I was thinking the same thing and I re-read it. I think the newly arrested individuals stole over 2 mil, the whole organization has probably stolen quite a bit more.
It's really not a resounding no to the merger, the basicly said we won't support any sort of deal if MSFT lowers the offer. They also said if you pay an extra $1 a share we'll support your takeover 100%.
Good point, there is a minimum threshold for crap where you are better off not releasing it. I sit in on a lot of high level planning meettings, and the sales/marketing teams are actually scary accurate with their projections unless an X factor jumps out that makes your product less relavant.
Unfortunatly sales/marketing has to be the people setting deadlines to a large extent it's up to engineers and programmers to push back when the objectives are horribly unreasonable. As much as it sucks to us technical people companies don't stay in business by making superior products, they stay in business through superior sales. A buggy half assed product delivered early will yield by far more sales than a complete product will that comes to market a couple quarters later.
For enterprise class storage this seems pretty reasonable. the two high end drives which only produce ~150 IOPS are 146GB FC and 300GB FC these drives typically run in 1k-2k range, although they can be purchased cheaper then that someplaces. So 852GB is ~6 146GB drives, or about 6k on the low end, which put the price on par for a superior performing product. Now add in the power savings from having fewer drives, using a more power efficient technology, using less cooling for the disk. Look at the money saved on drawers, extra FC cables,SFPs, and actually fewer enterprise level disk systems because in all Enterprise level systems the bottle neck is the number of spindles. Big pictures the flash drives will probably be cheaper up front, provide significant power savings( Go Green ), and reduce very expensive data center footprint. I'm actually surprised there isn't a big flash based data enterprise class storage solution out there yet.
How old is your 360. I have an original release 360 and it's a little loud. My friend baught one a couple months ago and it's like night and day on the noise issue.
The only two companies in that list that shouldn't be merged are Microsoft and Google. The other comapnies, Exxon, AT&T, and either Microsoft or Google could merge no problem. Event though they are huge, they don't have that much overlap in business and if the combined Googattxon tried to do something like block other search engines from AT&T network or something else they would shoot themselves in the foot so bad they would start falling apart immediatly. There are very few conglomerates that really opperate efficiently and just because you are the biggest oil company and the biggest phone company you don't really gain that much more leverage then you had before.
IBM didn't decide to switch from a lease model to a sales model they were forced to switch to it as a result of the anti-trust lawsuit that was filed against IBM.
The link you posted lists alternative fuels as sales and use tax free not fuel tax free. The fuel tax in NC was suppose to be used maintance and expansion of the roads. Since NCs fuel tax is part flat fee and part percentage they have actually been running surplus with the high gas prices, which they have been using to pay down some of the education bonds and teacher raises.
I have, most of the time not intentionaly mind you, but there have been several occasions where I mistyped a domain got the page o' ads and then a window from another program pops up I click somewhere and end up accidentally clicking on an ad. There are other times where I'll go to some URL like www.aspecificplace.com the ad screen pops up and the first ad says the official web page of specific place, so I give it try, A good potion of the time 80% plus, it's the right palce. The other thing is how many people do you really need to click to be sucessful. You can get a domain for $9/year, let's say the ad get's you $.50, get 18 people to click on it and you paid the domain registration. So all you need is 2 mistaken or shot in the dark clicks a month and you're making money. If you get a couple a day you're making several hundred a year. I do know people who will go to an ad trap and click on each item, all 20+ of them thinking one click might take them where they want to go.
Trolling for a fight much. I was just stating a couple facts. People need to do what people need to do, but here are a couple of more facts for ya. Dell got 242 million in incentives, wich is estimated to cost the States 72 million over 20 years, so yes the state will get some of it back through new taxes and rest absorbed by the tax payers. What I find interesting in the Dell deal was that the next nearest incentive package totalled 37 million, 6.5 times less then the NC deal. So you might say that over a 30 year window it might pay off, and it might if Dell doesn't skip town the moment there is a better deal, which if you look at past history is pretty much what they do. The other question you have to ask is in this case might 242 million been better spent elsewhere, incentives for smaller businesses maybe, businesses without 6 billion in cash.
I'm not opposed to states, towns, counties offering incentives but incentives that cost the taxpayers millions of dollars need to be looked at closely, which I'm sure you would agree with.
