You don't need to idiot-proof it, you just need to make sure it doesn't behave ungracefully when assaulted by an idiot.
It's impossible to do that. Turing demonstrated that it is not possible to determine whether any given algorithm will execute to completion for all possible inputs. As the library in question is a mathematical one, it will undoubtedly contain algorithms which will not complete for some input or inputs, and all the bounds-checking in the world cannot guarantee security from input which will cause an infinite execution time. If it was possible, it would be a solution to the Turing machine halting problem, and such a thing cannot be, by definition.
I applied for JobSeeker's Allowance a few months ago and got turned down because I didn't pay enough National Insurance [~12% tax on your income] 4 years ago
You have to watch out for the two checkboxes right at the start of the claim form. If you only tick "Contributions-based Jobseekers Allowance", you can miss out that way (although it shouldn't, IIRC, be based on 4 year old contributions). You must tick "Income Based JSA" as well, and fill in the 15 or so extra pages of means testing questions. If you've got more than £3000 savings, you'll lose money; if you've got more than £8000, you're not entitled at all. You have to sign an agreement stating the kinds of work you'll be looking for and how many job applications per week (at least two, though a graduate would be expected to do more), and be able to prove when they ask that you have done what the agreement states (they write the agreement). After 6 months, you're off contributions-based allowances no matter what.
If you aren't entitled to JSA, you should be able to get Income Support at a lower rate. IIRC, JSA at the moment is just over £50/week, Income Support rather less; those figures are also reduced to the ~£43/week if you're under about 25, as you're apparently supposed to sponge off your parents, whether you have any or not. Those figures are for singletons; if you have a partner, you'll be jointly means tested, but if you're both on the dole, you won't get twice as much.
Even when you get those sorted out, you have to deal with Housing Benefit to cover the cost of a roof over your head; and that itself won't cover the full cost, so you'll need about a quarter of the JSA/IS to make up your rent. If you've got a mortgage, start collecting the cardboard boxes; you get sod-all help, so you'll be living in them soon.
Still, if you think it's bad now, you should have tried it around 1990. Graduates had to sign that they were willing to work anywhere in the country; if you lived in Cornwall and they came up with a job in Aberdeen, you were supposed to move to get the job, just because you had a degree. Then that mad witch Thatcher got booted out, and a very slight amount of sanity began to creep back into the country.
Check out the Department of Work and Pensions for more info. Also, see if your local council has a welfare rights department (tip: if they're Tory, they probably won't).
Actually, I think they are just able to buy international call access at bulk rates. There's a lot of spare capacity on the international phone network, and buying it in bulk makes this sort of thing cost effective.
It's like those shops full of phones offering cheap calls abroad - the UK city I live in has about 30% population of Indian/Pakistani/Bangladeshi descent, and in certain areas there are loads of these places. They are reselling phone time that has been bought in bulk from someone like BT. Same thing for the cards that give you so many minutes of cheap international calls from your home phone by dialling a special prefix code.
AFAIK, UK businesses have to respect your registration with the Telephone Preference Service, even if the calls are originated in an offshore call centre.
Slashdot will be closed this Thursday, as it is necessary to shut it down and clean it once a year.
Re:you think the bunny was bad...
on
Easter Humor
·
· Score: 1
And if you really want to have fun, you can open his net statistics in a separate window/tab, reload the original page, then reload the statistics to see if you're in the "last 10 visitors"! Took me three goes to get there in time at all, and I was already down to number 8:-)
Well, it's a holiday, so I really do have too much free time. Anyway, there's someone at Boeing who isn't doing any work, either.
Nice bug. They really encourage people to pirate so-called corporate versions (no activation needed).
If you check out the Reg story, you'll see that the bug causes those corporate "no activation" versions to demand activation. That is the fundamental problem.
My Ericsson A2618s has such a bad user interface, I've thought of writing a paper pointing out its flaws from the perspective of HCI.
One of its amazing features is that the longer a text message gets, the longer it takes for it to respond to letter selection. (No predictive text, thank God.) I've easily managed to get 10 to 12 characters ahead of it. Presumable every time it adds a character to a string, it copies the whole string to a buffer one character longer and then appends the new char...
