Re:Because you'll end up at Lisp.
on
Lisp and Ruby
·
· Score: 3, Insightful
I would have to disagree with your statement that Lisp doesn't let you group things. Not only does it let you group things, it lets you write entire languages built specifically for the things you want to group. Macros in Lisp are effectively little compilers that are very easy to write, that let you make whatever language is most useful for working with the concepts you want to "collect, isolate, and distinguish". Combine this with CLOS and generic and functional programming techniques, and you have one of the best languages around for dealing with concepts at a very high level.
Lisp has its weaknesses, but expressiveness and abstraction are not among them.
Wait, Java is as high-level as you can go? It doesn't have higher-order functions, proper tail recursion, continuations, or any kind of macro system, and you can still work directly with byte arrays without working against the language. "Garbage collection" does not mean "as high-level as you can go".
If someone doesn't get medicine because some idiot software company wrote a system that can't handle otherwise correct information typed in normal case, I would hope said software company would get sued into the ground.
Congratulations, you've made the most amazing mental contortion I've ever seen. Software can't be analyzed because of the halting problem, but the simple regurgitation that Peano arithmetic is undecidable is wrong because we can simulate a 32-bit multiplier in C using only addition. Good job!
I think you're misunderstanding what I want. I do want all of the elements in the pfor loop to run in parallel, with the parent thread blocking until all the jobs in the pfor loop are completed. So, if the VM knew how many processing resources were available, it could allocate each element of the pfor loop to an available processing resource.
The way I handle this situation now is I launch each parallel element in its own thread using a semaphore to control how many threads are available (I set the thread count with a commandline parameter or environment variable), and a counter to show how many jobs are outstanding. The parent thread monitors the value of the counter and resumes when that hits 0. It's usually not even worth doing...
I don't intend this as a general model for parallelism or threading, it just seems like a quick and easy way to get really basic parallelism with virtually no programming effort.
You can screw up a program with side effects even if you're just doing regular serial coding. How is this more cmplex than that? This isn't meant to be a full-blown replacement for threading - just a simple language construct to introduce a very simple form of parallelism...
Yeah, Java is what I typically program in, and that's the problem with Runnable. You have to explicit block the parent thread. Not a big deal, but it seems like extra complexity for something that should be so simple...
I haven't done a lot of multi-threaded programming, so maybe this is actually commonly available, but I think a nice language-level parallelism feature would be something that could handle a really basic "for each" type loop:
serialCode();
pfor(element in collection) { element.parallelCode(); }
serialCode();
without having to worry about manually setting up the threads, etc - if there are multiple resources available, they get used, if not, then it happens in serial. Is there anything like this out now?
btw... how do you get proper indentation using this ecode tag?
Wait, what? The DMV? Try getting a job at ANY place that requires the skillset of the DMV. It's not like shitty jobs are magically less shitty (or less filled with shitty people) just because they're private sector. Have you called a customer service center recently? Have you called their competitors' customer service centers? Obviously not, otherwise you would have been disabused of the notion that the magic power of choice has any kind of measurable effect in any kind of reasonable timeframe.
You talk about private corporations being 100% voluntary. Is that why I receive hundreds of spam emails a day? Or why I used to get at least ten telemarketing calls at dinnertime until the government stepped in? Or why my grandmother almost got scammed by some pushy boiler room cold caller? Why my windshield never goes a day without collecting a flier advertising the dry cleaner down the street (who I've chosen not to go to, with very little noticeable effect)? Why Sony felt it necessary to install a rootkit on my computer without telling me or giving me the opportunity to opt out, when all I wanted to do was play some music? How many examples do I have to give?
As far as bankruptcies leading to efficiency: fine. That's a great thing to say about a company that's selling me toothpaste. What the fuck happens when it's the company supplying me with sewage service? Or the company that's treating my water supply? Or the company that's providing police or fire service? Look at Katrina: the local and state governments were overwhelmed, the federal government sat on its ass because it's run by Norquist nutjobs who want to privatize everything, and the private companies protected their own shit but did nothing for the city as a whole. Sure, government didn't look that great there, but how would you like to see a breakdown like that resulting from a private police force hiding the fact they were going bankrupt until the day they became insolvent? Or worse... a private police force realizing they could charge per incident and have people by the nutsack. Let me guess: insurance! Or hell, I could just form a vigilante mob with my neighbors, right?
