1. You can say a program is (or may be) buggy when it isn't.
2. You can say a program is free of bugs when it is actually buggy.
3. You can accept the possibility that your tool will run forever.
Actually, (1) characterizes tools like Coverity. I've also run into cases where a Hindley-Milner styled type inference needs a type annotation hint, or it wouldn't infer a type for my program.
(2) characterizes almost every language where the type system is not expressive enough. For example, the type of the fgets() function in C doesn't establish any relationship between the char *s and int n. I could write this code:
int main() { char buf[80]; fgets(buf, 1000, stdin); }
And the C compiler will accept as correct, but my program is buggy. It is more obvious for dynamically typed languages like Python or Javascript, where the program runs despite latent type errors in the program.
As for (3), I vaguely remember a passing fact that type inference for impredicative polymorphism is undecidable. Maybe that's something you're working on? I'd like to hear about it.
Coverity is really a code review tool. From your code, it tries to construct a model that shows your code is correct (static analysis + type inference). If it can't, the code is flagged, and it should be reviewed by a human. The flagged code may or may not be a bug, only that Coverity couldn't prove its correctness. If anything, I would advocate that the code should be rewritten in order to pass Coverity check, in the same spirit that if another competent person doesn't understand your code, you should probably rewrite it to make it more clear.
However, I've not seen any formal soundness proof of Coverity itself. As a result, Coverity may very well accept buggy programs as correct. This would certainly limit the tool's usefulness.
When I said context and train of thought, let me highlight a few places that it went on wreck in the conversation.
Since the bot first indicated that she was born in Calais, France, there is an implicit assumption that Calais is in France. The bot contradicted her earlier statement (if she remembers it) by saying "Shall I assume that's in France?"
I asked about University of Hawaii's fine arts program because the bot indicated that she is studying a doctorate in fine arts there. Then she contradicted herself by saying she never went there.
It's not just about keeping track of a topic. It is about being able to relate the statements said recently and deduce new information. For example, I could tell you that a + b = c, and c + d = 5. Then I ask you what is (a + b) + d? You would answer 5. The computer can't answer that by logical deduction. It can if you build a linear systems solver into the program.
You would be a genius if you build a chat bot that can do that, either with proper logical deduction, or with a linear systems solver.
xyzzy - where are you from? [ Af:0 At:7 Sp:0] I was born in Calais, France. xyzzy - Where is that in France? [ Af:0 At:7 Sp:0] It's that way. (points east). xyzzy - Where is Calais in France? [ Af:0 At:7 Sp:0] Shall I assume that's in France? xyzzy - You said you were born in Calais, France. [ Af:0 At:7 Sp:0] You can't trust what I say. I suffer from Multiple Personality Disorder. xyzzy - Where are you from? [ Af:0 At:7 Sp:0] I live in Kailua, since I am an exchange student doing my doctorate at the University of Hawaii. xyzzy - What do you study? [ Af:0 At:7 Sp:0] My major is Fine Art. xyzzy - How is University of Hawaii's fine art program? [ Af:0 At:7 Sp:0] I wish I knew. Never been there. Always wanted to go, though.
I think the challenge for computer AI bot is to remember the context of the conversation. Unlike humans, AI bots have no train of thought. I think the bot designers know that, and they design the bot to coerce you into a contextless conversation.
The part that most of the right-wing nutjobs seem to forget is that with the coming of the Messiah and the creation of the New Testament, the Old Testament was fulfilled, and the rules set forth no longer apply. That was the purpose of the NT: to set forth a new set of rules to live your life by, and to create a new covenant with God.
Matthew 5:17-18 records Jesus to have said this during the sermon on the mountain,
Do not think that I have come to abolish the Law or the Prophets; I have not come to abolish them but to fulfill them. I tell you the truth, until heaven and earth disappear, not the smallest letter, not the least stroke of a pen, will by any means disappear from the Law until everything is accomplished.
You said,
I also have issues with a god who doesn't want his followers to address him by name. ("God" is a job description. The name is either Yahweh, or Jehovah, depending on which pronunciation of the Hebrew you prefer.)
Deuteronomy 5:11, as part of the Ten Commandment, says,
"You shall not misuse the name of the LORD your God, for the LORD will not hold anyone guiltless who misuses his name.