That's a performance issue, but what about the way it unlock the keypad when it receives a text? Knocking the "yes" key brings up the message, knocking it again brings up the subsequent menu, and knocking it a third time selects the first option of the subsequent menu - which is "Delete".
Several times I've received texts which had been deleted by the time I got the phone out of my pocket.
f you want to write code that *anyone* will be able to run, you can't use Java
I keep meaning to write a Java app on my Mac and then get it running on my PC, just to see what the difficulties are these days. In theory, the write-once-run-anywhere stuff is true these days, but I'm going to suck it and see before I make any claims in that department;-)
I haven't tried any inline assembler yet; do you have any tips?
Afraid not; the last time I mixed assembler with high level was in Forth, which probably isn't much help when it comes to gcc.
However, you're spot on with the idea of coding in high level first, then converting to assembler. Specifically, I'd suggest getting an app working, then if you need extra performance, profile it and recode the bits that are bogging you down. That's the way we used to work in Forth, and it offers several advantages: first, you know that your logic is correct; second, if you find a bug in your logic, you can go back, switch in the high level stuff, fix the logic in there, then convert your changes back into the assembler code; third, if your unit tests (you did write your unit tests, right?) pass with the high-level and not the assembler, you know that the mistake is in your conversion, not your logic.
Cautionary tale:
Back in the early 80s, a friend at a company where I used to work had a bug in his assembler conversion of a Forth word (== subroutine). He spent hours trying to work out how he'd gone wrong - the high level code worked perfectly, the low level failed every time. Eventually he called in the project manager, a phenomenal coder who should never have been put onto managment. This guy sat there testing the whole thing from basics (never believe what the guy who wrote it says about it working;-), and finally brought a hex dump of the machine code up on the screen. He scanned through it, and suddenly pointed at a single byte: "It's there!" He switched back to the source, pointed at the relevant mnemonic, and stated "That's correct: it's a bug in the assembler." And it was...
Among other morals of this tale, people should note one thing: if a programmer is phenomenally good, don't make him a manager; just pay him more. None of the programmers there would have resented him being paid more, because they all recognised his superiority. Trouble is, management judge others by status, and coders judge by ability. He was the worst manager I've ever come across, but I learnt more about finding bugs from one 45 minute debugging session with him than from 10 years of programming.
That's about all you have to do, really, to keep people happy. Leave them alone, let them do their thing, keep up the supply of interesting things to do, don't push them unless you really have to, feed them lots of coffee, and let them dress comfortably.
I've worked at places like this. Funnily enough, they always seem to be managed by people who used to write code, and owned by people who know how to listen to their management. It's so simple, you'd think the rest of the industry would have cottoned on by now. Then again, our species' capacity for exalting mediocrity never ceases to astonish me.
I'm not surprised you're happy in your work. I hope it keeps up for you:-)
I can't argue with you, and don't want to. I think, though, that you're talking about writing software you will use; I'm sadly trapped in the hideous world of doing stuff to pay the rent. (Not that there is any stuff to do the moment, but we all have to tighten our belts in wartime;-)
And, I like the libraries I can get for C++. CPAN's great, but... I genuinely *like* C++.
Love the language that works for you, but as the song says, "If you can't be with the one you love, love the one you're with":-)
I have to say that if there's one language I really like to work in, it's Forth. I've written my own Forth language implementation and development environment before now (for the Atari ST) ; ultimately, when I'm working for my own pleasure, I want the language I like, and the devil take the hindmost. (I haven't had the chance to get paid for it since 1989, but I still solve Countdown's Numbers game on a stack in my head.)
Unfortunately, those people who employ - or fail to employ - consultants like me want stuff to work on their systems, and so I sell my soul. How much work does a Forth specialist get these days, as a regular gig?
Specifically, if we don't have a lot of money (I definitely don't) we can't afford a particularly fast machine. For example, I use a Pentium 167MMX laptop with 96MB of Ram and 4GB of disk at home.
My PC, brushing against my leg as I type, started life as an empty case that I put my dad's old 286 motherboard in; believe me, I know about making do with what you can afford. (FWIW, it's currently a 500MHz AMD I got as the cheapest available from the local computer shop when the second-hand AMD 100MHz went west. Still got a 4Gb drive in it.) I'm thinking about philosophical differences between approaches.
