Indeed, in Silicon Valley back in the 1980s I remember people saying not only "Vaxen" and "Unix boxen" but "Macintoshen" and "Chipmunken" (the Chipmunk was the nickname for the HP 9836 workstation).
As it happens three of the four books I mentioned are CS books, but Software Tools seems to me to be more of a programming book. And I have read books like The Mythical Man Month and, just
recently, ESR's The Art of Unix Programming,
so my reading has not been restricted to pure computer science.
I suspect that a major factor is that almost all of my programming experience (other than very early Fortran on mainframes and assembly language and Basic on an old minicomputer) has been in a Unix environment. I've never written for the Mac or MS Windows environments, and only a little under DOS. As you say, the PC world is somewhat different from the Unix world, and books popular in one may not have the same status in the other.
Of course I can only speak of detected bugs, but no, I can't recall ever having a buffer overflow bug. I've written tens of thousands of lines of C, so it isn't because I haven't programmed or have only used scripting languages. Although the security issues with buffer overflows were not an issue when I learned C, I remember learning early on to be careful about buffer size. As far back as I can remember, I used input routines that either terminated input on reaching the end of a fixed buffer or allocated storage dynamically, used strncpy instead of strcpy, and allocated fixed size buffers large enough to contain the largest possible input rather than something typical.
Its hard to be sure, but I suspect that one reason I have always been particularly sensitive to this issue is that a lot of the software that I have written has been for the purpose of processing text, which draws ones attention to the text.
It's probably easier to get sloppy about textual input when the focus of the program is something else and the textual input is something secondary such as configuration information,
I said nothing about being better than anybody, nor did I suggest that I had read only the classics. I merely compared real classics with what the article claimed to be a classic.
Writing Solid Code may well be a perfectly useful book - I have no opinion of it since I haven't read it. What I wondered was whether it is really a classic since, unlike many real classics, I had never heard of it. Thus far the response I've seen makes me think I was right - it may be a good book, but calling it a classic is hyperbole - but I was prepared to learn that I was wrong and had missed a true classic.
When someone mentions things that he has read and you respond with a reference to the books he "claims" to have read, you're casting doubt on whether he has read them. That's the insinuation to which I referred.
The books I listed are examples of real classics. I never suggested that they dealt with secure programming. Your insinuation that I haven't read them is baseless but fitting for an Anonymous Coward.
Maybe I'm just ignorant and ill-read, but I've never even heard of Writing Solid Code, which according to the article is a classic. I somehow missed it while reading The Art of Computer Programming, The Dragon Book, The Structure and Interpretion of Computer Programs, Software Tools, and the like.
I'm also amazed at the idea that competant programmers in a decently run company can't avoid
writing software full of bugs because C and C++
lead to buffer overflow errors. They're easy enough to avoid. I've never had one in anything
I've written and its not as if I've never had a bug.
Actually, the NYT provides a way of linking directly to stories so that readers of blogs and the like can bypass the registration system. You go to this page and enter the URL of the story you want to link to. When you click "Go", it returns a link to the NYT archives that bypasses registration.
The so-called Patriot Act is bad, but it isn't a secret. Here's a link
to the complete text. Here's another.
The ACLU didn't sue to see the complete Act. We (I'm not directly involved, but I'm proud to be a card-carrying member) sued to oppose certain actions under the act. The Act makes it illegal to disclose that some actions have been taken, e.g. that a search has taken place. That's why even mentioning the actions at issue was arguably illegal and a risk for the ACLU. Here's the ACLU press release.
The ACLU also took action, initially in the form of a Freedom of Information Act request, to find out how the government has been using the Act.
Here's a link
to the ACLU's press releases on the initial FOIA
request and subsequent activity. The ACLU has all sorts of information about the "Patriot Act" here.