As for your "billions of decisions every minute": you've got to be joking. The market has billions of transactions every second, yes, but that also means that my influence is directly correlated to my dollar. Want to kill a particular group of people? Want to hire some mercenaries to hunt down and shoot people with six toes? I don't, but if you're a billionaire and I'm not, you win, don't you? Letting the market be the ultimate arbitrator means that a persons influence in society is directly tied to how much money they have. We need to move AWAY from that scenario, not further towards it.
What, exactly, are you suggesting we do with the information you provide? Overthrow the government and replace everything it does with private for-profit corporations? Or elect every single little position paid for with tax dollars? Do shareholders typically vote on every janitor who gets hired?
I also don't understand the point you make about how the free market corrects for cronyism, and how this is better than the government. It sounds like the company has to go bankrupt or otherwise catastrophically fail for the correction to take place; is that what we would have to look forward to if we adopted that model for government? Isn't that kind of what we already have?
Bush has made a series of crappy appointments, and the public is starting to get upset about the apparent effects of that. In the next election, a politician would be wise to latch on to this incompetence. If people care about it enough, they'll vote on it; if they have other priorities, they'll vote on those. Sounds suspiciously market-like to me, and it happens regularly every 2-4-6 years.
This isn't about keeping speech private. Most of these professors would (and do) gladly stand up and loudly, uh, profess their ideology. That's kind of what they do. The issue here is the idea that their speech is somehow evil and should be turned over to the mob for the sake of doling out punishment.
It's not the recording, it's the motivation for the recording.
As regards the money thing, I think another post in this thread put it nicely: your money isn't paying for the professor, it's paying for the University and the great benefits of the university system. Something that comes along with that is smart people who won't toe anyone's party line. I'm not including myself in that group, but I am a university employee and I do get paid with your tax dollars (thanks for all the beer).
Wait a second. You made a post claiming that 0 virus was a myth that could be exposed by doing some basic research, and someone replies asking you to name just one...
To which you reply with some sarcastic diatribe making it look like the reply was completely unreasonable and vindictive towards you. WALB...
Anyways, so can you or can't you name one virus for Mac?
I don't know about that. Apple budged up about as much in 1998 as they did with the announcement of OS X in 2000, before riding along on the tech bubble and burst. They also continued to lose market share after 1998, which was why Dell felt they were doomed. The first graph also shows when their stock really took off - after the iTunes Music Store was announced in 2003. I agree that the late 90s were very important for Apple, but to say the iMacs were bigger than the iPod and iTunes? I think that's debatable.
I also agree that Apple wouldn't be where it is today if Dell were running the company (I said as much in my original post). I don't know where Dell gets such a crappy reputation... I would much rather own a Dell than a Gateway, HP (now), Sony (rootkits preinstalled!), etc, and all my past experiences with Dell (my own computers and working tech support on a college campus) have been fine (far more Windows problems than Dell problems). The world needs cheap computers, and Dell does a good job meeting that need. Of course, I prefer Apple most of all...
So, wait... a very successful CEO who built his company from scratch to market leader, not a Carly Fiorina or some other overpaid asshole, makes an observation that Apple probably won't survive with its proprietary offerings and such a small market share, and he's got a big mouth? In April 2001 (pre-iPod, days after OS X v10.0, in the wake of the bubble burst) would anyone have disagreed very strenuously with this conclusion? The Appleatchiks might have been always faithful but the rest of the world could be forgiven for doubting.
In October of that year came the iPod, and in April of 2003 Apple opened the iTunes Music Store. Dell's failure was not that he was wrong in analyzing the fate of a company dedicated to making proprietary niche-market computers, but that he lacked the imagination to see that Apple wasn't always going to bank on proprietary niche-market computers. He says he would never take the job of Apple CEO. Honestly, would anyone have wanted Dell to head Apple? Would the iPod have come out under Dell? Would Apple have ever expanded into the media market under Dell? There's no way. He's just not that creative. So, thankfully, Michael Dell would never take the job of Apple CEO. So what?