The reason we regarded people who programmed in C with a degree of pity (this was 1987 - 88) was because setting up and breaking down all those stack frames used up valuable cycles. On an 8086, a
mov ax,0
was less efficient than a
sub ax,ax
or (my opcode of choice) a
xor ax,ax
by a couple of cycles. That meant everything when a single byte read from/write to a CGA card's video RAM cost betwen ~20 to ~50 wait states for an 8MHz 8086.
That's the level we were working to when I developed my dislike of C; my dislike of later manifestations of C and C++ comes down to reading those ads in Doctor Dobbs Journal saying "Can you tell where the bug is? PC-Lint can!". I should be able to tell where the bug is; I wrote it, for goodness sake!
I just feel that the point in the original article about us programmers being more attracted to cool, tricky stuff than to meeting real-world requirements was right generally, as well as in the context of application security. I have to combat my desire to do cool stuff every day, because I have to do stuff that satisfies my customers' desire for functional, safe stuff. It's boring, but it keeps me off the streets - or, at the moment, fails to.
BTW, I don't actively recommend or discommend Perl; I think Perl is just bizarre. If that's how Larry Wall thinks about programming (see the Camel Book intro), I'm just glad there's room for a lot of opinions in this world:-)
As an aside, can I ask, do you optimize your stuff by dropping into assembler for critical portions of your code? I'd highly recommend it, because in my experience, anybody who appreciates the power available from C or C++ really gets off on assembler. (Sorry if that sounds patronising; don't want to teach my grandmother to suck eggs.)
I don't have anything against any language, and I certainly appreciate your situation as regards processing power
Things like memory allocation problems, dangling references, etc. It's possible to do all this stuff yourself, but why bother when tools and libraries are available to do it for you? This was the point made in the article, and I agree.
You're right that one should choose the correct tool for the job; I just believe that a heck of a lot of stuff is done in C or C++ that could be done with higher-level tools. You doubtless remember that COBOL was one of the most widely used languages for business applications for years; I get the impression that a lot of people nowadays are using C/C++ for applications that would, frankly, be better written in COBOL. (And if there's one language I don't want to deal with, it's COBOL).
Charles Moore, inventor of Forth, is quoted in Leo Brodie's book Thinking Forth:
(Programmers) like to do these elaborate difficult things. But there comes a time when the world is going to have to quit programming keypads and converting numbers to binary, and start solving problems.
1984 edition, p. 117
C was developed primarily as a tool for writing Unix. C++ was developed as a way of adding this cool, new (back then) OO approach to C. There's a place for both languages, but it would take a lot to convince me that either of them is the only appropriate choice for the majority of purposes to which they are actually put.
Programmers like stuff that's difficult because if you don't, you wouldn't be a programmer. But I prefer the language to stay out of the way of my exploration of my ideas as much as possible, and buggering about with malloc() is simply not relevant to the majority of problems. It's a distraction. I can make serious errors in the logic of my apps without having to be held up by null pointer exceptions that wouldn't happen with another language doing the system-level stuff for me. In the same way, if I could afford it I'd have a cleaner in; I want to live in the house, not be a housekeeper.
All IMHO, of course:-) And the next time I want to write an OS, I'll certainly use a tool like C.
a tendency to believe that one's own code is well-written, and a corresponding belief that real coders, like fighter pilots, work as close as possible to the bare metal: Real programmers manipulate the system at the lowest possible level, for the maximum possible effect.
The many posts here by people saying "I gotta use C/C++, that other stuff's too slow" prove exactly this point.
In the 80s, I worked for a number of years as an assembly language programmer (6502, 6809, Z80, x86, 680x0, ARM); we didn't use C (in fact, we sneered at people who did) because it just couldn't offer the performance of hand-crafted assembler (and I'm talking 30,000+ lines here).
Now I do almost all of my design prototyping in JavaScript (quick turnaround on the modify/compile/test cycle), and code real stuff in Java or (occasionally) C#. Why? Because I like making stuff work.
I find it hard to understand why people torture themselves with C/C++ when there are more productive languages available. I mean, wasting a day tracking down a dangling pointer? Life's too short!
