That says it all. Gentoo is not for the faint of heart. It's not widely supported by people developing third party software. Hardware support is not automatic. If you want something that has more of that "just works" feel you need to run fedora or suse.
That's not to say there's anything wrong with Gentoo. I've been using it for years. I only run Linux and it's been that way for 10 years now. I've actually had no problem in recent memory with new hardware or getting software to work. Of course I know exactly what to do and where to go.
Put me in front of Windows, and I'm hopelessly lost. I had the most horrible time getting my parents computer to hook up to a webdav server. SP2 this, edit registry key that, it's all confusing to all hell.
You're a windows user. You know windows well. You know exactly what to do when something goes wrong in windows. That's why it seems easier. If you spend as much time working through problems in Linux as you have in the past working through problems in Windows, you'd find they're pretty much equally usuable (or lacking).
Until IBM ports Lotus Notes to Linux and starts using it, anything they say about the Linux desktop should be absolutely ignored.
It's called iNotes. It is officially supported under Linux too. Check out the spec page. Retargeting large pieces of software is not something that happens over night. This is the direction that Lotus is supposedly heading though.
There are internal deployments of iNotes too. However, notes under wine works so well for me that I doubt I'd switch anytime soon...
Now, if you know an actual reason to be sceptical of the worldwide scientific consensus
Precisely because it's scientific consensus, and not science. See Michael Crichton's lecture on this.
I had a similiar discussion recently with some friends. I was talking about nuclear winter and how there's not only no scientific evidence to support such a scenario, it's actually highly unlikely. One of my friends responded that that was ok, it was better for people to fear nuclear winter regardless of it's scientific basis.
Yes, pollution is bad. Let's not invent a world-ending scenario though to try and make people understand that. There are plenty of reasons to significantly reduce pollution.
I think people don't understand that championing global warming only hurts the cause of the environmentalists by taking the focus away from the proven effects of pollution.
Xen is an open source hypervisor for intel hardware. A hypervisor allows multiple operating systems to run side-by-side simultanously. Don't think VMware, think partitioning on a mainframe.
Intel's VT technology is hardware support for partitioning. Google it.
sHype is a research hypervisor at IBM that implements advanced security mechanisms much in the same way that SELinux does.
So, think mainframe style partitioning with the security of SELinux.
1) All distros clearly say that their disro ver X is LSB ver Y compliant and stand behind that.
The major distros (SuSE and RedHat) do this.
2) LSB mandates a sufficiently detailed configuration and fileset that a developer can build an app under any LSB ver Y.Z and expect it to install and run (with no missing libaries, re-configuration, config file editing etc) on any other LSB Y.Z compliant disro installation.
Have you read the LSB? This is what it does.
3) Oracle ver nn runs under LSB ver Y.Z NOT ONLY RH AS3.x and Suse EL 9.x (or whatever).
You're never going to see this. Why? Because of testing. Since there's no "LSB" distro to test on, Oracle can only test on things like AS3.x and SuSE EL 9.x and therefore will only certify that it works on what it's tested on.
However, if it works on one LSB compliant distro, it should work on another. Keep in mind though, not everything is covered in the LSB. Otherwise, there'd be no point to having a distro, we'd all just use the LSB.
4) There's an automated validation that can determine if an initial distro install is (or is still) valid LSB ver Y.Z configuration.
There is. My office mate over the summer was an intern hired to run this test suite against various distros. I think you have to be an LSB member or something to get to it though.
You also have things like string theory (which, admittedly, is still theory, but it is a theoretical advancement),
See, this is the problem with post WWII physics. The major advances are all either pure theory or just comparatively small advances on existing theories. All sciences progress but the major breakthroughs in modern physics occurred early in the 20th century. That's my point.
I understand what you're saying about understanding the mathematical basics. My point though is that the math hasn't really gotten harder but now the tediousness of it has been reduced. Erwin Schrodinger did not have a computer.
Sorry, but theoretical computer science is only closer to math because...well, it is math (a branch of it, at least).
Yup, and theoritically physics is somewhat closer to philosophy. Theoritically physics takes observations and tries to build theories on top of it.
I disagree about the particle physics example too. There's two different types of math, high-level maths like the various forms of calculus and then very low-level maths like set theory and predicate calculus (various forms of logic). Computer science deals with lower level maths.
I think you bawk at Computer Science's lack of mathematics simply because you're not aware of what the major of Computer Science really is. Not that you don't have a right to, most people with degrees in Computer Science have not really been exposed to it.
