The only difference is that cell phones are encrypted. Maybe if Google and breaking encryption and listening in on network's, then your analogy would work.
I so responded to another post. Someone was responding to your post saying you where wrong.
"Really? seems to differ [arstechnica.com] and wasn't the only reference I could find for microsoft.com defaced [bing.com] (seventh link)."
They were saying that MS already had issues with security when really it was just bad programming on the app side and his examples had nothing to do with OS security.
Really? seems to differ and wasn't the only reference I could find for microsoft.com defaced (seventh link).
SQL injection != Kernel Flaw
Not an example of bad security but bad programming.
http://www.xatrix.org/article.php?s=3640 "They found the administration page and performed a SQL injection attack, allowing them to manage the content of the section."
OMG!!! Linux is SUPAR UNSAFE!!! It is vulnerable to SQL injection attacks!... Every OS has this issue because some moron decided to not validate their SQL string and/or didn't use parameterized variables.
Actually, I went through and googled a bunch myself and all the results for the past decade where SQL injection or they didn't specify but mostly SQL.
Next time you feel like showing off how un-secure and OS is, I'll load up SELinux, set it up to let root telnet in, give root a blank password, open up all the ports on my firewall and see how long it takes for SELinux to be "hacked"
it should be more along the line of "if you offer a free version of something and it becomes industry standard, you lose your rights to it".
It gets rid of this bait-and-switch bullshit where they make a codec free/easy and it becomes standard then they decide to charge for it once it everyone NEEDS it.
Because it's a pain in the ass and very hard for most coders.
What we need is either a simple library for threading or a new language (like haskell) for auto-parallelization
Parallel programming is like working with SQL. You have to structure your data in a way that's multi-thread friendly.
I was in a discussion in another forum about Multi-threaded programmer. The general consensus was "If you make multi-threaded programmer easier for idiot programmers, you'll just be giving them a bigger gun to shoot themselves with".
The issue with all of these new language tools is you can make an easy parallel for call on an array, only to find out you either made it slower or you sometimes get bad data because you still don't understand how it works. The new parallel language extensions are there to give programmers a shorthand version, but some programmers are going to think that these new addons magically make stuff run faster and you won't have to worry about anything.
Even lowly i5 cpu can boot Win7 is under 10 seconds with an SSD and can run any game and decode any video with ease.
Most people may not need a quad core, but even a dual core makes your OS experience seem much more responsive.
AMD and Intel do not "convert" their chips into single cores. It's just logical that if you're loading the whole cpu, why not make one part faster. No point in wasting electricity with 3 of of your cores being powered on. XP is great at single threads and does not scale almost at all with SMP. Single threaded programming has a HUGE diminishing return on ROI for transistor count. Better off making a a whole separate core.
What it comes down to is there are only a few ways to increase single threaded speed. #1. Out-of-Order Execution. This takes quite a bit of transistors and is only as good as your instruction windows size. You cannot change the order of many instructions. Also, the larger your instruction window, the slower your chip has to run as it needs more time to check all of the instructions first.
#2. Increase register count. More registers means fewer moves and fewer memory load/stores (for tight loops anyway), but this also has diminishing returns because you can only use so many registers at a time with your instructions and eventually you have to move on anyway and load/store more data from memory which means after a few extra registers, your bottleneck moves back to memory.
#3. SIMD. MMX/SSE/ETC are all great at crunching matricies of data and sometimes using special instructions that effectively perform many general isntructions in fewer steps than using those general purpose instructions. SIMD itself is also exclusinvely useful for Graphics/Math/Science/Games and near useless for everyday computing.
#4. MHZ. And who could forget the P4(aka space heater). Seems increasing clock speed takes a a huge amount more power as it goes up as you need to use more and more power to quickly oscilate your clock. They also found more interesting problems with higher clock speeds. Seems electricity can only travel so fast and they had to add extra steps in the pipeline to give the signals enough time to propigate accross the chip.
#5. Pre-fetching and/or more cache. You can add some instructions to help prefetch data but this has limited use as it consumes at least a cycle where you could have done real work and you don't always know yourn next memory address to make use of prefetch. More cache has the lovely problem of adding more latency which would make pre-fetching more useful but at a heavy cost of affecting everything else.
#6. anything else I forgot about.
In the end you have two things. 1) single threaded performance about hit a brick wall and was going almost no where. 2) If you don't go SMP, someone else will and you'll die a corprate death for not advancing.