I'd still happily turn to assembler if I really needed performance, but my crappy old PC and my uncrappy, not-so-old Mac (under OS X) seem to manage everything I want.
If I need to confuse myself with intractable problems, I'll contemplate my life.
FWIW, The Hot Rock was based on a novel by Donald E Westlake, who was nominated for an Academy Award for his screenplay of The Grifters. I read a number of his caper novels (including The Hot Rock) years ago, and they were hilarious.
Granted, that is partially tongue-in-cheeck, but it does raise the question of how do you choose a pair?
You don't. Today, Charly's looking over your shoulder; tomorrow, you'll be looking over Oliver's. That way, knowledge of the whole project is shared among the team. Oh, and everybody gets to have a go on the keys;-) And in case you're wondering how somebody is going to pick up and run with the code you wrote yesterday, that's why you code the unit tests first. They help to define what the code should do.
I have to disagree. JavaScript is a pretty neat language; although it is most commonly encountered in a web browser, it would be important to ensure that a beginner understood the separation between the language and the environment in which it was hosted. I feel that being made aware of the distinction between the DOM and the core language would help take a beginner a long way down the road to understanding the principles of OOSD.
Although it is true that JS is a prototype-based language rather than a true OOL, I always use it in an object-oriented manner(check out the description of implementing encapsulation and inheritance in Netscape's JS docs - which makes no mention of anything to do with browsers;-). The speed of development and ease of testing make it my language of choice for all kinds of prototyping, even if I'm ultimately aiming at (say) a Java implementation.
In case you're wondering, I've been coding for 25 years in everything from about 10 different flavours of assembler (anybody else remember loading bootstraps into a PDP8/e using the toggle switches on the front?) to Java via BASIC and Forth, and I wouldn't recommend BASIC to anybody expecting to acquire programming skills which will be of value in TRW. I know VB is going to finally get some kind of bastard object-orientation, but a kludge on a hack isn't a good example to set anyone.
Tell them that their concerns will be addressed in the next Service Pack for the discussion, whilch will be available 2nd quarter next year. Until then it is suggested that they disable ActiveSlashdot in their browser. Oh, and reinstall CommonSense while they're about it.
Did the email have an attachment with a.vbs extension by any chance?
I just went over to www.microsoft.com - not a word about this on the home page. So I did a search on "iloveyou" - it returned 0 hits. So I went into the MS Office area of the site, just to be sure, and got exactly the same result. I just love to see MS facing up to its responsibilities. Still, at least I was able to find out about their commitment to innovation.
Slashdot Mirror
It's impossible to do that. Turing demonstrated that it is not possible to determine whether any given algorithm will execute to completion for all possible inputs. As the library in question is a mathematical one, it will undoubtedly contain algorithms which will not complete for some input or inputs, and all the bounds-checking in the world cannot guarantee security from input which will cause an infinite execution time. If it was possible, it would be a solution to the Turing machine halting problem, and such a thing cannot be, by definition.
You have to watch out for the two checkboxes right at the start of the claim form. If you only tick "Contributions-based Jobseekers Allowance", you can miss out that way (although it shouldn't, IIRC, be based on 4 year old contributions). You must tick "Income Based JSA" as well, and fill in the 15 or so extra pages of means testing questions. If you've got more than £3000 savings, you'll lose money; if you've got more than £8000, you're not entitled at all. You have to sign an agreement stating the kinds of work you'll be looking for and how many job applications per week (at least two, though a graduate would be expected to do more), and be able to prove when they ask that you have done what the agreement states (they write the agreement). After 6 months, you're off contributions-based allowances no matter what.
If you aren't entitled to JSA, you should be able to get Income Support at a lower rate. IIRC, JSA at the moment is just over £50/week, Income Support rather less; those figures are also reduced to the ~£43/week if you're under about 25, as you're apparently supposed to sponge off your parents, whether you have any or not. Those figures are for singletons; if you have a partner, you'll be jointly means tested, but if you're both on the dole, you won't get twice as much.
Even when you get those sorted out, you have to deal with Housing Benefit to cover the cost of a roof over your head; and that itself won't cover the full cost, so you'll need about a quarter of the JSA/IS to make up your rent. If you've got a mortgage, start collecting the cardboard boxes; you get sod-all help, so you'll be living in them soon.