We can probably agree on one thing, CS is more popular because it's somewhat more immediately practical. Therefore there's a ton more people in it than in physics. However, if you compared the number of physists with an equal number of computer scientists (that were the top in their field) you'd find that both fields were equally intensive.
It'll all even out though once outsourcing pushes out the code monkeys to third world countries:-)
November 1998: IBM kills study on software patents by the Whitehouse
Your link is just about as shady as it gets. Look, it doesn't matter what patents IBM files. Has IBM (in recent times) used it's patent portfolio to squash competition or to do generally evil things? No.
IBM has invented a good chunk of the technology out there today. The article mentions pursuing patent claims against Oracle, well, IBM invented the relational database! A lot of companies exist and make lots of money based on technologies that IBM invented but did not pursue.
What group did you work for? The IEBs are organized at the divisional level. In Systems group and Software group it's the only way to file a patent. Since those two groups make up a large majority of all patents filed, I don't know if other organizations have IEBs in place.
It's a full time calling and you can't just do it while working as a patent clerk anymore.
Yeah because physics has advanced so much since the introduction of relativity and quantum mechanics. Oh, wait a minute...
Physics is just as hard if not easier. Now there are very sophisticated computers that take a large portion of the brunt mathematics out of physics.
As for your comments about the "lesser" physics, I say bah to that. Physics is just a cuddly version of math after all. Theoritical computer science is closer to pure math than physics is I'd wager.
What you don't hear about is that fact that to file a patent, one must present the idea in front of a committee (called an Invention Evaluation Board) which does an initial search on the idea and evaluation of business value, then the patent lawyers do their own internal search (before sending to the patent office), then you write up the patent with a lawyer (all the time modifying to take into account any existing work), then IBM sends the patent to the patent office so the patent office can do it's search.
By the time IBM sends out a patent, it's already gone through an exhaustive evaluation by very intelligent people. Patents cost a lot of money to file. IBM has no interest in filing useless patents. And yes, there is a culture that if an idea seems at all novel then file a disclosure because we have such a strong process in place to determine if that idea should become a patent.
And is IBM using it's portfolio to do negatively? Nope. Patents are a necessary evil. Any large company has to file patents to protect itself. Being that IBM is the largest technology company in existance (320,000 employees, revenue of $86 billion a year), it's only fitting that it files the most patents.
My elementary school had a computer lab. A very rare thing (this is circa 1988). The had a small elective class thing for second graders that taught logo. You learned the basic commands then had some time to make your own drawing.
Most students wrote simple circles and squares that took about 3 lines of logo. I drew a house with trees and stuff. It was a few pages of logo. It was enough that the teacher called my parents and told them I should go to a special school to learn programming.
My parents said no. They thought it was a little too weird. However, my parents got a computer (an IBM PS/1) and within the next two years my Uncle while visiting showed me how to use Basica on it.
That was it. I don't know why but it sparked an interest. I went out and continously checked out the two books on programming the local library had (one on Basic and one on C). I read them cover to cover and saved up 500 to buy my first used laptop around 1993. It didn't come with an operating system so I put this "hacker operating system" called Linux on it. Took me a couple years to figure out how to get X to work but I was able to use gcc which was all I cared about.
So, at the end of the day, I think I would have gotten into programming no matter what. It may have been later than I did but I do believe it still would have happened.
My advice? Don't try to introduce your children into computers. Expose them to everything, see what they take to, and nuture it. I know most people want their children to be successful, but I also think people are most successful when they're doing what they love to do.
One bug in two weeks? Come on. That's not impossible by a long shot.
I want to expand on this a bit. My math was based on the most conservative assumption possible. That an 'A' average meant a 90 average for the other 40% of the grade. The most liberal assumption is 100 for the other 40% which would then reduce the passing number to just 4 bugs. That's 3 weeks per bug.
Plus, 300 hours of work over 12 weeks is just a mere 3 hours per week. We have courses at my University that require 20 hours per week (plus class time). That's 2,600 hours of work. I'm telling you, this dude is full of BS. Plus, I'm sure if everyone is doing as bad as he says, the prof either hates these kids (cause they all didn't do the required work) or he's gonna curve it like mad.
I've had professors who at the end of the semester tell us that all the tests were targetted to an average grade of 50% and that the class would be curved 40% As 50% Bs and 10% other. The purpose of making impossibly hard classes is to let star students shine.
Don't underestimate the capabilities of a college student. They're much more capable than you think. A professor's job is to have them realize that capability. The originally poster of this story is a tool. I'm most certainly convinced.