The support for Bush from the Middle East makes prefect sense. Who do you think Osama bin Laden wants to win? Bush has done the Islamofascists an enormous favor. They couldn't have imagined in their dreams a US President so stupid as to let up on the hunt for actual terrorists and instead invade a country that had nothing much to do with them and in the process piss off people who were formerly not angry at the US.
The Charter of Rights and Freedoms is a very good thing, but it has one serious defect, section 33, the infamous "Notwithstanding Clause", which allows both Canada and the provinces to over-ride the protections in section 2 and sections 7-15 by passing legislation to that effect. Fortunately, Canada is a sensible country and the Notwithstanding Clause hasn't been invoked in order to allow serious abuse. (It has been invoked three times. Québec has invoked it twice, once symbolically to provide a general exemption for Québec legislation and once to protect its language laws, which arguably violate the rights of non-francophones. The province of Saskatchewan has invoked it once in a labour dispute with government employees.) Still, it is a blot on an otherwise fine document. Fundamental human rights shouldn't be subject to such exceptions.
Re:How did they decide?
on
Press freedom
·
· Score: 1
The rankings are accompanied by numbers in the righthand column which seem to be some sort of numerical index on the basis of which they constructed the rankings. Unfortunately, they don't, as far as I can tell, explain what these numbers mean.
It is possible, however, to get a better idea of the basis for their characterization of the various countries. The English language section doesn't have much information (and has some screwed up links), but if you go to either the French or Spanish sections and click on the annual report and from there click on one of the regional reports, you get a summary for that region with a list of countries in a column to the right. Click on the name of a country to get the report on that country. These give you a pretty good idea of what the issues in that country are.
Once a precedent is on the books, my understanding is that it's binding in the US, no?
It isn't that simple. First, the decision of a trial court creates no precedent; only decisions of appellate courts create precedent. Secondly, the decision of an appellate court is binding only within its own jurisdiction. So, if you get a decision from the Sixth Circuit Court of Appeals, the precedent is binding within the Sixth Circuit but not in the other circuits. A decision of the US Supreme Court sets precedent binding throughout the US. So, to set precedent for the entire US, you have to go to trial and have one side or the other appeal. Only if the appeals reach the Supreme Court and only if the Supreme Court issues an opinion, is precedent set for the entire US.
I strongly disagree with Leahy's stance on IP issues, but in other respects he has been an excellant senator. (He was the speaker at my high school graduation, when he was district attorney. I did volunteer work for his first senatorial campaign.) Much as I dislike his IP stance, it would be hard to find a candidate I would consider preferable. It would have to be someone whose overall position I considered so much preferable that it was worth replacing a very senior senator like Leahy with someone with no seniority. In my case, this is academic since I don't vote in Vermont anymore, but I would hope that those who do would think twice about voting just on his IP stance.
Downloading music seems a little narrow. If they can provide the infrastructure for that, why not make them into more general network access points? Since phone booths are so prone to being vandalized, it probably won't work to provide a computer, but they could provide ports to which people could plug in their laptops, PDAs, etc. as well as devices like iPods. Maybe a wireless access point too.
One of the things revealed by the internal Diebold documents that were released some time ago is that Diebold employees repeatedly made changes to their software AFTER it had been "certified" and without notifying anyone outside of Diebold. So we know for a fact that the software actually used had NOT been subjected to independent scrutiny. In any case, both the documents and published studies of the system indicate that the system is not designed in a secure fashion and does not provide a reliable audit trail. Whatever scrutiny it may have had did not result in a secure system.
One factor may be that TRON is basically a real-time embedded OS, whereas the applications for which automotive Linux is intended are not subject to real-time constraints. I don't know TRON very well, but it is likely easier to write code for a non real-time OS than for TRON. Another factor may have to do with internationalizing textual interfaces. Linux supports Unicode. The last I knew, TRON used only its own competing encoding and did not support Unicode. (I know that TRON fans consider the TRON encoding superior.