Much of this latency could be compensated by your more registers idea.
Too bad you're still limited to one instruction per second. All those registers are only good at reducing requests back out to the L1.
'All in all, you'll have to have a ton of code re-written and to gain a small sub double-digit percentage on single threads and running much slower but costing the same as a multi-core setup on thread-heavy tasks.
In other words, this is only good for a niche market
You also run the issue that a single core can only get so large before you start to do what they did with the P4 and put stall stages in the pipeline to give signals time enough to propagate across the chip.
when you want to measure a certain value, you try your best to isolate it. It's hard to "see" how fast a page is rendered when you start to add a ton of other variables that could slow down the response.
if you knew the anatomy of male plumbing, you'd notice that a vasectomy disconnect the testicles from the prostate. No, you can't remove sperm from the testicles anymore, but an orgasm still removes fluid from the prostate just no more sperm.
At first it *sounds* like ScienceDaily's report shows a negative link with masturbation and prostate cancer, but really it's just bad journalism. If you actually read past the journalist's bull crap and read what the research says you'll see that their result was that men with *high sex hormone levels* are more promiscuous more horny and tend to have higher prostate cancer risks.
Based on what the research says, it sounds more that men with higher hormones WANT to jack off more because of how horny they are. The masturbation isn't the cause, it's the hormones.
The journalist just wanted a catchy headline. Not to mention if this was true, I'd have a hard time taking a single research with a small group over multiple researches with MUCH larger groups and more diversity.
Few other points(quotes) in ScienceDaily's article:
#1. Men with prostate cancer were more likely to have had a sexually transmitted disease than those without prostate cancer. #2. “Hormones appear to play a key role in prostate cancer and it is very common to treat men with therapy to reduce the hormones thought to stimulate the cancer cells #3. “A possible explanation for the protective effect that men in their fifties appear to receive from overall sexual activity, and particularly masturbation, is that the release of accumulated toxins during sexual activity reduces the risk of developing cancer in the prostate area. #4. (non quote) they also mention that younger men tend to have more partners than the older men. So it seems the risk of cancer is more of an issue of multiple partners and risky sex while older men tend to have safer sex typically with their wife or partner.
As you can tell, it's just a headline to grab attention and the 3rd point even agrees with the other previous studies, even though it has NOT be proven as of yet.
I other OLD news, men who ejaculate 5 times per week had a 66% less chance of prostate cancer EVER in their life than men who only did once or less per week. They has to be started young and you must continue this into your 50s. 3 different studies on this from 3 different universities in 3 different countries and all 3 go roughly(damn near the same) the same results even though they went about different ways of testing. They one test alone was a periodic questioner that followed over 30,000 men. Is that a good sample size? There seems to be a semi linear link, so each day you clean the pipes per week gives you and ~X% reduced chance of getting this dreaded disease.
I say an orgasm of prevention is worth more than a pound of cancer.
The current theory is that the prostate is great at concentrating carcinogens because of how it excretes and re-absorbs fluids which essentially filters and captures bad crap. Gotta flush it out.
Remember, a 66% reduction of the chance of cancer is like saying "you have a 200% INCREASED chance of caner if you don't".
P.S. remember to tell your wife to put out or close the door, because you're busy curing cancer.
I use Charter at home to. Typo a domain and get their crappy advertising. Typically, my Chrome Address bar does an extremely good job auto completing and/or fixing my typos anyway.
One nice thing about using the Charter DNS is it's very fast. I'm getting an average 30ms response time via DNS from them when running a DNS benchmark. Even Google's DNS 8.8.x.x is coming back with ~70ms. Now mind you, I have a 30ms ping to Google at ALL times of the day, so it's 15ms there 40ms to process the request and 15ms back. Charter's DNS is something like 5ms there, 20ms to process and 5ms back.
My response values where acquired using some DNS benchmark tool from Google and averaged over 34k DNS entries.
Obviously my local branch DNS server is going to have much better network response to me and much less load than Google's servers, but the only metric I'm using to "grade" the servers is response. Google may give better "quality" response.
I'd just say that they're interfering with my encrypted connection and I have probable cause that they're doing a man-in-the-middle attack. Put the burden of proof on them by doing a large audit.
It's not the abuses anyone is complaining about, it's the cover ups. Sure, every profession is going to have people who piss on the ethical standards of that profession, and there's no reason a religious profession would somehow dodge that.
[...]
