On an investor level, your comment scares me a bit.
Solaris is not really the only thing that makes Sun go. Sun's huge benefit to me, as a system administrator, is that the SPARC-based servers (not desktops) are very solid. Solaris is also quite solid, but not especially fast.
I'd choose Solaris/SPARC over Linux/x86 for anything that needs stability over speed.
What about client programs that cache DNS lookups (I think some web browsers do this)?
Many web browsers do, nscd does, DNS caches do...
Speaking of DNS caches, think about the case when an ISP is providing DNS for their customers - even cycling once per minute isn't good for load-balancing the hits routed via a large DNS cache. Further, when I used to run DNS for a large ISP, I set a minimum timeout for data, because I explicitly did NOT want my caches pulling zone data once per minute. (I set it to five minutes, I think, but I'm not sure any more, and that was long ago, in a land far away, and besides, the wench is dead.)
This is a clever idea, and neatly works around the problem of multiple A records not being tried in the real world, but it does have problems caused by non-compliant DNS cacheing.
Has anyone given any thought as to how a programming language should be structured so that it would be easy to code by waving things with particular smells in front of the computer? If not, why not?
Seriously, this is not a design goal for programming languages. Programming languages are meant to a) be not ambiguous, and b) match or impose a way of thinking. Neither of these map particularly well onto spoken languages, because the bloody syntax (to keep it nonambiguous) gets in the way.
"We" being Canada? We certainly don't have a copy tax in the USA on all blank media.
Yes, we do. Cassettes, CDs, whatever can be recorded onto. It's just not charged seperately from the price on the shelf (unlike, for example, sales tax, which is assessed at checkout).
Re:Is NavTech Data OpenSource/Freely available?
on
Best Online Mapping Site?
·
· Score: 3, Informative
It is, although not preparsed or incredibly well formatted. All of the mapping systems (for the US) use the US Census TIGER/LINE data, which can be downloaded in original form. It's not the simplest of data to decode, and the reason that NavTech et al. charge for theirs is that they went through the trauma of fixing the more horribly broken bits, and realigning the data to be more usuable.
Don't do this. Just wget sitefinder directly if you want to impose load. There is collateral damage from looking up formerly nonexistant names imposed upon DNS caches, which have to remember each and every name now in.com, for the TTL of the wildcard record.
For the good of the DNS infrastructure, please just wget from sitefinder directly, if you intend to cause mayhem. I, of course, do not think people should do anything naughty, because naughtiness is mean. Yeah.
It sounds like the FFT DLL would be a seperate file. Why did they do this? If they want to obscure the code getting rid of symbols...
The FFT DLL was seperate, but distributed with the client (at least, it was, at the time, if I remember correctly). I no longer run Windows or SETI@home, so I dunno if this has changed. As to why it was a DLL? I have no idea.
I am also suprised that you were able to figure out what the arguments to the DLL functions were.
The person (who wasn't me) who figured out that the FFT DLL was the problem was a wizardly coder, and I have no clue how he did it, but he did - I ran the client with the new DLL and it was way faster.
Why they would not accept an improved Windows client is a mystery.
The only reason we could come up with for not taking our blocks was that they were mad that the client had been improved without their blessing. Like I said, they're welcome to keep their toys, but if the goal for their Windows client is that it run as terribly as possible, then heck, I'm not going to run any of their clients.
I actually happened to be working for them when this happened. Sun, for example, had gotten a source license to SETI@home, and had rewritten some of the more laughably terrible portions of the client to make it faster. They, of course, were using these results as unofficial "proof" that Solaris was faster than Windows.
One of the bright chaps at Microsoft noticed that the FFT DLL was dog slow. So he replaced it with a faster, equally accurate FFT library. A while later, the MS team got an email from SETI@home demanding the source code back. They were somewhat nonplussed when the MS SETI team told them that they didn't have any source.
Yes, the concern SETI stated was that the team was chewing through blocks too quickly and consuming all the data. But realistically, that isn't a huge problem. If the data gets processed, it gets processed. The goal is to determine one way or the other if there is anything interesting in a block, not just to use idle cycles from now until the end of time.
When we heard through the team that SETI had issued a cease and desist of using the client, we all more or less agreed that it was their toys and they could take them away if they wanted to, but we also more or less agreed that if SETI@home was going to demand that people run intentionally crippled clients for political benefit rather than scientific benefit, then the entire SETI goal was not really being pursued by the fine folks at SETI@home.
I don't run the client any more, although I think it would be neat if we found ETI. There's obviously no point.
If you're intending to flame me for something, please choose something more interesting that "You suck because you worked for Microsoft." Also: this was long ago, and in another country, and besides, the wench is dead.
