Emigration 20:1-17
spake king James version
1 and God all this word, say
2 I gentleman thy gott to have that thee track Egypt, house bondage. shalt
3 Thou outside to bring every other God not to have, before does not shalt
4 Thou to arrive around thee each load picture or each similarity to make each thing at skies or at earth under or at water under earth: from
5 Thou serves shalt neither arqueamien to thyself downward at them nor them: for Mr. Deity of I thy jealous God, which visits iniquity from the family members on the children, the third and fourth generation of them the hate i; to show
6 and at the thousands compassion of them that the i and maintenance my
7 entries shalt auftraege of Thou, fallen in love, not the name of Mr. Deity thy in unnecessary; for GENTLEMAN the taketh at innocent at it will not bring its name in unnecessary up to date.
8 remembers on on day sabbath to bring over it santo up to date the work thou shalt from
9 six days and from the characters the whole thy work:
10, but the sieved day thy is sabbath from Mr. Deity: in it shalt thou, in order to execute each work, thou, neither maidservant manservant threads thy, neither thy foreign and neither thy and nor thy thy and won thy and neither the girl, who is in the thy tueren: in six days the GENTLEMAN made
11 close the sky and the earth, the sea and everything in them is, and the sevened day rested: the wherefore GENTLEMAN approved the day sabbath, and he sanctified more thy father from 12 honours and from more thy mutter: day time does not out thy long power on track the thee thy giveth Mr. Deity tearing off shalt
13 Thou number shalt
14 Thou, in order adulterous confidence shalt
15 Thou not to make, around a false witness bear shalt
16 Thou not against neighbour thy. house
17 Thou shalt nonneighboring ovetthy to steal wife neighbour thy covet shalt thou not, neither its manservant neither its, neither its cattle, neither its donkey, nor each thing maidservant neighbouring thy.
Actually, there's a nice piece about tele-immersion and telepresence in this month's Scientific American written by the "father of Virtual Reality," Jaron Lanier.
According to the article, this technology's intimate relationship with Internet2 comes mostly from the fact that there were very few applications around which NEEDED Internet2's impressive network stats to actually run. Consequently, the peeps at I2 contacted Jaron to lead up the project. And... well, you can read about it here.
Actually, rumor amongst the underground is that someone (possibly a non-australian government agency) fucked up a Carnivore-esque program running on the cable so that it doesn't pass IP packets anymore.
This is what I don't understand: if our goal is to expand the amount of domain names available, why are businesses allowed to register every TLD associated with anything they have a trademark for? Allowing companies to do things like: microsoft.com, microsoft.biz, microsoft.info, microsoft.name, microsoft.coop, etc. effectively defeats the purpose of having multiple TLDs.
I think something needs to be done by ICANN to prevent this, Dammit. Otherwise a domain name crunch is really not helped by adding more TLDs.
Slashdot Mirror
Something seems fishy with this whole "just replace one cell at a time" philosophy. Why one, why not two? Would you really destroy conscienciousness if you replaced two of your cells with machine-based dopplegangers? If not two, what about three? five? 300? 5,000,000? Is there some magic number of cells that you can replace before destroying consciousness?
I will try to respond to the better points in your comment. (Some of your comments are outside the domain of this discussion, others are completely unintelligible).
Truth is the subject of philosophy. In science, psychology included, the most recent theories are best ones we have. NCTM standards are based on the best scientific research to date.
Indeed, this is discussed in the NCTM standards, and it incorporated to NCTM-based curricula. Certainly, we cannot expect that our students discover all of mathematics entirely on their own. In IMP, for instance, students first explore the concepts behind "spread" in a normal distribution, and they try various measures of spread (a couple intuitive suggestions are given to them), but eventually, when they get to standard deviation the formula is basically just given to students. However, now, students understand what standard deviation does and that its based on their simpler, more intuitive ideas of spread. There is no invention of standard deviation. Likewise, students do not invent their own terminology, and they are not expected to miraculously derive standard mathematical notation.
These are false dichotomies. In fact, NCTM uses individual work AND groupwork, portfolios AND testing. As for counselling and Prozac -- these are medical decisions, and certainly not part of any NCTM standard.
Though I think your reasoning is overgeneralized and mostly unfounded, I will comment that I am fairly soft on the use of portfolios in the classroom. I think they're nice to students to summarize work, and gain a sense of accomplishment. I am skeptical of any significant value beyond this, though.