It is only a tradition of the people, not a law of God, that forbids the use of the name. However, the exact name need not appear in an expression if the expression is meant to dishonor God. You're judged by the effect of your conduct, not by whether you observe the technicality of the law. I would invite you to read Matthew 15 where Jesus talks about how teachers and experts of the law actually break law for the sake of tradition.
My parting word for you is that don't let homosexuality separate you from the love of God. There are many sins people have that are not as visible as homosexuality. The lack of visibility of those other sins doesn't make them more right to criticize your homosexuality, but don't let homosexuality drive you away from God.
If we take energy out of the global climate system using more sail boats, we would still have the global warming, but we might end up with fewer severe storms.
My guess is that the state of Minnesota already has developed custom software in-house that depends on the Microsoft platform. You migrate to a Microsoft cloud if your existing software is already locked into Microsoft platform. That's only natural. At any rate, the whole cloud computing concept is very simple: let someone who is good at running data center do their job.
You could very well argue that redeveloping the software to base on an open-source platform might be a better plan in the long run, and I would tend to agree with that. But the redevelopment will surely cost the state some millions of dollars more in the short run.
You also probably don't realize that software costs money to develop. Even when the software is offered to you for free, someone, somewhere is paying for it. That's because someone has to spend time doing something. In order to sustain the livelihood, that person needs to eat, drink, pay rent, and once in a while use medical help. When software gets open sourced, the person is donating his time and effort and has absorbed the cost of writing the software.
And don't forget that free software is not really about the cost. It's about the freedom to learn from the software, to modify the software, and to distribute your modifications.
They first observe that plants start to spontaneously grow again in contamination sites despite the high radioactivity. Then they brought in seeds from uncontaminated origin. One batch goes to the contamination site, and another batch (the controlled group) goes to a decontaminated area near the site. Seeds grow fine in both batches, showing that seeds from uncontaminated origin is able to survive the radioactivity in the very first generation. The study is about the mechanism how plants naturally resist radioactivity. No evolution is taking place here.
I think he's worried that you could IM a friend during an exam to work the answers out for you, as if you're a thin client, with all that computing power over in the cloud.
Why limit your imagination at the boot loader? It is my opinion that these Windows applications should flash your BIOS instead of screwing with master boot record for the purpose of copy protection. That way, if they mess up, the computer won't even power up.;-)
The leaks were found to be true not because of the credibility of its founder. In fact, the organization itself has had no credibility at all, only the material being leaked speaks for itself. I really don't care if Julian rapes women as a hobby. If he gets arrested, what is important is that someone must continue his work.
Well, be careful with Jazz because there is also big-band Jazz.
It is true that we already have musical pieces with a quartet, quintet (or more) kind of setup for classical and modern music genres, but I don't believe I've seen a piece where you have 20-40 musicians all playing different parts. It's been done with actors in an opera or musical, but not with just the musical instruments.
Avatar uses real human actors and capture their action as well as facial expression and emotion, and then use that as the basis to synthesize a performance. Notion3 is actually similar but the motion/emotion capturing is much more primitive. The live performance mode in Notion3 allows a conductor---or a technician following a conductor---to use just one key on a MIDI keyboard to play a score. The MIDI keyboard captures the dynamics by recording key velocity as well as tempo. They then use that information to synthesize a performance based on audio samples recorded from London Symphony Orchestra.
While Avatar probably wouldn't be successful if they only had one person play all characters, the success of Notion3 where one person plays the whole orchestra is kind of interesting. It shows that when you're part of an orchestra ensemble, the amount of individual character you contribute to the group is negligible. This would probably motivate more musicians to pursue a solo career, or inspire a music genre where all the instrument pieces are part of a dialogue rather than just playing in unison.
The shape deformation is caused by the centrifuge that spins the rubber band, as a result of that particular experiment setup. I argue that no such deformation would happen if you use a treadmill instead.
Let's say the record producer asks your son if he likes the idea of making some money by selling the CD to his friends. Sure, it beats running a lemonade stand. The producer asks your son to give back $2 for every copy he sells, but he can sell it at any price he wants (the producer suggests $5). So your son starts making some money. And then one day, he notices that his friend is giving free copies to other friends, and your son stops making any more money. Knowing your son is the only person allowed to sell copies around the neighborhood, wouldn't you and your son get upset at that friend?