Aye, right, you sound like a regular mama's boy. Raced cars at 160? What'd you drive, a pinto? My foot won't let off the gas of my Ferrari at anything less than 200 (even in neighborhoods!). Bikes on public roads? Why, I bike on some private roads - which happen to be on a military artillery range! Scuba dived below 60m? Might as well just be dipping your toe in. Hell, I can hold my breath to 100m while I'm hunting for sharks! You climb rocks, I climbed Everest with only a single Sherpa (who I left behind at 8,000m) and my bare hands and feet. You kayak whitewater, I do that too, but without a paddle and with lead weights tied to my ankles! Rugby is a fun game (for girls and homosexuals), and jiu-jitsu is Japanese for "slap fighting" (yes, I speak Japanese, as well as 43 other languages. I am the last remaining speaker for 16 distinct dialects). Instead of all that pussy stuff I wrestle gorillas that I've loaded up with steroids and bred to be super-intelligent, usually with an arm tied behind my back, sometimes blindfolded.
All women want to have sex with me (which means my ability to maintain an erection for up to 3 weeks at a time, with only an hour of downtime in between, really comes in handy), and all men want to be me (which they someday can, thanks to my extensive genetics research... did I mention that I'm a geneticist, as well as a nuclear physicist, meteorologist, and concert pianist? I also wrote an operating system called Calambracix that is used, interestingly enough, to run candy factories).
You would think that with all I've accomplished, I'd be a bit arrogant, but I'm actually very humble (possibly the most humble of anyone), as is mandated by the spiritual laws of Calambracism (a religion that I founded and, incidentally, am a primary spiritual figure of). It's therefore disappointing to me to hear you brag about your personal exploits as if they should be an example to the rest of the world. Considering how unimpressive your feats are, it would be most disappointing if a young child were exposed to the notion that they could settle for a life like yours. If you were humble like me, you would recognize your inferiority and hide your head in shame, never speaking in this forum again.
Legacy crap that deserves to die, like OSI? WTF? Between OSI and TCP/IP, TCP/IP is the legacy system. The reason OSI never took off is because everybody had already invested their money in making TCP/IP work, and OSI was seen as too difficult to implement in practice...
Argh, transcription error...
p(n) = Price per song per month at month n = 1 / (100 + 25(n-1)) -> p(1) = 0.01, p(2) =.008...
should actually be
p(n) = Price per song per month at month n = 1 / (100 + 10(n-1)) -> p(1) = 0.01, p(2) =.009...
I had originally tried to find out the pricing if you downloaded 250 songs a month, but it was taking too long to get to 99 cents...
Let's say I download 1000 songs my first month (all my favorites) and 100 songs per month (roughly 10 cds) thereafter, at $10/month. What is the lifetime cost of a song downloaded in the first month over the course of a year? Two years? Five years? Ten years? How many months before that song downloaded in the first month is the cost of an iTune?
p(n) = Price per song per month at month n = 1 / (100 + 25(n-1)) -> p(1) = 0.01, p(2) =.008...
One year cost is p(1) + p(2) + p(3) +... + p(12) and so on...
So, a quick little bit of code tells me that one of those original songs has cost: One year: 0.076 Two year: 0.122 Five year: 0.197 Ten year: 0.260
Also, it takes 189,418 months until it's 99 cents... which is quite a while. I guess my point is, it's a pretty good deal*
*Though it's pretty obvious that the only reason these companies have these prices is because iTunes is kicking their ass, and this is the only way they can turn heads. There's no way they'll keep these prices forever. Plus the DRM they use to 'protect' these songs is as draconian as it gets. Plus they don't work with iPods. All the same, it's hard to imagine that iTunes could raise prices with these guys practically giving music away...
Wait a second. Nobody's advocating throwing their hands up and saying "Oh, well" - that's the exact opposite of what I said.
When you say, "These bad things are completely the responsibility of those who do them," you're correct. The bad thing is absolutely the fault of the person who commits the act, and they should be punished accordingly.