I know you're just trying to look out for the kid, but I think you're actually underestimating his ability.
You're a dick. "You had a semester." Nice attitude. They also had several other classes, undoubtedly demanding classes.
You make way too many assumptions. I am a student. I take a full course load in an honors CS-program at a top-10 university plus I work for IBM's Linux Technology Center.
I work on with Open Source software while going to school. Yes, it's hard, I don't doubt that. No, I don't expect everyone to have the kind of background I do, but I know, that it is possible.
I've seen, first hand, what most students are like in courses that are considered hard. The guy spewing all this president of ACM crap just makes me more upset.
We have classes at my school where you have to design an operating system from scratch, write a pascal compiler from scratch, reverse engineer large portions of the pentium 4 architecture and implement in VLSI (with no non-public information).
There are hard classes at Universities. They aren't required classes. If it was completely impossible it'd be one thing but they aren't. 10 bugs over a semester is pretty hard. But I looked at the syllabus and to fail the class with an A test average, you would have found less than 6 bugs. So, you don't fail if you find one bug every two weeks (semesters are twelve weeks).
One bug in two weeks? Come on. That's not impossible by a long shot.
then you find 10 security holes in released open-source software.
Most common security holes are because of off by one errors or because of printf/scanf vunerabilities. You run a piece of open source software in valgrind and you're likely to find an off by one.
I cannot undertake this challenge because 1) I work for IBM and am restricted in what I can publish about Open Source projects 2) it would take me a week of my free time to find these 10 holes. Why in the world would I waste my limited free time doing that?
Look, I understand the money-where-your-mouth is argument but you cannot expect me to spend that sort of time to disprove some kid who can't make a decent grade and decides to bitch about it on slashdot.
I'm paid to do it. In a corporation, testers are expected to find at least 2-3 bugs a week. You had a semester.
Did you try using static analysis tools? Or a tool like valgrind? There are methodologies to doing this sort of stuff. If you just picked some big project and started poking through C code, you have no chance.
The course lists fluency in C as a requirement and states that each student has to find 10 new bugs in Open Source software.
That's not hard. That would take about a days work for any proficient C hacker. If you attack a large OS program you're screwed. However, looking over his lectures, he suggests both group work and searching for projects through sourceforge.
Hell, half of the projects on sourceforge don't even compile. There's ton of opportuntities to find holes...
Unfortunately businesses are too preoccupied by a meaningless certificate of certification than can do spirited and capable scientists.
Businesses looking to hire IT staff may be but if you go to a company that specializes in solving software problems you will be turned down just for having certificates on your resume.
Just remember, there are jobs for IT specialists and jobs for Computer Scientists. They aren't the same thing. There are far more jobs in IT than there are in CS but there are still far too few good Computer Scientists for the number of jobs available.
I am frequently surprised that so many people consider themselves to be an X-Language programmer (for some particular X-Language). I think of myself as a computer scientist
I mostly agree with you with one caveat, I do think it's appropriate to label yourself as an X-Language programmer is you have extremely deep skills in a particular language. Skills that takes years and years to develop.
Most computer scientists can solve just about any problem in a typical programming language (if given enough time to learn it). The solution probably won't be optimal and most likely won't be elegant (as judged by an expert in the language).
Many people call themselves C, C++, or Java programmers. However, unless you can discuss the differences between char's range on different platforms, you're not a C programmer. If you cannot talk about the virtues of virtual inheritence or multi-paradigm programming, you're not a C++ programmer. If you cannot talk about the nuances of using J2EE you're not a Java programmer. Obviously, I'm not a Java programmer or I would have come up with a better example.
Java's a good example. I would never consider myself a Java programmer, but I am confident I know more about Java than most of my peers thatdo consider themselves as Java programmers.
However, you really can't blame most people. I don't think people realize how complicated most modern languages really are and how difficult it is to write 100% bullet-proof code.
Coverity is a start-up formed from the group at Stanford who developed the Stanford Checker. The Stanford Checker is a static analysis tool (quite primative actually as far as the state-of-the-art is concerned) who's main claim to fame was the number of bugs it found in the Linux kernel.
What they're probably claiming is that their tool found less bugs in Linux than in most commericial software. I buy this. The tool finds a lot of bugs that aren't necessarily easy to exploit or things that are really just not best practices.
Open source software tends to get highly critiqued (at least, the popular projects). It often becomes a programmer pissing-contest to see who can come up with problems in someone elses code. Commericial software, on the other hand, tends to just be whatever gets the job done.
