However, you could implement this in a library in C++ using compile-time templates, implying very little run-time overhead (if any) for stuff that can be computed directly in hardware.
Proving that those who don't use Common Lisp are doomed to reimplement it...
If the OS scheduler only know about one core, how in the world would it ever know to set two threads in the execute state simultaniously to take advantage of the extra horsepower
It won't and that doesn't matter. When will you SMP-fetishist's learn that two simultaneous threads won't be running each in their own CPU? If you have two threads (or processes) running, that doesn't mean that each gets its own CPU; they'll share the 2 cpu's along with the dozens of other running processes. If AMD is right that they can get the scheduling overhead lower with this than you can get the scheduling overhead for SMP, then you'll see performance improvements.
Contact info in an easily accessible location?
on
Is It Time For .tel?
·
· Score: 3, Funny
the proposed.tel TLD would allow individuals and companies to keep all of their contact information in an easily accessible location.
If someone's telling you that they used visudo to edit the file, then saved, then tried to run sudo and it failed because of an error it/etc/sudoers, they're either fibbing or you should report a bug.
Umm... no. There's nothing invalid about taking the account you can log in as out of the wheel group (OK, that would be gpasswd, which is a 3rd one never to use sudo for), or removing sudo privileges from your own account, or removing sudo privileges from the group you're in. Or removing sudo privileges for one of the commands your account needs to edit passwd or sudoers. Note that OpenBSD effectively forbids running visudo through sudo and requires an interactive login, since/etc/sudoers' permissions must be 600 for sudo to run. This in my opinion is the Right Thing.
I can think of two exceptions: never ever run vipw or visudo through sudo. Always run it through su, log in separately and test that the change you made does what you want it to, then log out of your su shell. I can't tell you how many thousands of dollars I've charged my hosting clients to undo a passwd or sudoers change they made using sudo and screwed up.
Incidentally, rdesktop, a Linux RDP client, offers the ability to attach to the root console of a 2003 server, which is very handy if 3 of your colleagues have inconsiderately left RDP sessions running, locking you out of the box. With rdesktop -0 foo.bar.com, you can attach to the root console, log them out, and then get a "normal" RDP session. I've searched high and low for a similar feature in the MS client, but I can't find it. So, I take my Linux laptop to work for the sole purpose of being able to log in to my Windows servers. Go figure.
What's missing is that employees are an organization's best assets.
Sigh...
An employee is not an asset. He is neither owned nor acquired. He neither depreciates nor amortizes. An employee is *PART OF THE COMPANY*, and is investing labor just like the shareholder is investing money.
I don't hear people talking about shareholders as a company's "asset", though the case is probably a little stronger than for employees: money is money, and so shareholders are largely replaceable. An employee can't be replaced as easily.
Isn't this exactly why domains have multiple MX records?
Come to think of it, if you look at the A records for www.yahoo.com (actually yahoo.aka-dns.com) I'd bet a good number of those hosts are down at any given moment too.
We have redundant systems so that a given host being down is not crippling. We have multiple responses to DNS queries so that we can make use of those redundancies.
I know I was never an ape. My distant relatives are a different story...
Carlous Linneaus, a creationist (by default), defined humans as apes long before Darwin was born. An ape is a primate with no tail and certain other diagnostic characteristics.
They had no real proof of torture or any other wrong doing (except holding terrorists without trials)
Ah, you're close. What you meant is "except holding people they called terrorists without trials". They've released some of those people after 3 years (some of them, incidentally, British subjects), saying "oops, you really weren't an enemy combatant after all... somebody just turned you in to get reward money."
You'd think they could have figured that out in a few days, at most a week. But years?
The government says every person in Gtmo is an enemy combatant or a terrorist (those, incidentally, are two disjoint sets). Except for the few dozen they have released after a few years. Do you trust the government? If so, why not just do away with due process for all of us? If we can trust the government then the Constitution is an outdated, byzantine document.
Just like your MBA will point you to the definitions of P&L, EBITDA, Gross Profit, etc.