I'm not sure what official union policy is on this one. But, there are many problems with standardized testing. The scores are often misused to label a district as inadequate. Reporters do not take into account the fact that a district may have 80% students in poverty, and 40% english language learners before comparing said district to another. Also, state/federal testing requirements often come with little additional funding to prepare/administer/evaluate a test. And, most frustrating for me, tests take away time from other things. I lose more than a week of teaching to standardized testing, and the worse a student does, the more time she/he's taken out of class. Finally, many standardized tests are redundant, or unnecessarily lengthy. All these factors can do more harm than good to a student's education. I feel some testing is important, but America's on a testing frenzy as of late.
The education community does question the benifits and costs of NCTM standards. It's asinine to think that we blindly go about nodding our heads because someone, somewhere told us this is good. We debate and discuss within the mathematics community and within the education community. Those outside the education community are probably more cynical because they haven't read the NCTM standards, or have little experience teaching NCTM-based curricula. But, I applaude such healthy skepticism, and encourage you to continue to expore the issue. Read the standards, and observe a classroom in which they are implemented. Or, better yet, conduct a study in which student performance with NCTM and non-NCTM curricula is examined. Of course, you will need to adjust for student/teacher/environmental variations... very difficult.
Before it was public, government subsidized, American education was non-compulsory and only the wealthy were educated. Both Thomas Jefferson and Horace Mann felt that public, free education would be the "great equalizer" in society.
In an effort to overcome our country's mathematics woes, the National Council of Teachers of Mathematics (NCTM) put together a monumental group of standards and principles revolutionizing the way that Mathematics is taught at the High School level.
The NCTM-based curriculum is different. Some teachers and college professors believe it to be weak on mathematics because it doesn't look like the curriculum they grew up with. Traditional curriculum (teacher does a couple examples, students practice solving 30+ problems similar) has not been good enough though.
The new curriculum, based on psychology and education research from the latter half of this century, focuses on understanding in addition to the traditional acquisition of skills. It is mathematics rich with connections to other areas, and deep in content. Students start in 6th grade learning basic algebraic concepts, number theory, geometry, probability, etc. Obviously mastery of all these concepts does not happen in a single year. In fact, the curriculum spirals around the same concepts, building new understanding and making new connections with each pass so that, ideally, when students graduate their skills AND understanding will be better than that of previous generations.
Sometimes this math is called "Fuzzy Math" or the "New 'New Math'". Some educators, professionals, parents, and children feel the curriculum is weak on "real math." My concerns were similar before I started teaching the Interactive Mathematics Program (IMP).
Between 9th and 10th grade, students master basic algebra, learn the basics of the trig functions, work with standard deviation and the chi-squared measure, build and solve and maximum profit linear programming (something most math majors don't do until grad school), derive and prove the pythagorean theorem, work with exponential and logarithmic functions, do all sorts of number-theory related problems, and so much more. Still IMP and other standards-based curricula have their problems. In my opinion, although there's plenty of problem-solving and understanding-based activities, there needs to be more traditional skill work. I supplement my lessons with such work where appropriate. Any teacher worth their stuff would do the same. Additionally, the curricula is very wordy, which is fine for middle-class suburbanites, but when you're teaching in a city where 25% of the students don't speak english as their first language, and 75% are in poverty (typically correlated with smaller vocab and weaker reading/writing skills), a wordy curriculum is just one more thing making it tough to teach/learn math. In sum, there's a lot of hostility from the non-math-teacher world toward this new curriculum because it's so different. But, with the abismal performance of American mathematics when compared internationally, it can't be business as usual. The curriculum is already working well in the classes I've seen. And the research points to positive improvements after curriculum implementation (no large study has been completed as far as I'm aware). NCTM-based curricula is no panacea, but it's a definite improvement over the more archaic traditional curricula.
Yeah! I actually used this when teaching stats to my 11th graders. They linearly regressed the past data to predict what the record would be. Everybody in the class gasped when they saw 50.5!
Off-topic -2
Anyway, if you're serious about learning mathematics but scared of the cost, go to your nearest University and just sit in on the class. Listen to the lecture, ask questions, take notes, do homeworks, take tests, just don't pay. I teach mathematics at the college level... if a students showed up in my classroom who seriously wanted to learn, but didn't want to pay tuition, I would be more than supportive of his/her presence in my class. A number of my colleagues feel the same way -- learning should transcend economic boundaries. (On the other hand, though, some of my peers in our University's Physics department like the fact that tuition weeds out the middle-aged crackpots with their pseudo-scientific TOEs). For math books freely downloadable online, dig around at http://www.math.umn.edu/~garrett
Ooh, ooh. I got another one:
... and I'm spent.