This doesn't answer the most important question though: Which method is more succeptiable to attack? The question is very hard to answer.
Google's all about HTML/JS compression. I remember reading an article about how Google goes to great lengths to reduce their HTML/JS fingerprint. It ends up resulting in real savings.
Is it why in most disadvantaged African societies, where young boys do not wear pants, one finds them very very sexualy potent well into their adult life.
I'd wager it's a few other factors. As someone else pointed out, contraceptives are used far less often than in the Western world. In fact, this is often attributed as one of the leading factors in the spread of AIDS. I had read an article in Scientific American that stated that something on the order of 66% of women had, at some point, exchanged money for sex in Zaire and that in these sorts of exchanges, men refuse to use any sort of birth control.
As for the age question, I'd also wager that the individuals that make it to such an old age are exceptionally healthy individuals. Since the average life expectancy is so much lower, those that live to an older age naturally represent stronger individuals.
As for the larger gentalia comment, I think you'll find that size has most to do with the size of the females hips. Men and women have evolved together such that everything fits just right. The rising rate of C-sections tends to have to do with inter-ethnic offspring. For instance, an african man marrying a chinese woman would likely result in complications as the babies head might be much larger than the woman's hips are designed to accomodate.
Why evolve to have larger bodies? I'm not sure. My first guess is it has something to do with how quickly a baby is expected to fend for themselves. A larger baby, I'm guessing, would more quickly be big enough to defend itself.
I'm sure there's been some paper written about the subject...
Any pointer toward good discussion about why existing LDAP daemon can't be used for AD interop
We discussed this mostly in person or on #samba-technical. Basically, there are a number of things about AD that won't ever be in OpenLDAP because it would comprimise OpenLDAP's standards compliance.
A very well-known example (that's quite personal for me) is connectionless-LDAP support. Active Directory supports a form of LDAP over UDP. There's a conflicting standard that's not really in use for this type of transport. The code's in OpenLDAP to support it but Kurt won't let it be easily enabled for fear that it will hamper future standization efforts.
Samba and OpenLDAP have an amicable relationship. It's just that we have different goals at the end of the day.
And as for reinventing the wheel, it's not really. Samba4 has a generic backend and an ASN1 library. Adding LDAP server-support doesn't require that much. There are some difficult parts but basic functionality isn't so bad.
Any idea what the cost is? This might be a workable solution, but I'd like to reduce the complexity as much as possible.
I'm not a sales guy but I'd reckon it'd be overkill. Here's what you should do, reproduce the problem with the latest version of Samba compiled from source. Take a network trace of the Windows BSOD. Post something to bugzilla that links to the network capture and the appropriate Samba logs.
It'll get fixed. The LDAP integration in Samba is a rather important feature and there's quite a few people on the team that tackle those sort of problems. Once it's in bugzilla, it's just a matter of time until it's fixed.
If you don't go through the effort of reproducing it with the latest version of Samba though, people will be much less willing to help out. The more leg work you do the quicker it'll get fixed.
Heck, if you've got a network trace of Samba 3.0.9 sending a bad packet to a Windows client, it'll get fixed really quickly. That's the sort of thing that catches people's eyes...
BTW, I should say that even though the solution we were trying to implement was not supported, the Redhat guy was nice enough to try and reproduce our setup and help us
We work closely with the RedHat guys. I'm surprised they didn't file a bug. Usually when they can't make progress on a bug they call us up.
people forget that samba is actually now about TWENTY FIVE separate protocols / APIs, about five of which are implemented in one program (nmbd), about TWENTY of which are implemented or used in smbd.
You'll appreciate this one. A couple years ago we were looking into better Active Directory interop. It was discovered that to do a real AD-style logon (the same way that XP and most Windows clients do it) required an RPC that was only available over TCP/IP (Windows refused to send it over SMB).
This led to a major hack job of Samba3 to support an end-point mapper and to support RPC over TCP. We eventually gave up and decided that AD would have to wait for Samba4.
In all fairness, once you get used to the hand-coded RPC style, it's not that bad. It's a decent subsystem as far as Samba3 goes. However, as you said, the learning curve is steep. I'm sure I still don't understand completely when to pad and when not to pad.
Slashdot Mirror
I run a Gentoo box
That says it all. Gentoo is not for the faint of heart. It's not widely supported by people developing third party software. Hardware support is not automatic. If you want something that has more of that "just works" feel you need to run fedora or suse.