And I've got very high regard for MBAs who stick to things they know like P&L, EBITDA, Gross profit (and incidentally, I at least know a naive definition of all those; I don't expect a business person to have to baby step everything for me). It's the ones who think that that knowledge translates into all the knowledge neccessary to make every decision about a business that bug me.
How many copies of a one gigabyte file fit in two gigbytes of RAM, how many fit on a forty gigbyte hard drive, how many bits are in each of those bytes and how many seconds would the file take to send over a 1-kilobit-per-second link ?
In order: not quite 2, 39 in an ideal filesystem (rather less in most implemented ones), 8 on all modern architectures, 1,000,000 seconds plus carrier protocol overhead. "kilobyte" and "byte" do have different meanings in different contexts, but that does not imply ambiguity; the meaning in each context is defined. (Though "one gigabyte file" is not clearly defined; you could mean a file with a disk image of 1 gigabyte, or a file with a memory image of 1 gigabyte. Since gigabyte means something different in memory than it does in storage, that would need some clarification. The convention of capitalizing G,M, K etc. to mean powers of 2 and lowercasing g, m, k to mean powers of 10 is helpful, but not standard.)
Can techies even get "unambiguous" right in the first place ?
[snip of RE definition of "unambiguous"]
Now, see, I think that's actually a very good example. The meaning of "unambiguous" in the context of that standard needed to be clear enough that all implementations could be verifiably compliant. I don't see anything wrong with specifying what "unambiguous" means in that context.
What if the roles were reversed? Suppose the poster is a business major who has been thrust into the IT/S division of his company, asking us business folk if he should have to learn these ridiculous technical terms in order to communicate with the people he has to deal with every day. Your advice in that situation translates to: Hell no! Fight those socially inept geeks who try to confuse the real issue by loading up on technical terms and all that garbage.
Rubbish.
If someone doesn't know what TCP/IP means or what a CNAME record is, I can direct him to appropriate RFCs that define them.
Now, I wouldn't actually direct an MBA to an RFC, because his eyes would glaze over about the time he got to "this memo has unlimited distribution." But what matters is that I can direct him to such a document, because such a document exists. Tech-speak is done with well-defined terms that have standardized meaning, and it is used to clarify how we talk to each other.
If you can point me to a document or documents standardizing terms like "Web 2.0", "enterprise", "solution", "mission-critical", "partner", etc., then I will admit my criticism of corporate speak is wrong. However, I don't think you will be able to, because those documents don't exist. Because these words' meanings are not standardized. They mean to the speaker what he imagines he means, and they mean to the listener what he imagines he hears. That, I think, is what business types don't understand when they compare themselves to techs: what we say means something, because we had to learn something objective, verifiable, and repeatable to get where we are, while they didn't.
Eh... honestly in all my work with databases, I've wished for a lot of things but never "I wish this database were faster". I've always wished it had more and better referential integrity options, more inheritance features, better filesystem interfaces with more frequent writes, but the problem has never been "this database isn't doing queries fast enough". There have always been some queries that were slow, but they have been from dbs that are poorly designed -- a poorly designed query into a poorly designed database will probably still run slow on a "faster" database.
You must've studied your ass off for the RoR test, and, it seems to me, if you aced the company's test, then by definition you're an excellent candidate to work there, whether you had really implemented your CRM system in their language or not.
umm... have you used RoR? Hardly takes much studying to learn it.
Just don't try that with something like C++, which contains many pitfalls that newbies don't know about at first, or Java, which has a gianormous class library.
*shrug* I got a Java job Back In The Day on the same pretenses (I had done something in C++, but I said I had done it in Java to get the interview. The rest is history). Java libraries are not that complex. Google for them, read the readme, and compile with them. Not very hard.
Personally I don't understand why people make such a big deal over this or that language. The language is not the issue. Understanding the underlyind algorithm is. If you can do a quicksort in C, you can do it in whatever.
In fairness, the Greek was pronounced something closer to "oorenos" (with a long "o" at the end). The problem comes with putting a "y" before the "ou" vowel; that's an artifact of Latin.