Even if they are right, Unicode has received much more support outside of Japan so Unicode is probably a better bet for the automotive market.) A non-technical factor may be that TRON is Japanese. There may be a bit of the NIH syndrome at work.
The contracts are for the Medical Services Plan, which covers all residents of BC, and Pharmacare, which covers drugs.
The provincial government has amended the privacy laws in an effort to satisfy critics, but a lot of people aren't convinced that this will do the job. I can't find an article on the privacy law amendments right off, but here's a
link to a statement by the BC Government Employees Union (which admittedly has other concerns about outsourcing), and here is a BC government page with relevant information.
British Columbia just established safeguards to protect personal data against the United States. The province gave a contract to a US-owned firm, and many people were worried that the company would hand the data over to the US government. Rather than give the contract to a Canadian firm, the provincial government effectively banned the company from exporting the data to the US.
Goff seems to believe that high-level programming tools and genetic programming will soon eliminate the need for programmers, but nothing I have seen suggests that such tools are anywhere near able to do this. Maybe someday, but not in the near future. What we've really seen is that the combination of more advanced tools and greatly increased computer power has enabled people with very limited skills to do things that would once have required a real programmer. The result is that lots of people can make web pages and create spreadsheets and so forth, but these are essentially all additions to the work that must still be done by real programmers. Real programmers are still needed both for the more difficult tasks and for even fairly simple ones that nobody has taken the trouble to create high-level tools for.
It seems to me that there is a parallel in medicine. With all the modern medicines, tests, and instruments (such as thermometers, blood pressure meters, glucose meters) available to the average person, the non-physician can successfully diagnose and treat many illnesses. This hasn't, however, eliminated the need for real experts to deal with the harder stuff.
"The holocaust of German Jews was a hoax" will get you deported, because it is hatred based against a religious group.
No, it isn't. To begin with, Jews are not a religious group either under Jewish law or from the Nazi perspective, but that isn't my main point. It is probably true that most people who claim that the holocaust of German Jews was a hoax are Nazi apologists, but there is nothing intrinsically hateful about the statement or the belief. Someone who had no animosity toward Jews could hold this view. He or she would be wrong, but holding silly views and making silly statements isn't, and shouldn't be, a crime.
I am strongly opposed to the Canadian hate crime laws
as well as other such laws, for three reasons. First, they're subjective and easily abused to silence unpopular points of view. Second, the right of free speech is fundamental - we should be extremely cautious about interfering with it. Hurting people's feelings isn't a sufficient reason to interfere with freedom of speech, and I simply find no evidence that "hate speech" of the sort that can be controlled causes more serious harm. Finally, attempts at suppressing speech are counterproductive. They confirm the person whose speech is suppressed in his views - he is likely to think that people suppress his speech because they aren't able to counter it by facts and arguments. Other people are likely to come to the same view. The best response to wrong speech is always more speech, not suppression.
And lest anyone trot out the lame response that I'd think differently if I were involved, I am a Jew and a Canadian citizen.
I have a Trager laptop backpack with which I have been quite happy. It looks like a normal daypack but comes with a semi-rigid padded laptop case ("Vertical Transporter" TM) that fits into a cloth pocket at the bottom rear of the interior of the pack. In addition to the main section, there are two large pockets, one on top of the other, and two small pockets with vertical zippers. The pockets easily hold a mouse and power adapter. Indeed, the main section is just large enough to hold both a good-sized laptop and a large digital audio recorder), with the power adapters for both devices, microphones, cables, etc. in one pocket. I've used it to carry books too and found it comfortable with a heavy load. In addition to padded shoulder straps, it has a strong padded handle at the top. It also has a waist strap, though it is not the sort of wide, padded waist belt you get with a full scale backpack. The site I just looked at,
ebags,
lists these for US$159.99, but I'm sure I paid less than $100
for mine.
Evil intentions though he may have, Gates isn't an idiot. He may not like it, but he can see that SCO has made a complete cock-up of its anti-Linux scam.