Simultaneous to this obvious truth, I point out that this is all a big whoop-de-shit statement for someone who has been compromised. If you own or control any system, be it a computer, a house, a family, assigning responsibility after the fact is a hell of a lot less important than preventing the bad thing from occurring in the first place.
Deterrence is wonderful when it works, but how often does it work? In the absence of a perfect system, I make the evaluation that my interests are best served by taking strong action to defend myself and my systems, regardless of who is ultimately at fault for anything that may occur.
You're right, the bad guys are responsible. Congratulations, you get a cookie.
Now the actual question is: so what? If the ultimate responsibility is on the bad guy, does that mean I don't lock my door anymore? How does placing the responsibility correctly actually influence my actions? How is it useful information? Answer: it's not.
Slashdot Mirror
I would have to disagree with your statement that Lisp doesn't let you group things. Not only does it let you group things, it lets you write entire languages built specifically for the things you want to group. Macros in Lisp are effectively little compilers that are very easy to write, that let you make whatever language is most useful for working with the concepts you want to "collect, isolate, and distinguish". Combine this with CLOS and generic and functional programming techniques, and you have one of the best languages around for dealing with concepts at a very high level.
Lisp has its weaknesses, but expressiveness and abstraction are not among them.
Wait, Java is as high-level as you can go? It doesn't have higher-order functions, proper tail recursion, continuations, or any kind of macro system, and you can still work directly with byte arrays without working against the language. "Garbage collection" does not mean "as high-level as you can go".
If someone doesn't get medicine because some idiot software company wrote a system that can't handle otherwise correct information typed in normal case, I would hope said software company would get sued into the ground.
Don't other languages/cultures already have different keyboards?
Congratulations, you've made the most amazing mental contortion I've ever seen. Software can't be analyzed because of the halting problem, but the simple regurgitation that Peano arithmetic is undecidable is wrong because we can simulate a 32-bit multiplier in C using only addition. Good job!
A MacBook Pro isn't exactly the best computer to run Matlab on, either, considering that Matlab isn't available for the intel macs yet...
The way I handle this situation now is I launch each parallel element in its own thread using a semaphore to control how many threads are available (I set the thread count with a commandline parameter or environment variable), and a counter to show how many jobs are outstanding. The parent thread monitors the value of the counter and resumes when that hits 0. It's usually not even worth doing...
I don't intend this as a general model for parallelism or threading, it just seems like a quick and easy way to get really basic parallelism with virtually no programming effort.
Not exactly what I had in mind, but it looks like an interesting read, thanks!
You can screw up a program with side effects even if you're just doing regular serial coding. How is this more cmplex than that? This isn't meant to be a full-blown replacement for threading - just a simple language construct to introduce a very simple form of parallelism...
Yeah, Java is what I typically program in, and that's the problem with Runnable. You have to explicit block the parent thread. Not a big deal, but it seems like extra complexity for something that should be so simple...
You talk about private corporations being 100% voluntary. Is that why I receive hundreds of spam emails a day? Or why I used to get at least ten telemarketing calls at dinnertime until the government stepped in? Or why my grandmother almost got scammed by some pushy boiler room cold caller? Why my windshield never goes a day without collecting a flier advertising the dry cleaner down the street (who I've chosen not to go to, with very little noticeable effect)? Why Sony felt it necessary to install a rootkit on my computer without telling me or giving me the opportunity to opt out, when all I wanted to do was play some music? How many examples do I have to give?
As far as bankruptcies leading to efficiency: fine. That's a great thing to say about a company that's selling me toothpaste. What the fuck happens when it's the company supplying me with sewage service? Or the company that's treating my water supply? Or the company that's providing police or fire service? Look at Katrina: the local and state governments were overwhelmed, the federal government sat on its ass because it's run by Norquist nutjobs who want to privatize everything, and the private companies protected their own shit but did nothing for the city as a whole. Sure, government didn't look that great there, but how would you like to see a breakdown like that resulting from a private police force hiding the fact they were going bankrupt until the day they became insolvent? Or worse... a private police force realizing they could charge per incident and have people by the nutsack. Let me guess: insurance! Or hell, I could just form a vigilante mob with my neighbors, right?