I'm sure there other other reasons besides tax breaks, and you make a good point that most everywhere some sort of incentive is going to be offered but NC gives huge incentives, look at the Dell deal I mentioned before. The total incentive package is $242 million, the next closest bid was $37 million by a county in VA, that's 6.5 times the amount of incentive, so while everyone offers incentives NC offers huge ones.
My bet it has more to do with states that offer massive tax breaks to businesses for moving in. I know North Carolina is famous for it, especially with the new Dell facility in Greensboro and Lenovo in RTP. The Dell deal was so good the state could have employed everyone hired by Dell for 11 years with the tax breaks and loans.
Is comparing MTBF correct when saying different drive types are of the same reliability? By looking at an existing system you really aren't looking at idependant variables. Let's say I have two servers one that hits the drives alot and another that barely touches the drives so using what I'll call "the old rules of thumb" I would put the FC drives on the intensive server and SATA drives on the less intensive server. So after 1 year I get my first drive failure on each. One could conclude that SATA is as reliable as FC, but is it really? I setup my environment to more heavily hit the FC drives, when would the SATA drive have failed if I had placed SATAs where the FCs were? The only way to really compare drives would be to hit a large number of each different drive type with the same workload.
If you look at the way most places do tiered storage they'll put highly accessed data on an FC tier and then migrate less used data to a SATA tier, this might be one reason why the failure rates of the two drives look the same.
BlueGene runs on PPC440 cores not PPC405 cores. To the best of my knowledge 405 cores were never used in mobile phones. Most mobile phone software is designed to run on ARM processors and PPC and ARM code tend not to translate back and forth to each other very well. Never mind the fact that most of the mobile phone peripherils are designed to work with the AMBA bus and not the embedded PPC's(ePPC) PLB bus. Maybe you are referring to ePPC cores being used in chips for cell phone base stations.
IBM's older super computers were based on Power5 Technology, so IBM did use it in some of their most advanced computer systems.
While power and heat are very important chips like the Power5 are very important even though clusters of lower performance chips can get massive parallelization. Some application can be parallelized so your performance ultimatly becomes that of your fastest processing unit. So Power5 based systems work on entirely different problem sets then BlueGene.
One of the definitions of pod is "Something resembling a pod, as in compactness. "
IANAL but I did recently take a class on patents, trademarks, and copyrights and one of the things we discussed was that trademarks that include generic terms had trouble being enforced. We specifically talked about Krispy Kreme. Krispy Kreme's trademark is specifically on Krispy Kreme with K's. They have tried to sue several people for selling products, even donuts with the name crispy cream, but when business have fought back the courts have always pretty much ruled that it wasn't a violation of trademark and the only way it would be was if the company specifically called it a Krispy Kreme something or other.
That being said Apple is probably just trying to cast a wide net in trying protect it's iPod trademarks. I think that the precedent set by other products( I have a Game Pod sitting on my desk that I got back in 98 that has 40 different card games) that pod generically defines any product that is compact in nature.
Depending on the situation, yes I would trust the engineer if it's SOP.
I have a processor manual in front of me, and for some reason it has more commands then shift, and, or, and xor. It has an int mult command, that only take two clock cycles, so any mult done by any other means except a bit shift power of two mult is going to take longer then just executing the mult command.
You don't need to know how every low level function works, that's how we make progress. We use the work that other people have done and build off of it. Granted there are specialty positions where in depth knowledge of the use of processor level instructions is necessary, and these people should know how to do that but they learned those skills through special classes, such as a compilers class, an embedded programming class, or etc etc etc.
You could also take this to the next level, you're not a computer programmer unless you know how the underlying logic works to execute an add or shift, or you're not a computer programmer unless you know how to design a CMOS circuit in a particular technology to implement a shift register.
To say anyone that doesn't know how to do assembly level add-shift multiplies or basic logical operations is just a secretary pushing buttons is wrong. I know many a gifted programmer who write awesome code that does awesome things, and have no idea how this works.
I write device drivers on occasion and I need to do things like this on occasion for performance, but I do it so other people don't need to think about how the real interface works. I write the low level code so some other person can write the higher level stuff.
I see what you're saying and some people need to learn how to do the add-shift multiplier, I believe these are your Computer Engineers today, not your software engineers, they tend to be the ones involved in you're low level voodoo, but think of it like this. 20 years ago when you were in school you spent 4 years learning how to program in C and assembly in order to execute add-shift multipliers. Today students still go to school for 4 years but have so much more to learn about. Where is the time for object oriented programming techniques, advanced data structures, multi-threaded software design, comm. theory, vector programming, and everything else if people are spending their time learning about add-shift multipliers or whatever archaic multi-step register operation you can think of.