The result of this is that everyone charges the exact same price, which coincidentally is exactly what it costs them to produce the good in the first place.
Close... it's actually sold at the marginal price, which is the difference in price between producing X units and X+1 units. That is related to the cost of producing the good, but it's not the same.
All HR knows is they want someone with x years of Java and a dozen other skills. You can't just say "I learn languages quickly!" The attitude you get in response from the jaded cynical HR people is "stop giving us b.s., anyone can say that and nearly everyone does."
*shrug* you're clearly less cynical than I. I just look for what each job is looking for and alter my resume to make it seem like I had been using that. My current job asked for experience with Ruby on Rails. So, I changed my resume from saying "Implemented CRM package in Common Lisp" to "Implemented CRM package in Ruby on Rails" and studied up before the interview. The interview included a skills test in Ruby, which I aced, and when they asked to see source code for the CRM system I said, "oh, sorry, there was an NDA on it and I can't distribute, but I'll be happy to write something for you right now."
I got the job; they are none the wiser that I had never even looked at Ruby until 2 weeks before the interview.
HR people are stupid and dishonest (and recruiters are even worse). There's no reason to be honest with them. If you have the skills, lie. It's that simple. Otherwise you are letting jackasses with the right buzzwords win.
Since when did running Unix decide your processor type for you? Last I checked, BSD ran on X86 without much issue. Last I checked, Linux wasn't a flavour of Unix.
The point is, vendors classify servers based on their chip type. The assumption seems to be that a CISC system will run some flavor of WINNT while a RISC system will run some flavor of UNIX (largely because there isn't any other choice for most of them). I could see calling a RISC system UNIX based on the fact that a large percentage of Intel systems are not running UNIX, while virtually every RISC system is.
And, yes, Linux is not a flavour of UNIX, just like the toy I had my dog fetch this afternoon was really a "flying disc" rather than a "frisbee (tm)", since it wasn't made by Whammo (tm).
It's thought by some that we still haven't brought geometry to the state it was in before the Dark Ages.
Only by those who don't read Hebrew, Arabic, and Farsi (which, sadly, includes almost every scholar of Hellenic math and sciences). I remember once in a Hellenic conics class I was taking the prof. said "unfortunately, book III of whatever has been lost", which surprised me because we had just been studying the Arabic version of book III of whatever that week in my medieval Arabic class.
Incidentally, I agree about the Chinese proof of Pythagorus: it's much clearer than Euclid's I.47 (if it's the one I'm thinking of with the inscribed circle). Still, for clarity and elegance in a proof it's almost impossible to beat Euclid's VII.3 for the GCD (well... once it's edited to make the recursion explicit rather than assumed) or (my personal favorite) Euclid's IX.20 for the infinitude of primes.
*shrug* there's this assumption that people before the Enlightenment were stupid, but 'tain't so. Galen performed brain surgery. So did ancient Egyptians. People are very very good at figuring out the medicinal properties of herbs, so it wouldn't surprise me at all if there were some moderately effective anaesthesia they used.
Slashdot Mirror
Proving that those who don't use Common Lisp are doomed to reimplement it...
It won't and that doesn't matter. When will you SMP-fetishist's learn that two simultaneous threads won't be running each in their own CPU? If you have two threads (or processes) running, that doesn't mean that each gets its own CPU; they'll share the 2 cpu's along with the dozens of other running processes. If AMD is right that they can get the scheduling overhead lower with this than you can get the scheduling overhead for SMP, then you'll see performance improvements.
Wow! If only someone had thought of that before!
Umm... no. There's nothing invalid about taking the account you can log in as out of the wheel group (OK, that would be gpasswd, which is a 3rd one never to use sudo for), or removing sudo privileges from your own account, or removing sudo privileges from the group you're in. Or removing sudo privileges for one of the commands your account needs to edit passwd or sudoers. Note that OpenBSD effectively forbids running visudo through sudo and requires an interactive login, since /etc/sudoers' permissions must be 600 for sudo to run. This in my opinion is the Right Thing.