That's not to say there's anything wrong with Gentoo. I've been using it for years. I only run Linux and it's been that way for 10 years now. I've actually had no problem in recent memory with new hardware or getting software to work. Of course I know exactly what to do and where to go.
Put me in front of Windows, and I'm hopelessly lost. I had the most horrible time getting my parents computer to hook up to a webdav server. SP2 this, edit registry key that, it's all confusing to all hell.
You're a windows user. You know windows well. You know exactly what to do when something goes wrong in windows. That's why it seems easier. If you spend as much time working through problems in Linux as you have in the past working through problems in Windows, you'd find they're pretty much equally usuable (or lacking).
Until IBM ports Lotus Notes to Linux and starts using it, anything they say about the Linux desktop should be absolutely ignored.
It's called iNotes. It is officially supported under Linux too. Check out the spec page. Retargeting large pieces of software is not something that happens over night. This is the direction that Lotus is supposedly heading though.
There are internal deployments of iNotes too. However, notes under wine works so well for me that I doubt I'd switch anytime soon...
Now, if you know an actual reason to be sceptical of the worldwide scientific consensus
Precisely because it's scientific consensus, and not science. See Michael Crichton's lecture on this.
I had a similiar discussion recently with some friends. I was talking about nuclear winter and how there's not only no scientific evidence to support such a scenario, it's actually highly unlikely. One of my friends responded that that was ok, it was better for people to fear nuclear winter regardless of it's scientific basis.
Yes, pollution is bad. Let's not invent a world-ending scenario though to try and make people understand that. There are plenty of reasons to significantly reduce pollution.
I think people don't understand that championing global warming only hurts the cause of the environmentalists by taking the focus away from the proven effects of pollution.
The article states that since 1790 the earth's global temperature has risen 0.8 degrees.
In almost a quarter of a millenia this is the change. Excuse me if I remain skeptical that this was caused by human activity.
Xen is an open source hypervisor for intel hardware. A hypervisor allows multiple operating systems to run side-by-side simultanously. Don't think VMware, think partitioning on a mainframe.
Intel's VT technology is hardware support for partitioning. Google it.
sHype is a research hypervisor at IBM that implements advanced security mechanisms much in the same way that SELinux does.
So, think mainframe style partitioning with the security of SELinux.
1) All distros clearly say that their disro ver X is LSB ver Y compliant and stand behind that.
The major distros (SuSE and RedHat) do this.
2) LSB mandates a sufficiently detailed configuration and fileset that a developer can build an app under any LSB ver Y.Z and expect it to install and run (with no missing libaries, re-configuration, config file editing etc) on any other LSB Y.Z compliant disro installation.
Have you read the LSB? This is what it does.
3) Oracle ver nn runs under LSB ver Y.Z NOT ONLY RH AS3.x and Suse EL 9.x (or whatever).
You're never going to see this. Why? Because of testing. Since there's no "LSB" distro to test on, Oracle can only test on things like AS3.x and SuSE EL 9.x and therefore will only certify that it works on what it's tested on.
However, if it works on one LSB compliant distro, it should work on another. Keep in mind though, not everything is covered in the LSB. Otherwise, there'd be no point to having a distro, we'd all just use the LSB.
4) There's an automated validation that can determine if an initial distro install is (or is still) valid LSB ver Y.Z configuration.
There is. My office mate over the summer was an intern hired to run this test suite against various distros. I think you have to be an LSB member or something to get to it though.
You also have things like string theory (which, admittedly, is still theory, but it is a theoretical advancement),
:-)
See, this is the problem with post WWII physics. The major advances are all either pure theory or just comparatively small advances on existing theories. All sciences progress but the major breakthroughs in modern physics occurred early in the 20th century. That's my point.
I understand what you're saying about understanding the mathematical basics. My point though is that the math hasn't really gotten harder but now the tediousness of it has been reduced. Erwin Schrodinger did not have a computer.
Sorry, but theoretical computer science is only closer to math because...well, it is math (a branch of it, at least).
Yup, and theoritically physics is somewhat closer to philosophy. Theoritically physics takes observations and tries to build theories on top of it.
I disagree about the particle physics example too. There's two different types of math, high-level maths like the various forms of calculus and then very low-level maths like set theory and predicate calculus (various forms of logic). Computer science deals with lower level maths.
I think you bawk at Computer Science's lack of mathematics simply because you're not aware of what the major of Computer Science really is. Not that you don't have a right to, most people with degrees in Computer Science have not really been exposed to it.