Slashdot Mirror
Indeed, in Silicon Valley back in the 1980s I remember people saying not only "Vaxen" and "Unix boxen" but "Macintoshen" and "Chipmunken" (the Chipmunk was the nickname for the HP 9836 workstation).
As it happens three of the four books I mentioned are CS books, but Software Tools seems to me to be more of a programming book. And I have read books like The Mythical Man Month and, just recently, ESR's The Art of Unix Programming, so my reading has not been restricted to pure computer science.
I suspect that a major factor is that almost all of my programming experience (other than very early Fortran on mainframes and assembly language and Basic on an old minicomputer) has been in a Unix environment. I've never written for the Mac or MS Windows environments, and only a little under DOS. As you say, the PC world is somewhat different from the Unix world, and books popular in one may not have the same status in the other.
Of course I can only speak of detected bugs, but no, I can't recall ever having a buffer overflow bug. I've written tens of thousands of lines of C, so it isn't because I haven't programmed or have only used scripting languages. Although the security issues with buffer overflows were not an issue when I learned C, I remember learning early on to be careful about buffer size. As far back as I can remember, I used input routines that either terminated input on reaching the end of a fixed buffer or allocated storage dynamically, used strncpy instead of strcpy, and allocated fixed size buffers large enough to contain the largest possible input rather than something typical.
Its hard to be sure, but I suspect that one reason I have always been particularly sensitive to this issue is that a lot of the software that I have written has been for the purpose of processing text, which draws ones attention to the text. It's probably easier to get sloppy about textual input when the focus of the program is something else and the textual input is something secondary such as configuration information,
I said nothing about being better than anybody, nor did I suggest that I had read only the classics. I merely compared real classics with what the article claimed to be a classic. Writing Solid Code may well be a perfectly useful book - I have no opinion of it since I haven't read it. What I wondered was whether it is really a classic since, unlike many real classics, I had never heard of it. Thus far the response I've seen makes me think I was right - it may be a good book, but calling it a classic is hyperbole - but I was prepared to learn that I was wrong and had missed a true classic.
When someone mentions things that he has read and you respond with a reference to the books he "claims" to have read, you're casting doubt on whether he has read them. That's the insinuation to which I referred.
The books I listed are examples of real classics. I never suggested that they dealt with secure programming. Your insinuation that I haven't read them is baseless but fitting for an Anonymous Coward.
Maybe I'm just ignorant and ill-read, but I've never even heard of Writing Solid Code, which according to the article is a classic. I somehow missed it while reading The Art of Computer Programming, The Dragon Book, The Structure and Interpretion of Computer Programs, Software Tools, and the like.
I'm also amazed at the idea that competant programmers in a decently run company can't avoid writing software full of bugs because C and C++ lead to buffer overflow errors. They're easy enough to avoid. I've never had one in anything I've written and its not as if I've never had a bug.
Actually, the NYT provides a way of linking directly to stories so that readers of blogs and the like can bypass the registration system. You go to this page and enter the URL of the story you want to link to. When you click "Go", it returns a link to the NYT archives that bypasses registration.
The so-called Patriot Act is bad, but it isn't a secret. Here's a link to the complete text. Here's another. The ACLU didn't sue to see the complete Act. We (I'm not directly involved, but I'm proud to be a card-carrying member) sued to oppose certain actions under the act. The Act makes it illegal to disclose that some actions have been taken, e.g. that a search has taken place. That's why even mentioning the actions at issue was arguably illegal and a risk for the ACLU. Here's the ACLU press release.
The ACLU also took action, initially in the form of a Freedom of Information Act request, to find out how the government has been using the Act. Here's a link to the ACLU's press releases on the initial FOIA request and subsequent activity. The ACLU has all sorts of information about the "Patriot Act" here.