Hmm... Maybe I'll watch some FreeBSD tonight.
IBM says: "The world's first dedicated Linux mainframe server!"
I say: "The world's first dedicated Linux server requiring a 30-year mortgage."
I think the point of the article was not that software shouldn't have this sort of functionality, but rather that software should not have such non-security-conscience enabled by default, allowing Users or Administrators to choose whether to turn the functionality on if they deem the benefits outweigh the costs.
Take OpenBSD as an example. The good folks at OpenBSD review their code, flush out the bugs, and turn off non-essential stuff in their default install. BUT, they also document everything well and make configuration easy enough so that if you want to turn on something risky you can. Also, they have great linux emulation and an excellent port system so that you can install tons of software, both secure and insecure. They leave it up to the admin to decide between functionality and security, but the overwhelming default is security over functionality.
I think this is essentially what Schneier envisions.
Unlike previous changes in direction, Microsoft cannot blitzkrieg itself into a Secure OS. As the Security Community states again and again: computer security takes time. It takes line-by-line code auditing, careful design decisions, and years of testing and analysis.
If Microsoft is serious about it, securing Windows is definitely a Good Thing (tm) for Microsoft, *nix, and the computing community at large.
First, as Schneier et al. pointed out, good security in a Windows OS will take a complete redesign from the ground up. There's NO WAY they can hack on patches to Windows and claim that they're serious about operating system security. Furthermore, Microsoft will NEVER let the security community examine "trade secrets" from an already released copy of Windows (if only because they'd have too many fires to put out after bugs were found). And if Microsoft is planning to doall that Schneier et al. suggest, they won't be able to neatly build on existing NT/XP code. Thus, redesign seems like their only option.
Why is a complete design of Windows a good thing for Linux/BSD?
Of course, all of this is pure speculation. Still, no matter what Microsoft does, at worst we'll be unaffected by Microsoft's new move toward a secure OS. But, at best, I believe the Open Source community could gain a lot.
Although this conversation is only a few short steps from namecalling, I'd like to point out that fundamentally, the issues being discussed here are tremendously important to scientists, ethicists, and human-kind in general.
At what point does a living organism become autonomous? At what point can we consider an human cell, a full-fledged human? Is it ethical to create a human-based organism just to destroy it?
These ethical considerations were discussed in last month's Scientific American.
Well said.
Mod this comment up.
To this Edsger W. Dijkstra rebutts with this:
and this:
Our CS department has converted to Java as an introductory language to teach undergrads. We've found that most students find the language pleasant and fairly easy to learn (as opposed to Scheme, a language which most pupils and even some TA's dread... it made for a good "weed-out" course but not much else). Java provides a solid OO foundation for further software development, and enough C-like syntax so that if a student needs to learn C or C++, they can do so quickly. And though algorithms and data stuctures can be taught in almost any language, Java's clear and consistent syntax, and automatic memory management makes teaching the subject a lot easier than when I used C++ or C (ugh).
Still, from a more pragmatic perspective, Java sucks. We can't very well teach an OS course using Java (C is still preferred). When doing any major software engineering projects, Java is too slow to make it a very viable choice. And teaching Networking with Java is just too easy to be of any value to students (I use C again).
So Java is not universally used in all areas of CS instruction (at least in my classes). But, I find that more students learn more CS when using Java as an introductory language. And when it's time to shift to "real world" programming, it's not too difficult for them.
That being said, I would have to say that the switch from C++ to Java as an AP Computer Science Language is definitely a good move. And using Java in the college classroom is also a good thing for CS instruction (though less definitely so). However, I think its important for CS instuctors to force their students to learn "real-world" programming languages such as C, C++, and Perl. The more languages a student learns, the better they will adapt to future trends in industry.
It's ironical. Like rain on a rainy day. Like a free Unix for which one pays. It's the good advice that Windows just didn't take. Who'd 've thought it figured.
These Buffer Overflows were local holes (i.e. you had to have to have a foothold on the box to take advantage of it).
For the past four years, OpenBSD has had no remote holes in the default install.
What Alex says is right on the nose (i.e. mod that comment up!). I too switched from Linux to OpenBSD at home almost 3 years ago, and have been using it for various projects at work for the past 2 years. This is what I've discovered:
In conclusion, I'd say trying OpenBSD is something every geek should do. But, admittedly, my loyalty to the OS (Theo) is waning, and I'm beginning to think FreeBSD might be a better choice.