We can probably agree on one thing, CS is more popular because it's somewhat more immediately practical. Therefore there's a ton more people in it than in physics. However, if you compared the number of physists with an equal number of computer scientists (that were the top in their field) you'd find that both fields were equally intensive.
It'll all even out though once outsourcing pushes out the code monkeys to third world countries
November 1998: IBM kills study on software patents by the Whitehouse
Your link is just about as shady as it gets. Look, it doesn't matter what patents IBM files. Has IBM (in recent times) used it's patent portfolio to squash competition or to do generally evil things? No.
IBM has invented a good chunk of the technology out there today. The article mentions pursuing patent claims against Oracle, well, IBM invented the relational database! A lot of companies exist and make lots of money based on technologies that IBM invented but did not pursue.
What group did you work for? The IEBs are organized at the divisional level. In Systems group and Software group it's the only way to file a patent. Since those two groups make up a large majority of all patents filed, I don't know if other organizations have IEBs in place.
It's a full time calling and you can't just do it while working as a patent clerk anymore.
Yeah because physics has advanced so much since the introduction of relativity and quantum mechanics. Oh, wait a minute...
Physics is just as hard if not easier. Now there are very sophisticated computers that take a large portion of the brunt mathematics out of physics.
As for your comments about the "lesser" physics, I say bah to that. Physics is just a cuddly version of math after all. Theoritical computer science is closer to pure math than physics is I'd wager.
What you don't hear about is that fact that to file a patent, one must present the idea in front of a committee (called an Invention Evaluation Board) which does an initial search on the idea and evaluation of business value, then the patent lawyers do their own internal search (before sending to the patent office), then you write up the patent with a lawyer (all the time modifying to take into account any existing work), then IBM sends the patent to the patent office so the patent office can do it's search.
By the time IBM sends out a patent, it's already gone through an exhaustive evaluation by very intelligent people. Patents cost a lot of money to file. IBM has no interest in filing useless patents. And yes, there is a culture that if an idea seems at all novel then file a disclosure because we have such a strong process in place to determine if that idea should become a patent.
And is IBM using it's portfolio to do negatively? Nope. Patents are a necessary evil. Any large company has to file patents to protect itself. Being that IBM is the largest technology company in existance (320,000 employees, revenue of $86 billion a year), it's only fitting that it files the most patents.
My elementary school had a computer lab. A very rare thing (this is circa 1988). The had a small elective class thing for second graders that taught logo. You learned the basic commands then had some time to make your own drawing.
Most students wrote simple circles and squares that took about 3 lines of logo. I drew a house with trees and stuff. It was a few pages of logo. It was enough that the teacher called my parents and told them I should go to a special school to learn programming.
My parents said no. They thought it was a little too weird. However, my parents got a computer (an IBM PS/1) and within the next two years my Uncle while visiting showed me how to use Basica on it.
That was it. I don't know why but it sparked an interest. I went out and continously checked out the two books on programming the local library had (one on Basic and one on C). I read them cover to cover and saved up 500 to buy my first used laptop around 1993. It didn't come with an operating system so I put this "hacker operating system" called Linux on it. Took me a couple years to figure out how to get X to work but I was able to use gcc which was all I cared about.
So, at the end of the day, I think I would have gotten into programming no matter what. It may have been later than I did but I do believe it still would have happened.
My advice? Don't try to introduce your children into computers. Expose them to everything, see what they take to, and nuture it. I know most people want their children to be successful, but I also think people are most successful when they're doing what they love to do.
Just my thoughts.
It's a moot point now. They all got B's. Found it on one of the kids blogs. The class was curved.
This was just some stupid kid whining cause he doesn't understand how college works.
One bug in two weeks? Come on. That's not impossible by a long shot.
I want to expand on this a bit. My math was based on the most conservative assumption possible. That an 'A' average meant a 90 average for the other 40% of the grade. The most liberal assumption is 100 for the other 40% which would then reduce the passing number to just 4 bugs. That's 3 weeks per bug.
Plus, 300 hours of work over 12 weeks is just a mere 3 hours per week. We have courses at my University that require 20 hours per week (plus class time). That's 2,600 hours of work. I'm telling you, this dude is full of BS. Plus, I'm sure if everyone is doing as bad as he says, the prof either hates these kids (cause they all didn't do the required work) or he's gonna curve it like mad.
I've had professors who at the end of the semester tell us that all the tests were targetted to an average grade of 50% and that the class would be curved 40% As 50% Bs and 10% other. The purpose of making impossibly hard classes is to let star students shine.