The support for Bush from the Middle East makes prefect sense. Who do you think Osama bin Laden wants to win? Bush has done the Islamofascists an enormous favor. They couldn't have imagined in their dreams a US President so stupid as to let up on the hunt for actual terrorists and instead invade a country that had nothing much to do with them and in the process piss off people who were formerly not angry at the US.
The Charter of Rights and Freedoms is a very good thing, but it has one serious defect, section 33, the infamous "Notwithstanding Clause", which allows both Canada and the provinces to over-ride the protections in section 2 and sections 7-15 by passing legislation to that effect. Fortunately, Canada is a sensible country and the Notwithstanding Clause hasn't been invoked in order to allow serious abuse. (It has been invoked three times. Québec has invoked it twice, once symbolically to provide a general exemption for Québec legislation and once to protect its language laws, which arguably violate the rights of non-francophones. The province of Saskatchewan has invoked it once in a labour dispute with government employees.) Still, it is a blot on an otherwise fine document. Fundamental human rights shouldn't be subject to such exceptions.
The rankings are accompanied by numbers in the righthand column which seem to be some sort of numerical index on the basis of which they constructed the rankings. Unfortunately, they don't, as far as I can tell, explain what these numbers mean.
It is possible, however, to get a better idea of the basis for their characterization of the various countries. The English language section doesn't have much information (and has some screwed up links), but if you go to either the French or Spanish sections and click on the annual report and from there click on one of the regional reports, you get a summary for that region with a list of countries in a column to the right. Click on the name of a country to get the report on that country. These give you a pretty good idea of what the issues in that country are.
It isn't that simple. First, the decision of a trial court creates no precedent; only decisions of appellate courts create precedent. Secondly, the decision of an appellate court is binding only within its own jurisdiction. So, if you get a decision from the Sixth Circuit Court of Appeals, the precedent is binding within the Sixth Circuit but not in the other circuits. A decision of the US Supreme Court sets precedent binding throughout the US. So, to set precedent for the entire US, you have to go to trial and have one side or the other appeal. Only if the appeals reach the Supreme Court and only if the Supreme Court issues an opinion, is precedent set for the entire US.
I strongly disagree with Leahy's stance on IP issues, but in other respects he has been an excellant senator. (He was the speaker at my high school graduation, when he was district attorney. I did volunteer work for his first senatorial campaign.) Much as I dislike his IP stance, it would be hard to find a candidate I would consider preferable. It would have to be someone whose overall position I considered so much preferable that it was worth replacing a very senior senator like Leahy with someone with no seniority. In my case, this is academic since I don't vote in Vermont anymore, but I would hope that those who do would think twice about voting just on his IP stance.
Downloading music seems a little narrow. If they can provide the infrastructure for that, why not make them into more general network access points? Since phone booths are so prone to being vandalized, it probably won't work to provide a computer, but they could provide ports to which people could plug in their laptops, PDAs, etc. as well as devices like iPods. Maybe a wireless access point too.
One of the things revealed by the internal Diebold documents that were released some time ago is that Diebold employees repeatedly made changes to their software AFTER it had been "certified" and without notifying anyone outside of Diebold. So we know for a fact that the software actually used had NOT been subjected to independent scrutiny. In any case, both the documents and published studies of the system indicate that the system is not designed in a secure fashion and does not provide a reliable audit trail. Whatever scrutiny it may have had did not result in a secure system.
Sony even has stores in some relatively small towns in Canada. There's one at the Pine Centre Mall in Prince George, B.C., (population 70k).
One factor may be that TRON is basically a real-time embedded OS, whereas the applications for which automotive Linux is intended are not subject to real-time constraints. I don't know TRON very well, but it is likely easier to write code for a non real-time OS than for TRON. Another factor may have to do with internationalizing textual interfaces. Linux supports Unicode. The last I knew, TRON used only its own competing encoding and did not support Unicode. (I know that TRON fans consider the TRON encoding superior. Even if they are right, Unicode has received much more support outside of Japan so Unicode is probably a better bet for the automotive market.) A non-technical factor may be that TRON is Japanese. There may be a bit of the NIH syndrome at work.