I can think of two exceptions: never ever run vipw or visudo through sudo. Always run it through su, log in separately and test that the change you made does what you want it to, then log out of your su shell. I can't tell you how many thousands of dollars I've charged my hosting clients to undo a passwd or sudoers change they made using sudo and screwed up.
Incidentally, rdesktop, a Linux RDP client, offers the ability to attach to the root console of a 2003 server, which is very handy if 3 of your colleagues have inconsiderately left RDP sessions running, locking you out of the box. With rdesktop -0 foo.bar.com, you can attach to the root console, log them out, and then get a "normal" RDP session. I've searched high and low for a similar feature in the MS client, but I can't find it. So, I take my Linux laptop to work for the sole purpose of being able to log in to my Windows servers. Go figure.
Sigh...
An employee is not an asset. He is neither owned nor acquired. He neither depreciates nor amortizes. An employee is *PART OF THE COMPANY*, and is investing labor just like the shareholder is investing money.
I don't hear people talking about shareholders as a company's "asset", though the case is probably a little stronger than for employees: money is money, and so shareholders are largely replaceable. An employee can't be replaced as easily.
Isn't this exactly why domains have multiple MX records?
Come to think of it, if you look at the A records for www.yahoo.com (actually yahoo.aka-dns.com) I'd bet a good number of those hosts are down at any given moment too.
We have redundant systems so that a given host being down is not crippling. We have multiple responses to DNS queries so that we can make use of those redundancies.
Carlous Linneaus, a creationist (by default), defined humans as apes long before Darwin was born. An ape is a primate with no tail and certain other diagnostic characteristics.
Ah, you're close. What you meant is "except holding people they called terrorists without trials". They've released some of those people after 3 years (some of them, incidentally, British subjects), saying "oops, you really weren't an enemy combatant after all... somebody just turned you in to get reward money."
You'd think they could have figured that out in a few days, at most a week. But years?
The government says every person in Gtmo is an enemy combatant or a terrorist (those, incidentally, are two disjoint sets). Except for the few dozen they have released after a few years. Do you trust the government? If so, why not just do away with due process for all of us? If we can trust the government then the Constitution is an outdated, byzantine document.
And I've got very high regard for MBAs who stick to things they know like P&L, EBITDA, Gross profit (and incidentally, I at least know a naive definition of all those; I don't expect a business person to have to baby step everything for me). It's the ones who think that that knowledge translates into all the knowledge neccessary to make every decision about a business that bug me.
In order: not quite 2, 39 in an ideal filesystem (rather less in most implemented ones), 8 on all modern architectures, 1,000,000 seconds plus carrier protocol overhead. "kilobyte" and "byte" do have different meanings in different contexts, but that does not imply ambiguity; the meaning in each context is defined. (Though "one gigabyte file" is not clearly defined; you could mean a file with a disk image of 1 gigabyte, or a file with a memory image of 1 gigabyte. Since gigabyte means something different in memory than it does in storage, that would need some clarification. The convention of capitalizing G,M, K etc. to mean powers of 2 and lowercasing g, m, k to mean powers of 10 is helpful, but not standard.)
Now, see, I think that's actually a very good example. The meaning of "unambiguous" in the context of that standard needed to be clear enough that all implementations could be verifiably compliant. I don't see anything wrong with specifying what "unambiguous" means in that context.
Rubbish.
If someone doesn't know what TCP/IP means or what a CNAME record is, I can direct him to appropriate RFCs that define them.
Now, I wouldn't actually direct an MBA to an RFC, because his eyes would glaze over about the time he got to "this memo has unlimited distribution." But what matters is that I can direct him to such a document, because such a document exists. Tech-speak is done with well-defined terms that have standardized meaning, and it is used to clarify how we talk to each other.
If you can point me to a document or documents standardizing terms like "Web 2.0", "enterprise", "solution", "mission-critical", "partner", etc., then I will admit my criticism of corporate speak is wrong. However, I don't think you will be able to, because those documents don't exist. Because these words' meanings are not standardized. They mean to the speaker what he imagines he means, and they mean to the listener what he imagines he hears. That, I think, is what business types don't understand when they compare themselves to techs: what we say means something, because we had to learn something objective, verifiable, and repeatable to get where we are, while they didn't.