Still, if you think it's bad now, you should have tried it around 1990. Graduates had to sign that they were willing to work anywhere in the country; if you lived in Cornwall and they came up with a job in Aberdeen, you were supposed to move to get the job, just because you had a degree. Then that mad witch Thatcher got booted out, and a very slight amount of sanity began to creep back into the country.
Check out the Department of Work and Pensions for more info. Also, see if your local council has a welfare rights department (tip: if they're Tory, they probably won't).
HTH
No, they're trying to cut down on the dupes.
Don't you mean
Sorry, I'm feeling crass today ;-)
Actually, I think they are just able to buy international call access at bulk rates. There's a lot of spare capacity on the international phone network, and buying it in bulk makes this sort of thing cost effective.
It's like those shops full of phones offering cheap calls abroad - the UK city I live in has about 30% population of Indian/Pakistani/Bangladeshi descent, and in certain areas there are loads of these places. They are reselling phone time that has been bought in bulk from someone like BT. Same thing for the cards that give you so many minutes of cheap international calls from your home phone by dialling a special prefix code.
AFAIK, UK businesses have to respect your registration with the Telephone Preference Service, even if the calls are originated in an offshore call centre.
Slashdot will be closed this Thursday, as it is necessary to shut it down and clean it once a year.
And if you really want to have fun, you can open his net statistics in a separate window/tab, reload the original page, then reload the statistics to see if you're in the "last 10 visitors"! Took me three goes to get there in time at all, and I was already down to number 8 :-)
Well, it's a holiday, so I really do have too much free time. Anyway, there's someone at Boeing who isn't doing any work, either.
If He wanted to pay the ultimate sacriice, He should have stayed down here with us.
Or the Office Assho^H^Histant...
It looks like you're a dirty stinking software pirate. Do you want some help erasing your hard drive?
If you check out the Reg story, you'll see that the bug causes those corporate "no activation" versions to demand activation. That is the fundamental problem.
My Ericsson A2618s has such a bad user interface, I've thought of writing a paper pointing out its flaws from the perspective of HCI.
One of its amazing features is that the longer a text message gets, the longer it takes for it to respond to letter selection. (No predictive text, thank God.) I've easily managed to get 10 to 12 characters ahead of it. Presumable every time it adds a character to a string, it copies the whole string to a buffer one character longer and then appends the new char...
That's a performance issue, but what about the way it unlock the keypad when it receives a text? Knocking the "yes" key brings up the message, knocking it again brings up the subsequent menu, and knocking it a third time selects the first option of the subsequent menu - which is "Delete".
Several times I've received texts which had been deleted by the time I got the phone out of my pocket.
I keep meaning to write a Java app on my Mac and then get it running on my PC, just to see what the difficulties are these days. In theory, the write-once-run-anywhere stuff is true these days, but I'm going to suck it and see before I make any claims in that department ;-)
Afraid not; the last time I mixed assembler with high level was in Forth, which probably isn't much help when it comes to gcc.
However, you're spot on with the idea of coding in high level first, then converting to assembler. Specifically, I'd suggest getting an app working, then if you need extra performance, profile it and recode the bits that are bogging you down. That's the way we used to work in Forth, and it offers several advantages: first, you know that your logic is correct; second, if you find a bug in your logic, you can go back, switch in the high level stuff, fix the logic in there, then convert your changes back into the assembler code; third, if your unit tests (you did write your unit tests, right?) pass with the high-level and not the assembler, you know that the mistake is in your conversion, not your logic.
Cautionary tale:
Back in the early 80s, a friend at a company where I used to work had a bug in his assembler conversion of a Forth word (== subroutine). He spent hours trying to work out how he'd gone wrong - the high level code worked perfectly, the low level failed every time. Eventually he called in the project manager, a phenomenal coder who should never have been put onto managment. This guy sat there testing the whole thing from basics (never believe what the guy who wrote it says about it working ;-), and finally brought a hex dump of the machine code up on the screen. He scanned through it, and suddenly pointed at a single byte: "It's there!" He switched back to the source, pointed at the relevant mnemonic, and stated "That's correct: it's a bug in the assembler." And it was...