Human greed is only obvious when you turn it around against you.
I assume that it is nearly impossible to do an exemplary job teaching one course for one semester.
In practice, the department rotates teaching fellow as well as the professor teaching the course every semester except for certain highly specialized courses. However, past experiences in teaching carries over to new subjects. I've done about 8 semesters already. The first few times I did see some improvements needed on my part. The more I did it, the more I realize that teacher-student is a mutual relationship, and unrequited teaching won't do the student any good.
For example, if the A+ students had complained you were going to slowly and explaining things too much, but the B & C students were very happy with the explanations and needed them to keep from getting a much lower grade, then the explanations should be kept
That makes sense, but here is the problem with our feedback system: you don't know who reported the feedback because evaluations are written anonymously.
Students are also inherently myopic; no student sees the big picture. One student self-identified as a smart student but sympathized for other slower students that I went too fast. He failed to consider that I also spent time outside of the classroom to help with the slower ones. I've always used up my classroom time, and it's not like I was rushing to get myself out of the class.
The preface of the new translation of Philagrius is interesting to read. The translation you referenced was mentioned on Slashdot a while ago, something about the earliest dead parrot sketch. I read this translation from 1920 but I skipped the introduction.
Maybe instead of blaming students (many of whom are going into many thousands of dollars in debt to hear you speak) for wanting to be "spoon fed", maybe you should think about how good of a job you actually did.
There is a fallacy in your argument. You assume that by blaming my students, I must have done a bad job teaching. The conclusion about my teaching skill may or may not be true, but it could not be justified by my comments towards my students alone. It would only be plausible if I told you I did this, and you tell me I could have done that. Even so, that only suggests room for improvement and not necessarily indicates that the baseline performance before the improvement was poor.
How do you appease a group of students, some of which says I covered too much X, and some says I didn't cover enough X? Do you find contradictory feedback reasonable?
I still maintain that the students are paying for things they could neither appreciate nor afford. Going to school is just a dumb decision for most of them. I hope you really get your money's worth of education. Sometimes it means you have to humble yourself and admit that you could have done better to learn from the teacher, not that the teacher could have done better to teach you.
I tried to do what Socrates would do as a teaching fellow. A student complained in the course evaluation that I should not ask questions before they learn about the subject. I also took a top-down approach, reviewing course material in a way hoping that if a student is interested or needed to know more, he would ask questions to clarify. Several students commented that I glossed over detail and that I was too vague. A number of students thought I was boring. I tried to explore related knowledge that they wouldn't find on textbook, and a few students said that I digressed, and I went off tangent.
On the positive side, one student did comment that he found my method of teaching intellectually challenging, and that he learned a lot, which he described as a strength.
Socrates would not have wanted to teach students who only want to be spoon fed, who do not want to learn, who only want to get good grades, and eventually their diploma, so they could work for an industry that they're just in it for the money. Students nowadays feel entitled. Entitled to be knowledgeable without making the effort to learn. Entitled to get good grades without learning anything. Entitled to get diploma without passing courses. Entitled to get a job. Entitled to get high salary. Entitled to get their material satisfaction. And finally, entitled to blame everyone else including their college professor if they don't get what they want.
On one hand, Apple could have very well done the same with other parts of the software, providing fix without disclosure. This goes on to say that vulnerability disclosure is a very poor indicator of software quality. However, in this case, it could have said something as trivial as "updated malware signature database." It's not fixing a vulnerability.
On the other hand, this article highlights the very interesting fact that there *is* a market for anti-virus software, even when the base OS is robust and secure. The base OS could be immune to virus and malware attack when there is no user action involved. However, the user could become the weak link to compromise their own system. Anti-virus software prevents high-risk users from being affected by their reckless action.
It's just like how only certain people need to be HIV tested regularly. You only need to worry about HIV infection if you received blood transfusion, or if you engaged in promiscuous sexual act (willfully or as a rape victim). If you did neither, then you don't need to be tested, hence you don't need to spend money on the pharmaceutical products for the HIV test. You should definitely be tested regularly if you know what you do carries a high risk of contracting HIV.
You may still need anti-virus software, depending on if what you do online carries a high risk of contracting malware. It has less to do with whether your operating system is secure.