Don't underestimate the capabilities of a college student. They're much more capable than you think. A professor's job is to have them realize that capability. The originally poster of this story is a tool. I'm most certainly convinced.
I know you're just trying to look out for the kid, but I think you're actually underestimating his ability.
You're a dick. "You had a semester." Nice attitude. They also had several other classes, undoubtedly demanding classes.
You make way too many assumptions. I am a student. I take a full course load in an honors CS-program at a top-10 university plus I work for IBM's Linux Technology Center.
I work on with Open Source software while going to school. Yes, it's hard, I don't doubt that. No, I don't expect everyone to have the kind of background I do, but I know, that it is possible.
I've seen, first hand, what most students are like in courses that are considered hard. The guy spewing all this president of ACM crap just makes me more upset.
We have classes at my school where you have to design an operating system from scratch, write a pascal compiler from scratch, reverse engineer large portions of the pentium 4 architecture and implement in VLSI (with no non-public information).
There are hard classes at Universities. They aren't required classes. If it was completely impossible it'd be one thing but they aren't. 10 bugs over a semester is pretty hard. But I looked at the syllabus and to fail the class with an A test average, you would have found less than 6 bugs. So, you don't fail if you find one bug every two weeks (semesters are twelve weeks).
One bug in two weeks? Come on. That's not impossible by a long shot.
then you find 10 security holes in released open-source software.
Most common security holes are because of off by one errors or because of printf/scanf vunerabilities. You run a piece of open source software in valgrind and you're likely to find an off by one.
I cannot undertake this challenge because 1) I work for IBM and am restricted in what I can publish about Open Source projects 2) it would take me a week of my free time to find these 10 holes. Why in the world would I waste my limited free time doing that?
Look, I understand the money-where-your-mouth is argument but you cannot expect me to spend that sort of time to disprove some kid who can't make a decent grade and decides to bitch about it on slashdot.
Really? Then you do it.
I'm paid to do it. In a corporation, testers are expected to find at least 2-3 bugs a week. You had a semester.
Did you try using static analysis tools? Or a tool like valgrind? There are methodologies to doing this sort of stuff. If you just picked some big project and started poking through C code, you have no chance.
The course lists fluency in C as a requirement and states that each student has to find 10 new bugs in Open Source software.
That's not hard. That would take about a days work for any proficient C hacker. If you attack a large OS program you're screwed. However, looking over his lectures, he suggests both group work and searching for projects through sourceforge.
Hell, half of the projects on sourceforge don't even compile. There's ton of opportuntities to find holes...
Unfortunately businesses are too preoccupied by a meaningless certificate of certification than can do spirited and capable scientists.
Businesses looking to hire IT staff may be but if you go to a company that specializes in solving software problems you will be turned down just for having certificates on your resume.
Just remember, there are jobs for IT specialists and jobs for Computer Scientists. They aren't the same thing. There are far more jobs in IT than there are in CS but there are still far too few good Computer Scientists for the number of jobs available.
I am frequently surprised that so many people consider themselves to be an X-Language programmer (for some particular X-Language). I think of myself as a computer scientist
I mostly agree with you with one caveat, I do think it's appropriate to label yourself as an X-Language programmer is you have extremely deep skills in a particular language. Skills that takes years and years to develop.
Most computer scientists can solve just about any problem in a typical programming language (if given enough time to learn it). The solution probably won't be optimal and most likely won't be elegant (as judged by an expert in the language).
Many people call themselves C, C++, or Java programmers. However, unless you can discuss the differences between char's range on different platforms, you're not a C programmer. If you cannot talk about the virtues of virtual inheritence or multi-paradigm programming, you're not a C++ programmer. If you cannot talk about the nuances of using J2EE you're not a Java programmer. Obviously, I'm not a Java programmer or I would have come up with a better example.
Java's a good example. I would never consider myself a Java programmer, but I am confident I know more about Java than most of my peers thatdo consider themselves as Java programmers.
However, you really can't blame most people. I don't think people realize how complicated most modern languages really are and how difficult it is to write 100% bullet-proof code.
Coverity is a start-up formed from the group at Stanford who developed the Stanford Checker. The Stanford Checker is a static analysis tool (quite primative actually as far as the state-of-the-art is concerned) who's main claim to fame was the number of bugs it found in the Linux kernel.
What they're probably claiming is that their tool found less bugs in Linux than in most commericial software. I buy this. The tool finds a lot of bugs that aren't necessarily easy to exploit or things that are really just not best practices.