With the catholic church, they covered up the pedophilia for decades, and now that they can't hide it any more, do they at least finally apologize, vow to fix it, and start making good on that promise by immediately kicking the most obvious offenders out of the clergy and turning them over the cops? Nope, they instead whine that that transparency of the internet is bad, because it makes their wrongdoing public. That isn't bad PR, that's a systemic failure of the morals they claim to uphold.
[..]
I agree. The worst kind of evil is having the power to stop evil, but doing nothing about it. When it comes to morals/ethics, you're only as "good" as your weakest link. If the Vatican sees a pedophile and does nothing of it, then the Vatican is giving it's reputation and telling people that this man/woman is of Vatican quality even though it knows what he/she actually does. They are no better than the people they willing support when knowing of their evils.
One thing others haven't pointed out yet is there are many other *fixed* power consumers. Like NICs, switches, motherboards in general.
Even with a CPU that is 1/40th the speed, but 1/100th the power is you still have to use 40 times more computers, which also means 40 times more network ports.
You also have to remember a mostly fixed amount of overhead from the DB/OS standpoint. It might only be 1/40th the speed, but you may take a relative performance hit from XXX amount of mips just to calculate indexes and requests/etc.
SSD cost is limited by the cost to refine and turn Silicon into Flash Memory.
The price will only go down as the process size goes down, currently at 32nm with Intel's Latest drives. Once it reaches 8nm or the like then the cost will truly be comparable to Hard Drives. Until then, don't expect a miracle.
I agree with this. Most of the price drops for SSDs was Flash memory suddenly going from 50 to 40 to 30nm as they quickly caught up with chip sizes. Now that SSDs are caught up with current fabrication processes, the prices will stagnate until a new better fabrication comes out.
There will obviously be price drops as the cost of initial investments are covered and when a slightly newer more competitive model comes out and they need to clear old-stock. But overall, they're like CPUs now. Expensive when they first come out, drop sharply shortly after and level out with a minor decline until the next big thing.
I find if you set your consumer threads to high priority and spin-lock while waiting on your producer threads which are idle priority, it makes your code fast.
Actually, I came out out college with $30k debt of student loans, $7k of CC debt, $2k of medical, no job, no car.
Unemployeed for 3 months after graduation, no unemployment pay because of student position. Wife was unemployeed for 6 month before that because of injury, with no pay, and I was making $300/month for 6 months before I graduated.
During my last year of college, I literally could make $10 pay for breakfast and lunch for a full week. Live on ramen and peanuts for over a year.
Shortly after I found a job, the CC debt collector called me up and said I had 14 days to pay off the full $7k or go to court because I was unable to pay them anything for 2 years.
All my friends have moved away because of jobs. I don't go out drinking, I don't go out to movies, I don't buy movies, I don't go out to eat, and don't travel. All I have is my computer for fun. Ohh, and I have anxiety/depression but I have severe reactions to any meds. If I get too stressed out, I completely lock down and can't function, yet I can't take drugs to level it out.
Wife still can't find a job because of market, medical limitations, and no transportation. We have almost no public transportation where I live.
Most of my friends, who make about the same amount as me, spend more in one night at the bar then I spend on non-essentials in an entire month since they don't have a mountain of debt and have to support another person.
I have about $20 set aside every month as "play" money. The rest goes all to bills.
I figure setting aside $2k of splurge money once per year is kind of important to my mental state.
Doesn't it annoy 45% of Americans that the government holds $2k of their money, giving it back in April? There must be a better way!
(Like the way I pay tax: the correct amount straight out of my paycheque every month.)
Thee things
#1. I like to put in too much since my wife likes to find things to buy when she has extra $$$ #2. When I get all that $$$ back, it get it all back around the same time as TONS of internet deals. This means, I have money ready when I want it. #3. My loans are based on my pay-check income. If I keep my taxes high, then my pay-checks looks smaller, so I can keep some of my loans in a deferred state. I'm paying off quite a few loans coming out of college, and it's nice to have a little bit of spending $$$. My tax return doesn't count towards my total income for my loans.
Slashdot Mirror
The only difference is that cell phones are encrypted. Maybe if Google and breaking encryption and listening in on network's, then your analogy would work.
listening != accessing
XP64 is considered as 2k3 as it derives from 2k3 while XP32 is derived from 2k
I so responded to another post. Someone was responding to your post saying you where wrong.