You have to admit, you actually have no evidence one way or another. I don't have to tell you anything about myself to win an argument.
Whatever we do in this life that is an 'achievement', we don't do "all by ourselves"... I do honestly feel that there IS a God, and he does "inspire us to greatness" @ times, & it's not just "us alone" doing it is all... which is largely what that quote from Corinthians is after all, ALL about.
I'm glad you at least humble yourself before God. I have the impression that you don't humble yourself before other men, so no matter what I say you won't listen. That's okay. Why don't you do a little research on your own about who wrote the books of Corinthians, and read at least Luke and Acts in entirety? That would give you enough context to understand where Paul was coming from. Then, go back to 1 Corinthians 15:9, the verse before the one you quoted. You really shouldn't take the bible out of its context. A lot of harm has been done that way.
You are the person who started using your credential to justify your argument, i.e. Argument from authority, therefore Argumentum Ad Hominem is all you're going to get. I, on the other hand, speak of objective arguments that don't depend on who says it. That's why I can afford to be an anonymous person, but you can't.
Who in their right mind is going to write hand-optimized inline assembly for a CD alarm clock program that spends 99.999% the time idling, waiting for the timer to go off? You did. I think you're crazy.
And finally, you read the bible (good for you), but you don't understand what it says.
Slashdot Mirror
Actually, (1) characterizes tools like Coverity. I've also run into cases where a Hindley-Milner styled type inference needs a type annotation hint, or it wouldn't infer a type for my program.
(2) characterizes almost every language where the type system is not expressive enough. For example, the type of the fgets() function in C doesn't establish any relationship between the char *s and int n. I could write this code:
And the C compiler will accept as correct, but my program is buggy. It is more obvious for dynamically typed languages like Python or Javascript, where the program runs despite latent type errors in the program.
As for (3), I vaguely remember a passing fact that type inference for impredicative polymorphism is undecidable. Maybe that's something you're working on? I'd like to hear about it.
Coverity is really a code review tool. From your code, it tries to construct a model that shows your code is correct (static analysis + type inference). If it can't, the code is flagged, and it should be reviewed by a human. The flagged code may or may not be a bug, only that Coverity couldn't prove its correctness. If anything, I would advocate that the code should be rewritten in order to pass Coverity check, in the same spirit that if another competent person doesn't understand your code, you should probably rewrite it to make it more clear.
However, I've not seen any formal soundness proof of Coverity itself. As a result, Coverity may very well accept buggy programs as correct. This would certainly limit the tool's usefulness.
When I said context and train of thought, let me highlight a few places that it went on wreck in the conversation.
It's not just about keeping track of a topic. It is about being able to relate the statements said recently and deduce new information. For example, I could tell you that a + b = c, and c + d = 5. Then I ask you what is (a + b) + d? You would answer 5. The computer can't answer that by logical deduction. It can if you build a linear systems solver into the program.
You would be a genius if you build a chat bot that can do that, either with proper logical deduction, or with a linear systems solver.
I think the challenge for computer AI bot is to remember the context of the conversation. Unlike humans, AI bots have no train of thought. I think the bot designers know that, and they design the bot to coerce you into a contextless conversation.
Matthew 5:17-18 records Jesus to have said this during the sermon on the mountain,
You said,
Deuteronomy 5:11, as part of the Ten Commandment, says,
It is only a tradition of the people, not a law of God, that forbids the use of the name. However, the exact name need not appear in an expression if the expression is meant to dishonor God. You're judged by the effect of your conduct, not by whether you observe the technicality of the law. I would invite you to read Matthew 15 where Jesus talks about how teachers and experts of the law actually break law for the sake of tradition.
My parting word for you is that don't let homosexuality separate you from the love of God. There are many sins people have that are not as visible as homosexuality. The lack of visibility of those other sins doesn't make them more right to criticize your homosexuality, but don't let homosexuality drive you away from God.
If I want to steal cold-war nuclear weapons, I would build a robot that appears just like that, in case a run-in encounter is unavoidable.
Also, the radio signal that these robots emit is like a beacon saying "I'm coming, triangulate me."
If we take energy out of the global climate system using more sail boats, we would still have the global warming, but we might end up with fewer severe storms.
Just a thought.
Wind, anyone? Zero carbon emission, no radioactive pollutant.