The "Tragedy of the Commons" is that when there is common land, no one maintains it. Think of a communal kitchen in a geekhouse or something - typically one person does more of the cleaning than the others. At least, that's the case with people I know. But I digress.
The tragedy of the commons was not the land grab but the apparent disrepair that the commons were suffering from because no one was willing to maintain them at their own expense so that all could profit. "Why should I not overgraze so that my neighbor's sheep can be nice and plump? If I let my sheep eat all the grass on the commons (leaving none for anyone else), my sheep will be the plumpest and I'll make the most money!"
States in the US with a sales tax typically have a use tax at the same rate, which applies to goods imported across states lines for which less or no sales tax was paid. So if I buy something in a state with 3% sales tax, and live in a state with 5% sales tax, I must pay my home state enough money to total 5% tax paid. If I paid nothing to the other state, all 5% goes to my home state. If I paid 3% to the first, 2% goes home. If I paid 7% sales tax, I'm out 2%.
Practically speaking, no one deals with interstate sales tax or use tax if they can help it. If a company has no presence in a state with sales tax, it seems they can pretty much ignore that sales tax, since the state tax collectors cannot cross state lines to collect it (state's rights and all that), and although the states sometimes make a heck of a fuss about use tax, it is quite hard to figure out how and where to pay it, let alone bother paying.
There was a brilliant Asimov short story with the premise of world war. The computers in the guided missiles were too big and heavy, and they used so much energy to run, so the missiles were doing significantly less well than either side had hoped. Someone came up with the idea of teaching people trigonometry. They could guide the missiles themselves (presumably on kamikaze missions, although this was not made incredibly clear), and would be much more effective than the computers. [I guess you have to make some allowance for Asimov having written this when computers were obviously useful but not obviously following (the as-yet unstated) Moore's Law.]
Anyway, the names for these mathematical wonders, who could do sines and square roots in their heads? Yup, Asimov chose for the culture of his story to use the word "calculator."
It's not worth trying to figure it out. Yes, it's 11.5, and no, I didn't use a calculator. But 109/18? I'm pretty sure it's between 5 and 6. If I needed the answer, I'd ask xcalc. Why? Because I don't care enough to find the answer. If it is on a test, and I have to know the answer to complete the problem, I'll do subtraction by hand (carry the one, etc.), but usually I'll leave it as... * 109/18 (or whatever). I can do integrals in my head (usually) and think in many dimensions and all that good stuff, but I do have trouble with basic math. I don't think it's because I used a calculator too much so much as a lack of general need-to-know most of the time.
That said, there is another reason I avoid doing such problems: I've found I'm mildly dyslexic, and often do something like 190/18 or something equally wrong. Missing two points for using a totally bogus constant is worse than dealing with the (extremely rare) prof who won't accept 109/18 in an answer.
And yes, my degree is in Math. Take this as you will.
This is at least as relevant today as when Heinlein wrote it about 30 years ago.
"There has grown up in the minds of certain groups in this country the notion that because a man or a corporation has made a profit out of the public for a number of years, the government and the courts are charged with the duty of guaranteeing such profit in the future, even in the face of changing circumstances and contrary public interest. This strange doctrine is not supported by statute nor common law. Neither individuals nor corporations have any right to come into court and ask that the clock of
history be stopped, or turned back, for their private benefit. That is all."
- Robert A. Heinlein ("Life-Line")
It's coming. We just got the switches in at the house, so our build server is going to come on line and start spewing out packages. Files should be available on our web site as soon as there's anything to actually install.
I really appreciate your interest - it makes me feel that this is actually worthwhile!
Some friends of mine and I have been working on a system to netboot Linux. It turns out that it's not that difficult - a combination of TFTP for a kernel, NFS for a root filesystem, and some custom code to pull all of the stuff over. Think Solaris JumpStart meets Linux.
It's actually surprisingly clean. Right now, it actually works without requiring NFS, so after a little bit of preconfiguration on the server side, you can configure a shockingly large number of systems just by putting a floppy or CD in the drive and booting. (The NFS is planned, mostly for completeness, since the system at large can actually work without.)
The scope of ii is from declaration through to the close curly braces of the block in which it is defined. So
int foo() {
for (int ii = 0; ii < 4; ii++) {
...
} ...
for (int ii = 0; ii < 8; ii++) {
...
}
}
has a duplicate definition of ii, since the first declaration is in scope until the last curly brace I wrote. Any other compiler is behaving questionably if it allows it.
How many people did you recognize there? (C'mon, I know that's why you were looking at most of the pictures - that's why I looked at more than just the happy couple.) I saw Rob and crew (big surprise), but I really think that the joker taking a picture of the camera was Mandrake, of Enlightenment fame. There's this shot of someone I'd swear is Geoff Harrison.
And after checking www.mandrake.net it seems that he did go to the wedding, which implies that the person he's standing next to is Tammy...