Among other morals of this tale, people should note one thing: if a programmer is phenomenally good, don't make him a manager; just pay him more. None of the programmers there would have resented him being paid more, because they all recognised his superiority. Trouble is, management judge others by status, and coders judge by ability. He was the worst manager I've ever come across, but I learnt more about finding bugs from one 45 minute debugging session with him than from 10 years of programming.
Aw, shucks. <simper />
Good luck, and (most important) have fun!
I've worked at places like this. Funnily enough, they always seem to be managed by people who used to write code, and owned by people who know how to listen to their management. It's so simple, you'd think the rest of the industry would have cottoned on by now. Then again, our species' capacity for exalting mediocrity never ceases to astonish me.
I'm not surprised you're happy in your work. I hope it keeps up for you :-)
I can't argue with you, and don't want to. I think, though, that you're talking about writing software you will use; I'm sadly trapped in the hideous world of doing stuff to pay the rent. (Not that there is any stuff to do the moment, but we all have to tighten our belts in wartime ;-)
Love the language that works for you, but as the song says, "If you can't be with the one you love, love the one you're with" :-)
I have to say that if there's one language I really like to work in, it's Forth. I've written my own Forth language implementation and development environment before now (for the Atari ST) ; ultimately, when I'm working for my own pleasure, I want the language I like, and the devil take the hindmost. (I haven't had the chance to get paid for it since 1989, but I still solve Countdown's Numbers game on a stack in my head.)
Unfortunately, those people who employ - or fail to employ - consultants like me want stuff to work on their systems, and so I sell my soul. How much work does a Forth specialist get these days, as a regular gig?
My PC, brushing against my leg as I type, started life as an empty case that I put my dad's old 286 motherboard in; believe me, I know about making do with what you can afford. (FWIW, it's currently a 500MHz AMD I got as the cheapest available from the local computer shop when the second-hand AMD 100MHz went west. Still got a 4Gb drive in it.) I'm thinking about philosophical differences between approaches.
The reason we regarded people who programmed in C with a degree of pity (this was 1987 - 88) was because setting up and breaking down all those stack frames used up valuable cycles. On an 8086, a
was less efficient than a
or (my opcode of choice) a
by a couple of cycles. That meant everything when a single byte read from/write to a CGA card's video RAM cost betwen ~20 to ~50 wait states for an 8MHz 8086. That's the level we were working to when I developed my dislike of C; my dislike of later manifestations of C and C++ comes down to reading those ads in Doctor Dobbs Journal saying "Can you tell where the bug is? PC-Lint can!". I should be able to tell where the bug is; I wrote it, for goodness sake!
I just feel that the point in the original article about us programmers being more attracted to cool, tricky stuff than to meeting real-world requirements was right generally, as well as in the context of application security. I have to combat my desire to do cool stuff every day, because I have to do stuff that satisfies my customers' desire for functional, safe stuff. It's boring, but it keeps me off the streets - or, at the moment, fails to.
BTW, I don't actively recommend or discommend Perl; I think Perl is just bizarre. If that's how Larry Wall thinks about programming (see the Camel Book intro), I'm just glad there's room for a lot of opinions in this world :-)
As an aside, can I ask, do you optimize your stuff by dropping into assembler for critical portions of your code? I'd highly recommend it, because in my experience, anybody who appreciates the power available from C or C++ really gets off on assembler. (Sorry if that sounds patronising; don't want to teach my grandmother to suck eggs.)
I don't have anything against any language, and I certainly appreciate your situation as regards processing power
Things like memory allocation problems, dangling references, etc. It's possible to do all this stuff yourself, but why bother when tools and libraries are available to do it for you? This was the point made in the article, and I agree.
You're right that one should choose the correct tool for the job; I just believe that a heck of a lot of stuff is done in C or C++ that could be done with higher-level tools. You doubtless remember that COBOL was one of the most widely used languages for business applications for years; I get the impression that a lot of people nowadays are using C/C++ for applications that would, frankly, be better written in COBOL. (And if there's one language I don't want to deal with, it's COBOL).