Now, explain to me, what exactly is your criteria for calling people names?
My guess is that the state of Minnesota already has developed custom software in-house that depends on the Microsoft platform. You migrate to a Microsoft cloud if your existing software is already locked into Microsoft platform. That's only natural. At any rate, the whole cloud computing concept is very simple: let someone who is good at running data center do their job.
You could very well argue that redeveloping the software to base on an open-source platform might be a better plan in the long run, and I would tend to agree with that. But the redevelopment will surely cost the state some millions of dollars more in the short run.
You also probably don't realize that software costs money to develop. Even when the software is offered to you for free, someone, somewhere is paying for it. That's because someone has to spend time doing something. In order to sustain the livelihood, that person needs to eat, drink, pay rent, and once in a while use medical help. When software gets open sourced, the person is donating his time and effort and has absorbed the cost of writing the software.
And don't forget that free software is not really about the cost. It's about the freedom to learn from the software, to modify the software, and to distribute your modifications.
Read their method.
They first observe that plants start to spontaneously grow again in contamination sites despite the high radioactivity. Then they brought in seeds from uncontaminated origin. One batch goes to the contamination site, and another batch (the controlled group) goes to a decontaminated area near the site. Seeds grow fine in both batches, showing that seeds from uncontaminated origin is able to survive the radioactivity in the very first generation. The study is about the mechanism how plants naturally resist radioactivity. No evolution is taking place here.
I think he's worried that you could IM a friend during an exam to work the answers out for you, as if you're a thin client, with all that computing power over in the cloud.
Why limit your imagination at the boot loader? It is my opinion that these Windows applications should flash your BIOS instead of screwing with master boot record for the purpose of copy protection. That way, if they mess up, the computer won't even power up. ;-)
The leaks were found to be true not because of the credibility of its founder. In fact, the organization itself has had no credibility at all, only the material being leaked speaks for itself. I really don't care if Julian rapes women as a hobby. If he gets arrested, what is important is that someone must continue his work.
Well, be careful with Jazz because there is also big-band Jazz.
It is true that we already have musical pieces with a quartet, quintet (or more) kind of setup for classical and modern music genres, but I don't believe I've seen a piece where you have 20-40 musicians all playing different parts. It's been done with actors in an opera or musical, but not with just the musical instruments.
Avatar uses real human actors and capture their action as well as facial expression and emotion, and then use that as the basis to synthesize a performance. Notion3 is actually similar but the motion/emotion capturing is much more primitive. The live performance mode in Notion3 allows a conductor---or a technician following a conductor---to use just one key on a MIDI keyboard to play a score. The MIDI keyboard captures the dynamics by recording key velocity as well as tempo. They then use that information to synthesize a performance based on audio samples recorded from London Symphony Orchestra.
While Avatar probably wouldn't be successful if they only had one person play all characters, the success of Notion3 where one person plays the whole orchestra is kind of interesting. It shows that when you're part of an orchestra ensemble, the amount of individual character you contribute to the group is negligible. This would probably motivate more musicians to pursue a solo career, or inspire a music genre where all the instrument pieces are part of a dialogue rather than just playing in unison.
The shape deformation is caused by the centrifuge that spins the rubber band, as a result of that particular experiment setup. I argue that no such deformation would happen if you use a treadmill instead.
Let's say the record producer asks your son if he likes the idea of making some money by selling the CD to his friends. Sure, it beats running a lemonade stand. The producer asks your son to give back $2 for every copy he sells, but he can sell it at any price he wants (the producer suggests $5). So your son starts making some money. And then one day, he notices that his friend is giving free copies to other friends, and your son stops making any more money. Knowing your son is the only person allowed to sell copies around the neighborhood, wouldn't you and your son get upset at that friend?
Human greed is only obvious when you turn it around against you.
In practice, the department rotates teaching fellow as well as the professor teaching the course every semester except for certain highly specialized courses. However, past experiences in teaching carries over to new subjects. I've done about 8 semesters already. The first few times I did see some improvements needed on my part. The more I did it, the more I realize that teacher-student is a mutual relationship, and unrequited teaching won't do the student any good.
That makes sense, but here is the problem with our feedback system: you don't know who reported the feedback because evaluations are written anonymously.