So I guess this is the next poll: How many people did you recognize at Hemos' wedding?
Slashdot Mirror
On an investor level, your comment scares me a bit.
Solaris is not really the only thing that makes Sun go. Sun's huge benefit to me, as a system administrator, is that the SPARC-based servers (not desktops) are very solid. Solaris is also quite solid, but not especially fast.
I'd choose Solaris/SPARC over Linux/x86 for anything that needs stability over speed.
What about client programs that cache DNS lookups (I think some web browsers do this)?
Many web browsers do, nscd does, DNS caches do...
Speaking of DNS caches, think about the case when an ISP is providing DNS for their customers - even cycling once per minute isn't good for load-balancing the hits routed via a large DNS cache. Further, when I used to run DNS for a large ISP, I set a minimum timeout for data, because I explicitly did NOT want my caches pulling zone data once per minute. (I set it to five minutes, I think, but I'm not sure any more, and that was long ago, in a land far away, and besides, the wench is dead.)
This is a clever idea, and neatly works around the problem of multiple A records not being tried in the real world, but it does have problems caused by non-compliant DNS cacheing.
Has anyone given any thought as to how a programming language should be structured so that it would be easy to code by waving things with particular smells in front of the computer? If not, why not?
Seriously, this is not a design goal for programming languages. Programming languages are meant to a) be not ambiguous, and b) match or impose a way of thinking. Neither of these map particularly well onto spoken languages, because the bloody syntax (to keep it nonambiguous) gets in the way.
This is also why Edsger Dijkstra felt that natural programming was not a good idea.
So why not just add builtin event notification to snmp?
You mean SNMP traps?
Yes, we do. Cassettes, CDs, whatever can be recorded onto. It's just not charged seperately from the price on the shelf (unlike, for example, sales tax, which is assessed at checkout).
It is, although not preparsed or incredibly well formatted. All of the mapping systems (for the US) use the US Census TIGER/LINE data, which can be downloaded in original form. It's not the simplest of data to decode, and the reason that NavTech et al. charge for theirs is that they went through the trauma of fixing the more horribly broken bits, and realigning the data to be more usuable.
Don't do this. Just wget sitefinder directly if you want to impose load. There is collateral damage from looking up formerly nonexistant names imposed upon DNS caches, which have to remember each and every name now in .com, for the TTL of the wildcard record.
For the good of the DNS infrastructure, please just wget from sitefinder directly, if you intend to cause mayhem. I, of course, do not think people should do anything naughty, because naughtiness is mean. Yeah.
It sounds like the FFT DLL would be a seperate file. Why did they do this? If they want to obscure the code getting rid of symbols...
The FFT DLL was seperate, but distributed with the client (at least, it was, at the time, if I remember correctly). I no longer run Windows or SETI@home, so I dunno if this has changed. As to why it was a DLL? I have no idea.
I am also suprised that you were able to figure out what the arguments to the DLL functions were.
The person (who wasn't me) who figured out that the FFT DLL was the problem was a wizardly coder, and I have no clue how he did it, but he did - I ran the client with the new DLL and it was way faster.
Why they would not accept an improved Windows client is a mystery.
The only reason we could come up with for not taking our blocks was that they were mad that the client had been improved without their blessing. Like I said, they're welcome to keep their toys, but if the goal for their Windows client is that it run as terribly as possible, then heck, I'm not going to run any of their clients.
I actually happened to be working for them when this happened. Sun, for example, had gotten a source license to SETI@home, and had rewritten some of the more laughably terrible portions of the client to make it faster. They, of course, were using these results as unofficial "proof" that Solaris was faster than Windows.
One of the bright chaps at Microsoft noticed that the FFT DLL was dog slow. So he replaced it with a faster, equally accurate FFT library. A while later, the MS team got an email from SETI@home demanding the source code back. They were somewhat nonplussed when the MS SETI team told them that they didn't have any source.
Yes, the concern SETI stated was that the team was chewing through blocks too quickly and consuming all the data. But realistically, that isn't a huge problem. If the data gets processed, it gets processed. The goal is to determine one way or the other if there is anything interesting in a block, not just to use idle cycles from now until the end of time.
When we heard through the team that SETI had issued a cease and desist of using the client, we all more or less agreed that it was their toys and they could take them away if they wanted to, but we also more or less agreed that if SETI@home was going to demand that people run intentionally crippled clients for political benefit rather than scientific benefit, then the entire SETI goal was not really being pursued by the fine folks at SETI@home.
I don't run the client any more, although I think it would be neat if we found ETI. There's obviously no point.
If you're intending to flame me for something, please choose something more interesting that "You suck because you worked for Microsoft." Also: this was long ago, and in another country, and besides, the wench is dead.