Open source software tends to get highly critiqued (at least, the popular projects). It often becomes a programmer pissing-contest to see who can come up with problems in someone elses code. Commericial software, on the other hand, tends to just be whatever gets the job done.
This doesn't answer the most important question though: Which method is more succeptiable to attack? The question is very hard to answer.
I'm surprised noone's caught on to this.
Google's all about HTML/JS compression. I remember reading an article about how Google goes to great lengths to reduce their HTML/JS fingerprint. It ends up resulting in real savings.
Is it why in most disadvantaged African societies, where young boys do not wear pants, one finds them very very sexualy potent well into their adult life.
I'd wager it's a few other factors. As someone else pointed out, contraceptives are used far less often than in the Western world. In fact, this is often attributed as one of the leading factors in the spread of AIDS. I had read an article in Scientific American that stated that something on the order of 66% of women had, at some point, exchanged money for sex in Zaire and that in these sorts of exchanges, men refuse to use any sort of birth control.
As for the age question, I'd also wager that the individuals that make it to such an old age are exceptionally healthy individuals. Since the average life expectancy is so much lower, those that live to an older age naturally represent stronger individuals.
As for the larger gentalia comment, I think you'll find that size has most to do with the size of the females hips. Men and women have evolved together such that everything fits just right. The rising rate of C-sections tends to have to do with inter-ethnic offspring. For instance, an african man marrying a chinese woman would likely result in complications as the babies head might be much larger than the woman's hips are designed to accomodate.
Why evolve to have larger bodies? I'm not sure. My first guess is it has something to do with how quickly a baby is expected to fend for themselves. A larger baby, I'm guessing, would more quickly be big enough to defend itself.
I'm sure there's been some paper written about the subject...
Any pointer toward good discussion about why existing LDAP daemon can't be used for AD interop
We discussed this mostly in person or on #samba-technical. Basically, there are a number of things about AD that won't ever be in OpenLDAP because it would comprimise OpenLDAP's standards compliance.
A very well-known example (that's quite personal for me) is connectionless-LDAP support. Active Directory supports a form of LDAP over UDP. There's a conflicting standard that's not really in use for this type of transport. The code's in OpenLDAP to support it but Kurt won't let it be easily enabled for fear that it will hamper future standization efforts.
Samba and OpenLDAP have an amicable relationship. It's just that we have different goals at the end of the day.
And as for reinventing the wheel, it's not really. Samba4 has a generic backend and an ASN1 library. Adding LDAP server-support doesn't require that much. There are some difficult parts but basic functionality isn't so bad.
Any idea what the cost is? This might be a workable solution, but I'd like to reduce the complexity as much as possible.
I'm not a sales guy but I'd reckon it'd be overkill. Here's what you should do, reproduce the problem with the latest version of Samba compiled from source. Take a network trace of the Windows BSOD. Post something to bugzilla that links to the network capture and the appropriate Samba logs.
It'll get fixed. The LDAP integration in Samba is a rather important feature and there's quite a few people on the team that tackle those sort of problems. Once it's in bugzilla, it's just a matter of time until it's fixed.
If you don't go through the effort of reproducing it with the latest version of Samba though, people will be much less willing to help out. The more leg work you do the quicker it'll get fixed.
Heck, if you've got a network trace of Samba 3.0.9 sending a bad packet to a Windows client, it'll get fixed really quickly. That's the sort of thing that catches people's eyes...
BTW, I should say that even though the solution we were trying to implement was not supported, the Redhat guy was nice enough to try and reproduce our setup and help us
We work closely with the RedHat guys. I'm surprised they didn't file a bug. Usually when they can't make progress on a bug they call us up.
that is my fault: i started that technique.
Yup.
people forget that samba is actually now about TWENTY FIVE separate protocols / APIs, about five of which are implemented in one program (nmbd), about TWENTY of which are implemented or used in smbd.
You'll appreciate this one. A couple years ago we were looking into better Active Directory interop. It was discovered that to do a real AD-style logon (the same way that XP and most Windows clients do it) required an RPC that was only available over TCP/IP (Windows refused to send it over SMB).
This led to a major hack job of Samba3 to support an end-point mapper and to support RPC over TCP. We eventually gave up and decided that AD would have to wait for Samba4.
In all fairness, once you get used to the hand-coded RPC style, it's not that bad. It's a decent subsystem as far as Samba3 goes. However, as you said, the learning curve is steep. I'm sure I still don't understand completely when to pad and when not to pad.