My highschool biology teacher taught us:
The contracts are for the Medical Services Plan, which covers all residents of BC, and Pharmacare, which covers drugs. The provincial government has amended the privacy laws in an effort to satisfy critics, but a lot of people aren't convinced that this will do the job. I can't find an article on the privacy law amendments right off, but here's a link to a statement by the BC Government Employees Union (which admittedly has other concerns about outsourcing), and here is a BC government page with relevant information.
British Columbia just established safeguards to protect personal data against the United States. The province gave a contract to a US-owned firm, and many people were worried that the company would hand the data over to the US government. Rather than give the contract to a Canadian firm, the provincial government effectively banned the company from exporting the data to the US.
Goff seems to believe that high-level programming tools and genetic programming will soon eliminate the need for programmers, but nothing I have seen suggests that such tools are anywhere near able to do this. Maybe someday, but not in the near future. What we've really seen is that the combination of more advanced tools and greatly increased computer power has enabled people with very limited skills to do things that would once have required a real programmer. The result is that lots of people can make web pages and create spreadsheets and so forth, but these are essentially all additions to the work that must still be done by real programmers. Real programmers are still needed both for the more difficult tasks and for even fairly simple ones that nobody has taken the trouble to create high-level tools for.
It seems to me that there is a parallel in medicine. With all the modern medicines, tests, and instruments (such as thermometers, blood pressure meters, glucose meters) available to the average person, the non-physician can successfully diagnose and treat many illnesses. This hasn't, however, eliminated the need for real experts to deal with the harder stuff.
No, it isn't. To begin with, Jews are not a religious group either under Jewish law or from the Nazi perspective, but that isn't my main point. It is probably true that most people who claim that the holocaust of German Jews was a hoax are Nazi apologists, but there is nothing intrinsically hateful about the statement or the belief. Someone who had no animosity toward Jews could hold this view. He or she would be wrong, but holding silly views and making silly statements isn't, and shouldn't be, a crime.
I am strongly opposed to the Canadian hate crime laws as well as other such laws, for three reasons. First, they're subjective and easily abused to silence unpopular points of view. Second, the right of free speech is fundamental - we should be extremely cautious about interfering with it. Hurting people's feelings isn't a sufficient reason to interfere with freedom of speech, and I simply find no evidence that "hate speech" of the sort that can be controlled causes more serious harm. Finally, attempts at suppressing speech are counterproductive. They confirm the person whose speech is suppressed in his views - he is likely to think that people suppress his speech because they aren't able to counter it by facts and arguments. Other people are likely to come to the same view. The best response to wrong speech is always more speech, not suppression.
And lest anyone trot out the lame response that I'd think differently if I were involved, I am a Jew and a Canadian citizen.
For those who want to use a regular backpack, the protective sleeve that comes with Trager laptop backpacks can be purchased separately, e.g. here.
I have a Trager laptop backpack with which I have been quite happy. It looks like a normal daypack but comes with a semi-rigid padded laptop case ("Vertical Transporter" TM) that fits into a cloth pocket at the bottom rear of the interior of the pack. In addition to the main section, there are two large pockets, one on top of the other, and two small pockets with vertical zippers. The pockets easily hold a mouse and power adapter. Indeed, the main section is just large enough to hold both a good-sized laptop and a large digital audio recorder), with the power adapters for both devices, microphones, cables, etc. in one pocket. I've used it to carry books too and found it comfortable with a heavy load. In addition to padded shoulder straps, it has a strong padded handle at the top. It also has a waist strap, though it is not the sort of wide, padded waist belt you get with a full scale backpack. The site I just looked at, ebags, lists these for US$159.99, but I'm sure I paid less than $100 for mine.
Evil intentions though he may have, Gates isn't an idiot. He may not like it, but he can see that SCO has made a complete cock-up of its anti-Linux scam.