"Really? seems to differ [arstechnica.com] and wasn't the only reference I could find for microsoft.com defaced [bing.com] (seventh link)."
They were saying that MS already had issues with security when really it was just bad programming on the app side and his examples had nothing to do with OS security.
Really? seems to differ and wasn't the only reference I could find for microsoft.com defaced (seventh link).
SQL injection != Kernel Flaw
Not an example of bad security but bad programming.
http://www.xatrix.org/article.php?s=3640
"They found the administration page and performed a SQL injection attack, allowing them to manage the content of the section."
OMG!!! Linux is SUPAR UNSAFE!!! It is vulnerable to SQL injection attacks!... Every OS has this issue because some moron decided to not validate their SQL string and/or didn't use parameterized variables.
Actually, I went through and googled a bunch myself and all the results for the past decade where SQL injection or they didn't specify but mostly SQL.
Next time you feel like showing off how un-secure and OS is, I'll load up SELinux, set it up to let root telnet in, give root a blank password, open up all the ports on my firewall and see how long it takes for SELinux to be "hacked"
it should be more along the line of "if you offer a free version of something and it becomes industry standard, you lose your rights to it".
It gets rid of this bait-and-switch bullshit where they make a codec free/easy and it becomes standard then they decide to charge for it once it everyone NEEDS it.
"Good Hygiene" means you're clean enough to not have health concerns, has nothing to do with smelling good.
So, yes, it's a good thing to point out.
And don't be anti-social, you're going to have to ask a lot of questions to others to find out whom to contact for what/etc.
Because it's a pain in the ass and very hard for most coders.
What we need is either a simple library for threading or a new language (like haskell) for auto-parallelization
Parallel programming is like working with SQL. You have to structure your data in a way that's multi-thread friendly.
I was in a discussion in another forum about Multi-threaded programmer. The general consensus was "If you make multi-threaded programmer easier for idiot programmers, you'll just be giving them a bigger gun to shoot themselves with".
The issue with all of these new language tools is you can make an easy parallel for call on an array, only to find out you either made it slower or you sometimes get bad data because you still don't understand how it works. The new parallel language extensions are there to give programmers a shorthand version, but some programmers are going to think that these new addons magically make stuff run faster and you won't have to worry about anything.
We just need fewer fail programmers.
#1. is correct, but #2 is not.
Even lowly i5 cpu can boot Win7 is under 10 seconds with an SSD and can run any game and decode any video with ease.
Most people may not need a quad core, but even a dual core makes your OS experience seem much more responsive.
AMD and Intel do not "convert" their chips into single cores. It's just logical that if you're loading the whole cpu, why not make one part faster. No point in wasting electricity with 3 of of your cores being powered on. XP is great at single threads and does not scale almost at all with SMP. Single threaded programming has a HUGE diminishing return on ROI for transistor count. Better off making a a whole separate core.
What it comes down to is there are only a few ways to increase single threaded speed.
#1. Out-of-Order Execution. This takes quite a bit of transistors and is only as good as your instruction windows size. You cannot change the order of many instructions. Also, the larger your instruction window, the slower your chip has to run as it needs more time to check all of the instructions first.
#2. Increase register count. More registers means fewer moves and fewer memory load/stores (for tight loops anyway), but this also has diminishing returns because you can only use so many registers at a time with your instructions and eventually you have to move on anyway and load/store more data from memory which means after a few extra registers, your bottleneck moves back to memory.
#3. SIMD. MMX/SSE/ETC are all great at crunching matricies of data and sometimes using special instructions that effectively perform many general isntructions in fewer steps than using those general purpose instructions. SIMD itself is also exclusinvely useful for Graphics/Math/Science/Games and near useless for everyday computing.
#4. MHZ. And who could forget the P4(aka space heater). Seems increasing clock speed takes a a huge amount more power as it goes up as you need to use more and more power to quickly oscilate your clock. They also found more interesting problems with higher clock speeds. Seems electricity can only travel so fast and they had to add extra steps in the pipeline to give the signals enough time to propigate accross the chip.
#5. Pre-fetching and/or more cache. You can add some instructions to help prefetch data but this has limited use as it consumes at least a cycle where you could have done real work and you don't always know yourn next memory address to make use of prefetch. More cache has the lovely problem of adding more latency which would make pre-fetching more useful but at a heavy cost of affecting everything else.
#6. anything else I forgot about.