As for your "billions of decisions every minute": you've got to be joking. The market has billions of transactions every second, yes, but that also means that my influence is directly correlated to my dollar. Want to kill a particular group of people? Want to hire some mercenaries to hunt down and shoot people with six toes? I don't, but if you're a billionaire and I'm not, you win, don't you? Letting the market be the ultimate arbitrator means that a persons influence in society is directly tied to how much money they have. We need to move AWAY from that scenario, not further towards it.
What, exactly, are you suggesting we do with the information you provide? Overthrow the government and replace everything it does with private for-profit corporations? Or elect every single little position paid for with tax dollars? Do shareholders typically vote on every janitor who gets hired?
I also don't understand the point you make about how the free market corrects for cronyism, and how this is better than the government. It sounds like the company has to go bankrupt or otherwise catastrophically fail for the correction to take place; is that what we would have to look forward to if we adopted that model for government? Isn't that kind of what we already have?
Bush has made a series of crappy appointments, and the public is starting to get upset about the apparent effects of that. In the next election, a politician would be wise to latch on to this incompetence. If people care about it enough, they'll vote on it; if they have other priorities, they'll vote on those. Sounds suspiciously market-like to me, and it happens regularly every 2-4-6 years.
It's not the recording, it's the motivation for the recording.
As regards the money thing, I think another post in this thread put it nicely: your money isn't paying for the professor, it's paying for the University and the great benefits of the university system. Something that comes along with that is smart people who won't toe anyone's party line. I'm not including myself in that group, but I am a university employee and I do get paid with your tax dollars (thanks for all the beer).
Wait... what? V8 sports cars are always marketed as much faster than four-cylinder economy cars.
Wait a second. You made a post claiming that 0 virus was a myth that could be exposed by doing some basic research, and someone replies asking you to name just one...
To which you reply with some sarcastic diatribe making it look like the reply was completely unreasonable and vindictive towards you. WALB...
Anyways, so can you or can't you name one virus for Mac?
I also agree that Apple wouldn't be where it is today if Dell were running the company (I said as much in my original post). I don't know where Dell gets such a crappy reputation... I would much rather own a Dell than a Gateway, HP (now), Sony (rootkits preinstalled!), etc, and all my past experiences with Dell (my own computers and working tech support on a college campus) have been fine (far more Windows problems than Dell problems). The world needs cheap computers, and Dell does a good job meeting that need. Of course, I prefer Apple most of all...
So, wait... a very successful CEO who built his company from scratch to market leader, not a Carly Fiorina or some other overpaid asshole, makes an observation that Apple probably won't survive with its proprietary offerings and such a small market share, and he's got a big mouth? In April 2001 (pre-iPod, days after OS X v10.0, in the wake of the bubble burst) would anyone have disagreed very strenuously with this conclusion? The Appleatchiks might have been always faithful but the rest of the world could be forgiven for doubting.
In October of that year came the iPod, and in April of 2003 Apple opened the iTunes Music Store. Dell's failure was not that he was wrong in analyzing the fate of a company dedicated to making proprietary niche-market computers, but that he lacked the imagination to see that Apple wasn't always going to bank on proprietary niche-market computers. He says he would never take the job of Apple CEO. Honestly, would anyone have wanted Dell to head Apple? Would the iPod have come out under Dell? Would Apple have ever expanded into the media market under Dell? There's no way. He's just not that creative. So, thankfully, Michael Dell would never take the job of Apple CEO. So what?
Aye, right, you sound like a regular mama's boy. Raced cars at 160? What'd you drive, a pinto? My foot won't let off the gas of my Ferrari at anything less than 200 (even in neighborhoods!). Bikes on public roads? Why, I bike on some private roads - which happen to be on a military artillery range! Scuba dived below 60m? Might as well just be dipping your toe in. Hell, I can hold my breath to 100m while I'm hunting for sharks! You climb rocks, I climbed Everest with only a single Sherpa (who I left behind at 8,000m) and my bare hands and feet. You kayak whitewater, I do that too, but without a paddle and with lead weights tied to my ankles! Rugby is a fun game (for girls and homosexuals), and jiu-jitsu is Japanese for "slap fighting" (yes, I speak Japanese, as well as 43 other languages. I am the last remaining speaker for 16 distinct dialects). Instead of all that pussy stuff I wrestle gorillas that I've loaded up with steroids and bred to be super-intelligent, usually with an arm tied behind my back, sometimes blindfolded.