Charles Moore, inventor of Forth, is quoted in Leo Brodie's book Thinking Forth:
C was developed primarily as a tool for writing Unix. C++ was developed as a way of adding this cool, new (back then) OO approach to C. There's a place for both languages, but it would take a lot to convince me that either of them is the only appropriate choice for the majority of purposes to which they are actually put.
Programmers like stuff that's difficult because if you don't, you wouldn't be a programmer. But I prefer the language to stay out of the way of my exploration of my ideas as much as possible, and buggering about with malloc() is simply not relevant to the majority of problems. It's a distraction. I can make serious errors in the logic of my apps without having to be held up by null pointer exceptions that wouldn't happen with another language doing the system-level stuff for me. In the same way, if I could afford it I'd have a cleaner in; I want to live in the house, not be a housekeeper.
All IMHO, of course :-) And the next time I want to write an OS, I'll certainly use a tool like C.
The many posts here by people saying "I gotta use C/C++, that other stuff's too slow" prove exactly this point.
In the 80s, I worked for a number of years as an assembly language programmer (6502, 6809, Z80, x86, 680x0, ARM); we didn't use C (in fact, we sneered at people who did) because it just couldn't offer the performance of hand-crafted assembler (and I'm talking 30,000+ lines here).
Now I do almost all of my design prototyping in JavaScript (quick turnaround on the modify/compile/test cycle), and code real stuff in Java or (occasionally) C#. Why? Because I like making stuff work.
I find it hard to understand why people torture themselves with C/C++ when there are more productive languages available. I mean, wasting a day tracking down a dangling pointer? Life's too short!
I'd still happily turn to assembler if I really needed performance, but my crappy old PC and my uncrappy, not-so-old Mac (under OS X) seem to manage everything I want.
If I need to confuse myself with intractable problems, I'll contemplate my life.
FWIW, The Hot Rock was based on a novel by Donald E Westlake, who was nominated for an Academy Award for his screenplay of The Grifters . I read a number of his caper novels (including The Hot Rock) years ago, and they were hilarious.
Hopefully not too off-topic...
problem is, I swear I take a karma beating for wearing that on /. ; every linux user who sees it just goes ape shit on me :)
Probably just six particular Linux users ;-)
Leo Brodie mentions that Thinking Forth is back in print, and he may be revising Starting Forth.
I wouldn't fret too much. The lights won't be going out all over the universe until next Tuesday at the earliest ;-)
You don't. Today, Charly's looking over your shoulder; tomorrow, you'll be looking over Oliver's. That way, knowledge of the whole project is shared among the team. Oh, and everybody gets to have a go on the keys ;-) And in case you're wondering how somebody is going to pick up and run with the code you wrote yesterday, that's why you code the unit tests first. They help to define what the code should do.
I have to disagree. JavaScript is a pretty neat language; although it is most commonly encountered in a web browser, it would be important to ensure that a beginner understood the separation between the language and the environment in which it was hosted. I feel that being made aware of the distinction between the DOM and the core language would help take a beginner a long way down the road to understanding the principles of OOSD.
Although it is true that JS is a prototype-based language rather than a true OOL, I always use it in an object-oriented manner(check out the description of implementing encapsulation and inheritance in Netscape's JS docs - which makes no mention of anything to do with browsers ;-). The speed of development and ease of testing make it my language of choice for all kinds of prototyping, even if I'm ultimately aiming at (say) a Java implementation.
In case you're wondering, I've been coding for 25 years in everything from about 10 different flavours of assembler (anybody else remember loading bootstraps into a PDP8/e using the toggle switches on the front?) to Java via BASIC and Forth, and I wouldn't recommend BASIC to anybody expecting to acquire programming skills which will be of value in TRW. I know VB is going to finally get some kind of bastard object-orientation, but a kludge on a hack isn't a good example to set anyone.
All this IMHO, of course.
<Script Language=JavaScript>
document.write("Best viewed with ");
if (navigator.appName == "Netscape") { } else { }
</Script>
Even works for Opera
Did the email have an attachment with a .vbs extension by any chance?
I just went over to www.microsoft.com - not a word about this on the home page. So I did a search on "iloveyou" - it returned 0 hits. So I went into the MS Office area of the site, just to be sure, and got exactly the same result. I just love to see MS facing up to its responsibilities. Still, at least I was able to find out about their commitment to innovation.