The "Tragedy of the Commons" is that when there is common land, no one maintains it. Think of a communal kitchen in a geekhouse or something - typically one person does more of the cleaning than the others. At least, that's the case with people I know. But I digress.
The tragedy of the commons was not the land grab but the apparent disrepair that the commons were suffering from because no one was willing to maintain them at their own expense so that all could profit. "Why should I not overgraze so that my neighbor's sheep can be nice and plump? If I let my sheep eat all the grass on the commons (leaving none for anyone else), my sheep will be the plumpest and I'll make the most money!"
States in the US with a sales tax typically have a use tax at the same rate, which applies to goods imported across states lines for which less or no sales tax was paid. So if I buy something in a state with 3% sales tax, and live in a state with 5% sales tax, I must pay my home state enough money to total 5% tax paid. If I paid nothing to the other state, all 5% goes to my home state. If I paid 3% to the first, 2% goes home. If I paid 7% sales tax, I'm out 2%.
Practically speaking, no one deals with interstate sales tax or use tax if they can help it. If a company has no presence in a state with sales tax, it seems they can pretty much ignore that sales tax, since the state tax collectors cannot cross state lines to collect it (state's rights and all that), and although the states sometimes make a heck of a fuss about use tax, it is quite hard to figure out how and where to pay it, let alone bother paying.
There was a brilliant Asimov short story with the premise of world war. The computers in the guided missiles were too big and heavy, and they used so much energy to run, so the missiles were doing significantly less well than either side had hoped. Someone came up with the idea of teaching people trigonometry. They could guide the missiles themselves (presumably on kamikaze missions, although this was not made incredibly clear), and would be much more effective than the computers. [I guess you have to make some allowance for Asimov having written this when computers were obviously useful but not obviously following (the as-yet unstated) Moore's Law.]
Anyway, the names for these mathematical wonders, who could do sines and square roots in their heads? Yup, Asimov chose for the culture of his story to use the word "calculator."
It's not worth trying to figure it out. Yes, it's 11.5, and no, I didn't use a calculator. But 109/18? I'm pretty sure it's between 5 and 6. If I needed the answer, I'd ask xcalc. Why? Because I don't care enough to find the answer. If it is on a test, and I have to know the answer to complete the problem, I'll do subtraction by hand (carry the one, etc.), but usually I'll leave it as ... * 109/18 (or whatever). I can do integrals in my head (usually) and think in many dimensions and all that good stuff, but I do have trouble with basic math. I don't think it's because I used a calculator too much so much as a lack of general need-to-know most of the time.
That said, there is another reason I avoid doing such problems: I've found I'm mildly dyslexic, and often do something like 190/18 or something equally wrong. Missing two points for using a totally bogus constant is worse than dealing with the (extremely rare) prof who won't accept 109/18 in an answer.
And yes, my degree is in Math. Take this as you will.
This is at least as relevant today as when Heinlein wrote it about 30 years ago.
"There has grown up in the minds of certain groups in this country the notion that because a man or a corporation has made a profit out of the public for a number of years, the government and the courts are charged with the duty of guaranteeing such profit in the future, even in the face of changing circumstances and contrary public interest. This strange doctrine is not supported by statute nor common law. Neither individuals nor corporations have any right to come into court and ask that the clock of history be stopped, or turned back, for their private benefit. That is all."
- Robert A. Heinlein ("Life-Line")
It's coming. We just got the switches in at the house, so our build server is going to come on line and start spewing out packages. Files should be available on our web site as soon as there's anything to actually install.
I really appreciate your interest - it makes me feel that this is actually worthwhile!
Some friends of mine and I have been working on a system to netboot Linux. It turns out that it's not that difficult - a combination of TFTP for a kernel, NFS for a root filesystem, and some custom code to pull all of the stuff over. Think Solaris JumpStart meets Linux.
It's actually surprisingly clean. Right now, it actually works without requiring NFS, so after a little bit of preconfiguration on the server side, you can configure a shockingly large number of systems just by putting a floppy or CD in the drive and booting. (The NFS is planned, mostly for completeness, since the system at large can actually work without.)
No it shouldn't.
The scope of ii is from declaration through to the close curly braces of the block in which it is defined. So
int foo() {
} has a duplicate definition of ii, since the first declaration is in scope until the last curly brace I wrote. Any other compiler is behaving questionably if it allows it.How many people did you recognize there? (C'mon, I know that's why you were looking at most of the pictures - that's why I looked at more than just the happy couple.) I saw Rob and crew (big surprise), but I really think that the joker taking a picture of the camera was Mandrake, of Enlightenment fame. There's this shot of someone I'd swear is Geoff Harrison.
And after checking www.mandrake.net it seems that he did go to the wedding, which implies that the person he's standing next to is Tammy...
So I guess this is the next poll: How many people did you recognize at Hemos' wedding?