Whatever the case, I'm sure Theo doesn't give a good god-damn...
Happy hacking,
The 'roid
Hang on, so Theo et al. in all their line by line code auditing never noticed that IPF didn't fall under the BSD license??! How many times has OpenBSD code been audited?!? How long has IPF been a part of OpenBSD??!
In the upcoming volume IV of Knuth's TAOCP, Seminumerical Searches for Hoaring Triplets, he conjectures the following about the Theory of Programming Languages:
Along these lines, I pose the following questions:
Thank you.
I don't think anyone is arguing Linux's popularity among Unixens (Unixers? Eunuchs?). The point is whether it is some sort of de facto standard like Apache has become in webservers, TCP/IP in networking protocols, and Perl in scripting languages (yes, that was a red herring).
If you ask me (which you didn't, but you should have!) I really don't think that this is the case. As much as Linux has contributed in technology to other Unix variants, it has stolen. HP-Unix, Tru64 (OSF/1), AIX, *BSD, and Solaris are still THRIVING in both corporate and academic circles. Indeed, the systems programming classes taught in our Computer Science department are Solaris-biased, not Linux. The Unix of choice for work in our Physics department is NOT Linux, but Tru64.
Yes, I agree that Linux is popular. And it is generally becoming true that where you find Unix, you will find some Linux. But if Linux is a de facto standard, then it is only the case among slashdotters. It is no more the de facto standard of Unix, than RedHat is the de facto standard of Linux distributions.
Just for shits and giggles here's the 10 commandments run through Babelfish a couple times (English ->Span ->French->German->English).
:)
Now imagine what 4000 years of the same thing has done to the original version of The Bible...
Emigration 20:1-17
spake king James version
1 and God all this word, say
2 I gentleman thy gott to have that thee track Egypt, house bondage. shalt
3 Thou outside to bring every other God not to have, before does not shalt
4 Thou to arrive around thee each load picture or each similarity to make each thing at skies or at earth under or at water under earth: from
5 Thou serves shalt neither arqueamien to thyself downward at them nor them: for Mr. Deity of I thy jealous God, which visits iniquity from the family members on the children, the third and fourth generation of them the hate i; to show
6 and at the thousands compassion of them that the i and maintenance my
7 entries shalt auftraege of Thou, fallen in love, not the name of Mr. Deity thy in unnecessary; for GENTLEMAN the taketh at innocent at it will not bring its name in unnecessary up to date.
8 remembers on on day sabbath to bring over it santo up to date the work thou shalt from
9 six days and from the characters the whole thy work:
10, but the sieved day thy is sabbath from Mr. Deity: in it shalt thou, in order to execute each work, thou, neither maidservant manservant threads thy, neither thy foreign and neither thy and nor thy thy and won thy and neither the girl, who is in the thy tueren: in six days the GENTLEMAN made
11 close the sky and the earth, the sea and everything in them is, and the sevened day rested: the wherefore GENTLEMAN approved the day sabbath, and he sanctified more thy father from 12 honours and from more thy mutter: day time does not out thy long power on track the thee thy giveth Mr. Deity tearing off shalt
13 Thou number shalt
14 Thou, in order adulterous confidence shalt
15 Thou not to make, around a false witness bear shalt
16 Thou not against neighbour thy. house
17 Thou shalt nonneighboring ovetthy to steal wife neighbour thy covet shalt thou not, neither its manservant neither its, neither its cattle, neither its donkey, nor each thing maidservant neighbouring thy.
According to the article, this technology's intimate relationship with Internet2 comes mostly from the fact that there were very few applications around which NEEDED Internet2's impressive network stats to actually run. Consequently, the peeps at I2 contacted Jaron to lead up the project. And ... well, you can read about it here.
Additionally there's some other teleimmersion sites at UNC and at Jaron's research site.
Actually, rumor amongst the underground is that someone (possibly a non-australian government agency) fucked up a Carnivore-esque program running on the cable so that it doesn't pass IP packets anymore.
Purely rumor, but still interesting.
This is what I don't understand: if our goal is to expand the amount of domain names available, why are businesses allowed to register every TLD associated with anything they have a trademark for? Allowing companies to do things like: microsoft.com, microsoft.biz, microsoft.info, microsoft.name, microsoft.coop, etc. effectively defeats the purpose of having multiple TLDs.
I think something needs to be done by ICANN to prevent this, Dammit. Otherwise a domain name crunch is really not helped by adding more TLDs.