In the end you have two things. 1) single threaded performance about hit a brick wall and was going almost no where. 2) If you don't go SMP, someone else will and you'll die a corprate death for not advancing.
More L1 cache = more latency
Much of this latency could be compensated by your more registers idea.
Too bad you're still limited to one instruction per second. All those registers are only good at reducing requests back out to the L1.
'All in all, you'll have to have a ton of code re-written and to gain a small sub double-digit percentage on single threads and running much slower but costing the same as a multi-core setup on thread-heavy tasks.
In other words, this is only good for a niche market
You also run the issue that a single core can only get so large before you start to do what they did with the P4 and put stall stages in the pipeline to give signals time enough to propagate across the chip.
when you want to measure a certain value, you try your best to isolate it. It's hard to "see" how fast a page is rendered when you start to add a ton of other variables that could slow down the response.
???
if you knew the anatomy of male plumbing, you'd notice that a vasectomy disconnect the testicles from the prostate. No, you can't remove sperm from the testicles anymore, but an orgasm still removes fluid from the prostate just no more sperm.
At first it *sounds* like ScienceDaily's report shows a negative link with masturbation and prostate cancer, but really it's just bad journalism. If you actually read past the journalist's bull crap and read what the research says you'll see that their result was that men with *high sex hormone levels* are more promiscuous more horny and tend to have higher prostate cancer risks.
Based on what the research says, it sounds more that men with higher hormones WANT to jack off more because of how horny they are. The masturbation isn't the cause, it's the hormones.
The journalist just wanted a catchy headline. Not to mention if this was true, I'd have a hard time taking a single research with a small group over multiple researches with MUCH larger groups and more diversity.
Few other points(quotes) in ScienceDaily's article:
#1. Men with prostate cancer were more likely to have had a sexually transmitted disease than those without prostate cancer.
#2. “Hormones appear to play a key role in prostate cancer and it is very common to treat men with therapy to reduce the hormones thought to stimulate the cancer cells
#3. “A possible explanation for the protective effect that men in their fifties appear to receive from overall sexual activity, and particularly masturbation, is that the release of accumulated toxins during sexual activity reduces the risk of developing cancer in the prostate area.
#4. (non quote) they also mention that younger men tend to have more partners than the older men. So it seems the risk of cancer is more of an issue of multiple partners and risky sex while older men tend to have safer sex typically with their wife or partner.
As you can tell, it's just a headline to grab attention and the 3rd point even agrees with the other previous studies, even though it has NOT be proven as of yet.
I other OLD news, men who ejaculate 5 times per week had a 66% less chance of prostate cancer EVER in their life than men who only did once or less per week. They has to be started young and you must continue this into your 50s. 3 different studies on this from 3 different universities in 3 different countries and all 3 go roughly(damn near the same) the same results even though they went about different ways of testing. They one test alone was a periodic questioner that followed over 30,000 men. Is that a good sample size? There seems to be a semi linear link, so each day you clean the pipes per week gives you and ~X% reduced chance of getting this dreaded disease.
I say an orgasm of prevention is worth more than a pound of cancer.
The current theory is that the prostate is great at concentrating carcinogens because of how it excretes and re-absorbs fluids which essentially filters and captures bad crap. Gotta flush it out.
Remember, a 66% reduction of the chance of cancer is like saying "you have a 200% INCREASED chance of caner if you don't".
P.S. remember to tell your wife to put out or close the door, because you're busy curing cancer.
I use Charter at home to. Typo a domain and get their crappy advertising. Typically, my Chrome Address bar does an extremely good job auto completing and/or fixing my typos anyway.
One nice thing about using the Charter DNS is it's very fast. I'm getting an average 30ms response time via DNS from them when running a DNS benchmark. Even Google's DNS 8.8.x.x is coming back with ~70ms. Now mind you, I have a 30ms ping to Google at ALL times of the day, so it's 15ms there 40ms to process the request and 15ms back. Charter's DNS is something like 5ms there, 20ms to process and 5ms back.
My response values where acquired using some DNS benchmark tool from Google and averaged over 34k DNS entries.
Obviously my local branch DNS server is going to have much better network response to me and much less load than Google's servers, but the only metric I'm using to "grade" the servers is response. Google may give better "quality" response.
I'd just say that they're interfering with my encrypted connection and I have probable cause that they're doing a man-in-the-middle attack. Put the burden of proof on them by doing a large audit.