Students are also inherently myopic; no student sees the big picture. One student self-identified as a smart student but sympathized for other slower students that I went too fast. He failed to consider that I also spent time outside of the classroom to help with the slower ones. I've always used up my classroom time, and it's not like I was rushing to get myself out of the class.
The preface of the new translation of Philagrius is interesting to read. The translation you referenced was mentioned on Slashdot a while ago, something about the earliest dead parrot sketch. I read this translation from 1920 but I skipped the introduction.
There is a fallacy in your argument. You assume that by blaming my students, I must have done a bad job teaching. The conclusion about my teaching skill may or may not be true, but it could not be justified by my comments towards my students alone. It would only be plausible if I told you I did this, and you tell me I could have done that. Even so, that only suggests room for improvement and not necessarily indicates that the baseline performance before the improvement was poor.
How do you appease a group of students, some of which says I covered too much X, and some says I didn't cover enough X? Do you find contradictory feedback reasonable?
I still maintain that the students are paying for things they could neither appreciate nor afford. Going to school is just a dumb decision for most of them. I hope you really get your money's worth of education. Sometimes it means you have to humble yourself and admit that you could have done better to learn from the teacher, not that the teacher could have done better to teach you.
I tried to do what Socrates would do as a teaching fellow. A student complained in the course evaluation that I should not ask questions before they learn about the subject. I also took a top-down approach, reviewing course material in a way hoping that if a student is interested or needed to know more, he would ask questions to clarify. Several students commented that I glossed over detail and that I was too vague. A number of students thought I was boring. I tried to explore related knowledge that they wouldn't find on textbook, and a few students said that I digressed, and I went off tangent.
On the positive side, one student did comment that he found my method of teaching intellectually challenging, and that he learned a lot, which he described as a strength.
Socrates would not have wanted to teach students who only want to be spoon fed, who do not want to learn, who only want to get good grades, and eventually their diploma, so they could work for an industry that they're just in it for the money. Students nowadays feel entitled. Entitled to be knowledgeable without making the effort to learn. Entitled to get good grades without learning anything. Entitled to get diploma without passing courses. Entitled to get a job. Entitled to get high salary. Entitled to get their material satisfaction. And finally, entitled to blame everyone else including their college professor if they don't get what they want.
On one hand, Apple could have very well done the same with other parts of the software, providing fix without disclosure. This goes on to say that vulnerability disclosure is a very poor indicator of software quality. However, in this case, it could have said something as trivial as "updated malware signature database." It's not fixing a vulnerability.
On the other hand, this article highlights the very interesting fact that there *is* a market for anti-virus software, even when the base OS is robust and secure. The base OS could be immune to virus and malware attack when there is no user action involved. However, the user could become the weak link to compromise their own system. Anti-virus software prevents high-risk users from being affected by their reckless action.
It's just like how only certain people need to be HIV tested regularly. You only need to worry about HIV infection if you received blood transfusion, or if you engaged in promiscuous sexual act (willfully or as a rape victim). If you did neither, then you don't need to be tested, hence you don't need to spend money on the pharmaceutical products for the HIV test. You should definitely be tested regularly if you know what you do carries a high risk of contracting HIV.
You may still need anti-virus software, depending on if what you do online carries a high risk of contracting malware. It has less to do with whether your operating system is secure.
You have to admit, you actually have no evidence one way or another. I don't have to tell you anything about myself to win an argument.
I'm glad you at least humble yourself before God. I have the impression that you don't humble yourself before other men, so no matter what I say you won't listen. That's okay. Why don't you do a little research on your own about who wrote the books of Corinthians, and read at least Luke and Acts in entirety? That would give you enough context to understand where Paul was coming from. Then, go back to 1 Corinthians 15:9, the verse before the one you quoted. You really shouldn't take the bible out of its context. A lot of harm has been done that way.
You are the person who started using your credential to justify your argument, i.e. Argument from authority, therefore Argumentum Ad Hominem is all you're going to get. I, on the other hand, speak of objective arguments that don't depend on who says it. That's why I can afford to be an anonymous person, but you can't.
Who in their right mind is going to write hand-optimized inline assembly for a CD alarm clock program that spends 99.999% the time idling, waiting for the timer to go off? You did. I think you're crazy.
And finally, you read the bible (good for you), but you don't understand what it says.