All women want to have sex with me (which means my ability to maintain an erection for up to 3 weeks at a time, with only an hour of downtime in between, really comes in handy), and all men want to be me (which they someday can, thanks to my extensive genetics research... did I mention that I'm a geneticist, as well as a nuclear physicist, meteorologist, and concert pianist? I also wrote an operating system called Calambracix that is used, interestingly enough, to run candy factories).
You would think that with all I've accomplished, I'd be a bit arrogant, but I'm actually very humble (possibly the most humble of anyone), as is mandated by the spiritual laws of Calambracism (a religion that I founded and, incidentally, am a primary spiritual figure of). It's therefore disappointing to me to hear you brag about your personal exploits as if they should be an example to the rest of the world. Considering how unimpressive your feats are, it would be most disappointing if a young child were exposed to the notion that they could settle for a life like yours. If you were humble like me, you would recognize your inferiority and hide your head in shame, never speaking in this forum again.
"Deserves to die" doesn't make it the legacy system, which was the point of contention...
Legacy crap that deserves to die, like OSI? WTF? Between OSI and TCP/IP, TCP/IP is the legacy system. The reason OSI never took off is because everybody had already invested their money in making TCP/IP work, and OSI was seen as too difficult to implement in practice...
Argh, transcription error... p(n) = Price per song per month at month n = 1 / (100 + 25(n-1)) -> p(1) = 0.01, p(2) = .008...
should actually be
p(n) = Price per song per month at month n = 1 / (100 + 10(n-1)) -> p(1) = 0.01, p(2) = .009...
I had originally tried to find out the pricing if you downloaded 250 songs a month, but it was taking too long to get to 99 cents...
Let's say I download 1000 songs my first month (all my favorites) and 100 songs per month (roughly 10 cds) thereafter, at $10/month. What is the lifetime cost of a song downloaded in the first month over the course of a year? Two years? Five years? Ten years? How many months before that song downloaded in the first month is the cost of an iTune?
.008...
... + p(12) and so on...
p(n) = Price per song per month at month n = 1 / (100 + 25(n-1)) -> p(1) = 0.01, p(2) =
One year cost is p(1) + p(2) + p(3) +
So, a quick little bit of code tells me that one of those original songs has cost:
One year: 0.076
Two year: 0.122
Five year: 0.197
Ten year: 0.260
Also, it takes 189,418 months until it's 99 cents... which is quite a while. I guess my point is, it's a pretty good deal*
*Though it's pretty obvious that the only reason these companies have these prices is because iTunes is kicking their ass, and this is the only way they can turn heads. There's no way they'll keep these prices forever. Plus the DRM they use to 'protect' these songs is as draconian as it gets. Plus they don't work with iPods. All the same, it's hard to imagine that iTunes could raise prices with these guys practically giving music away...
When you say, "These bad things are completely the responsibility of those who do them," you're correct. The bad thing is absolutely the fault of the person who commits the act, and they should be punished accordingly.
Simultaneous to this obvious truth, I point out that this is all a big whoop-de-shit statement for someone who has been compromised. If you own or control any system, be it a computer, a house, a family, assigning responsibility after the fact is a hell of a lot less important than preventing the bad thing from occurring in the first place.
Deterrence is wonderful when it works, but how often does it work? In the absence of a perfect system, I make the evaluation that my interests are best served by taking strong action to defend myself and my systems, regardless of who is ultimately at fault for anything that may occur.
Now the actual question is: so what? If the ultimate responsibility is on the bad guy, does that mean I don't lock my door anymore? How does placing the responsibility correctly actually influence my actions? How is it useful information? Answer: it's not.