... but I'm not.
wget (while they're waiting in the airport).
Eh... honestly in all my work with databases, I've wished for a lot of things but never "I wish this database were faster". I've always wished it had more and better referential integrity options, more inheritance features, better filesystem interfaces with more frequent writes, but the problem has never been "this database isn't doing queries fast enough". There have always been some queries that were slow, but they have been from dbs that are poorly designed -- a poorly designed query into a poorly designed database will probably still run slow on a "faster" database.
umm... have you used RoR? Hardly takes much studying to learn it.
*shrug* I got a Java job Back In The Day on the same pretenses (I had done something in C++, but I said I had done it in Java to get the interview. The rest is history). Java libraries are not that complex. Google for them, read the readme, and compile with them. Not very hard.
Personally I don't understand why people make such a big deal over this or that language. The language is not the issue. Understanding the underlyind algorithm is. If you can do a quicksort in C, you can do it in whatever.
Long live digg? I think you mean long live reddit, bitch!
In fairness, the Greek was pronounced something closer to "oorenos" (with a long "o" at the end). The problem comes with putting a "y" before the "ou" vowel; that's an artifact of Latin.
Close... it's actually sold at the marginal price, which is the difference in price between producing X units and X+1 units. That is related to the cost of producing the good, but it's not the same.
*shrug* you're clearly less cynical than I. I just look for what each job is looking for and alter my resume to make it seem like I had been using that. My current job asked for experience with Ruby on Rails. So, I changed my resume from saying "Implemented CRM package in Common Lisp" to "Implemented CRM package in Ruby on Rails" and studied up before the interview. The interview included a skills test in Ruby, which I aced, and when they asked to see source code for the CRM system I said, "oh, sorry, there was an NDA on it and I can't distribute, but I'll be happy to write something for you right now."
I got the job; they are none the wiser that I had never even looked at Ruby until 2 weeks before the interview.
HR people are stupid and dishonest (and recruiters are even worse). There's no reason to be honest with them. If you have the skills, lie. It's that simple. Otherwise you are letting jackasses with the right buzzwords win.
The point is, vendors classify servers based on their chip type. The assumption seems to be that a CISC system will run some flavor of WINNT while a RISC system will run some flavor of UNIX (largely because there isn't any other choice for most of them). I could see calling a RISC system UNIX based on the fact that a large percentage of Intel systems are not running UNIX, while virtually every RISC system is.
And, yes, Linux is not a flavour of UNIX, just like the toy I had my dog fetch this afternoon was really a "flying disc" rather than a "frisbee (tm)", since it wasn't made by Whammo (tm).
Only by those who don't read Hebrew, Arabic, and Farsi (which, sadly, includes almost every scholar of Hellenic math and sciences). I remember once in a Hellenic conics class I was taking the prof. said "unfortunately, book III of whatever has been lost", which surprised me because we had just been studying the Arabic version of book III of whatever that week in my medieval Arabic class.
Incidentally, I agree about the Chinese proof of Pythagorus: it's much clearer than Euclid's I.47 (if it's the one I'm thinking of with the inscribed circle). Still, for clarity and elegance in a proof it's almost impossible to beat Euclid's VII.3 for the GCD (well... once it's edited to make the recursion explicit rather than assumed) or (my personal favorite) Euclid's IX.20 for the infinitude of primes.
*shrug* there's this assumption that people before the Enlightenment were stupid, but 'tain't so. Galen performed brain surgery. So did ancient Egyptians. People are very very good at figuring out the medicinal properties of herbs, so it wouldn't surprise me at all if there were some moderately effective anaesthesia they used.
I mentioned in the post that we are assuming for simplicity's sake that only the top centimeter of water is affected.
Of course it came from the Sun! Where else would it come from?
The point remains that it's here now whereas previously it was being reflected back into space.