It's not the abuses anyone is complaining about, it's the cover ups. Sure, every profession is going to have people who piss on the ethical standards of that profession, and there's no reason a religious profession would somehow dodge that.
[...]
With the catholic church, they covered up the pedophilia for decades, and now that they can't hide it any more, do they at least finally apologize, vow to fix it, and start making good on that promise by immediately kicking the most obvious offenders out of the clergy and turning them over the cops? Nope, they instead whine that that transparency of the internet is bad, because it makes their wrongdoing public. That isn't bad PR, that's a systemic failure of the morals they claim to uphold.
[..]
I agree. The worst kind of evil is having the power to stop evil, but doing nothing about it. When it comes to morals/ethics, you're only as "good" as your weakest link. If the Vatican sees a pedophile and does nothing of it, then the Vatican is giving it's reputation and telling people that this man/woman is of Vatican quality even though it knows what he/she actually does. They are no better than the people they willing support when knowing of their evils.
encoding and encrypting are just different sides of the same coin. They're both *just* transformations of data.
Key word.. SEARCH. You can't search an encrypted field unless it matches exactly.
I agree. Google is anti-competitive just like Linux.
One thing others haven't pointed out yet is there are many other *fixed* power consumers. Like NICs, switches, motherboards in general.
Even with a CPU that is 1/40th the speed, but 1/100th the power is you still have to use 40 times more computers, which also means 40 times more network ports.
You also have to remember a mostly fixed amount of overhead from the DB/OS standpoint. It might only be 1/40th the speed, but you may take a relative performance hit from XXX amount of mips just to calculate indexes and requests/etc.
SSD cost is limited by the cost to refine and turn Silicon into Flash Memory.
The price will only go down as the process size goes down, currently at 32nm with Intel's Latest drives. Once it reaches 8nm or the like then the cost will truly be comparable to Hard Drives. Until then, don't expect a miracle.
I agree with this. Most of the price drops for SSDs was Flash memory suddenly going from 50 to 40 to 30nm as they quickly caught up with chip sizes. Now that SSDs are caught up with current fabrication processes, the prices will stagnate until a new better fabrication comes out.
There will obviously be price drops as the cost of initial investments are covered and when a slightly newer more competitive model comes out and they need to clear old-stock. But overall, they're like CPUs now. Expensive when they first come out, drop sharply shortly after and level out with a minor decline until the next big thing.
I find if you set your consumer threads to high priority and spin-lock while waiting on your producer threads which are idle priority, it makes your code fast.
Actually, I came out out college with $30k debt of student loans, $7k of CC debt, $2k of medical, no job, no car.
Unemployeed for 3 months after graduation, no unemployment pay because of student position. Wife was unemployeed for 6 month before that because of injury, with no pay, and I was making $300/month for 6 months before I graduated.
During my last year of college, I literally could make $10 pay for breakfast and lunch for a full week. Live on ramen and peanuts for over a year.
Shortly after I found a job, the CC debt collector called me up and said I had 14 days to pay off the full $7k or go to court because I was unable to pay them anything for 2 years.
All my friends have moved away because of jobs. I don't go out drinking, I don't go out to movies, I don't buy movies, I don't go out to eat, and don't travel. All I have is my computer for fun. Ohh, and I have anxiety/depression but I have severe reactions to any meds. If I get too stressed out, I completely lock down and can't function, yet I can't take drugs to level it out.
Wife still can't find a job because of market, medical limitations, and no transportation. We have almost no public transportation where I live.
Most of my friends, who make about the same amount as me, spend more in one night at the bar then I spend on non-essentials in an entire month since they don't have a mountain of debt and have to support another person.
I have about $20 set aside every month as "play" money. The rest goes all to bills.
I figure setting aside $2k of splurge money once per year is kind of important to my mental state.
Doesn't it annoy 45% of Americans that the government holds $2k of their money, giving it back in April? There must be a better way!
(Like the way I pay tax: the correct amount straight out of my paycheque every month.)
Thee things
#1. I like to put in too much since my wife likes to find things to buy when she has extra $$$
#2. When I get all that $$$ back, it get it all back around the same time as TONS of internet deals. This means, I have money ready when I want it.
#3. My loans are based on my pay-check income. If I keep my taxes high, then my pay-checks looks smaller, so I can keep some of my loans in a deferred state. I'm paying off quite a few loans coming out of college, and it's nice to have a little bit of spending $$$. My tax return doesn't count towards my total income for my loans.