Re:Only one way to resolve this...
on
Gmail vs Pine
·
· Score: 2, Insightful
Using this logic GMail should be disposed of for Yahoo, which had about 6x the number of results.
The CD will go away eventually, but what doesn't. While iTunes and P2P music sharing may be driving people away from CD's there is something else that drove me and other away, and that was free streaming music. I was thinking about this a couple months ago, when I was at the used CD store buying an album, the first album I had purchased in almost three years. My thoughts went like, wow I haven't purchased a CD in 3 years, ever since I started listening differnt streaming music stations. I pick the station to fit my mood and off I work.
Just my 2 cents.
You forgot to add, that after looking at hotmail and discovering that it has anti-virus capabilities MS has clearly ripped off Google again. Why can MS innovate? Stupid copy cats. They have a crappy business model that will be destoryed by Google. Google stops MS for using it's monopoly power to force users from using hotmail anti-virus. Yea Google, MS sucks.
On a realted topic has anyone ever heard or done wikipedia pulluting? I have several friends who claim to purposfuly introduce micro-lies into wikipedia, I'm often sent links to stories that have twists they claim to have inserted. It's always little things such as, when Political Person X goes to Europe they always bring a 32 pack of Snickers because they think European candy is horrible. I've never gone back to check and see if it was removed or not, so I don't know if it's getting cleaned as fast as it's entered. Has anyone here ever done it for kicks, or know of anyone who has?
I did the same thing 6 years ago, with a web cam on a windows 98 box, using Visual C++. So maybe you could say that Linux caught up with 6 year old windows technology, just kidding, although I'm sure I'll get plenty of anti-Microsoft snaps. I've also done many another web cam projects, center a web cam on a moving target, point a web cam at a speaking target, point a laser pointer at a moving target, recognive text on a white board as people write making a virtual white board.
All in all this sort of thing isn't that hard or complicated.
SMT isn't the straight up performance gimme it is often marketed to be. In some instances SMT is good, and some aspects of SMT are good most of the time. but SMT introduces issues that can sometimes trash performance. For instance If you have mutliple threads acting in a processor core you suddenly have to worry about register names, this is accomplished either by using register renaming or multiple sets of registers, both solutions adding complexity to the decode time of the chip. Another issue is cache thrashing. If multipe threads exercise the same cache lines you can get into a situation where everytime a thread switches in it invalidates the cache and then switches back out, there are multiple solutions to this but it's just one more thing that needs to be placed into your logic. So now lets say you have addressed every issue with SMT, many of which I haven't bothered to start to address, you have more gates which means longer delays, and if it's in a critical path that means a lower frequency. Accessing different caches or busses might take an extra cycle. So now you plug all this data into your "master performance equation" and you find that overall there isn't that much of a boost if any at all, or at least not enough to justify the cost of a bigger package, lower yield, a higher power consumption, and more heat. Which is why chips like the Power5 have SMT they are a pedal to the floor performance machines so you deal with the drawbacks because it lets you get a 1% more done. Which yields the question if there are so many drawbacks how does Intel do it with HT, well they really don't do it, at least not all of it. HT uses some aspects of SMT in order to get some performance boost while not implementing others, your classic engineering decision.
If you read the article you would have noticed that these two cases are one in the same. The early ruling was one part of the claim this ruling is the second part.
>However, that ruling, issued in December, left open for trial the question of whether the use >of the trade marked terms in the text of sponsored ads breached GEICO's rights.
Slashdot Mirror
I was thinking the same thing and I re-read it. I think the newly arrested individuals stole over 2 mil, the whole organization has probably stolen quite a bit more.
It's really not a resounding no to the merger, the basicly said we won't support any sort of deal if MSFT lowers the offer. They also said if you pay an extra $1 a share we'll support your takeover 100%.
Good point, there is a minimum threshold for crap where you are better off not releasing it. I sit in on a lot of high level planning meettings, and the sales/marketing teams are actually scary accurate with their projections unless an X factor jumps out that makes your product less relavant.
Unfortunatly sales/marketing has to be the people setting deadlines to a large extent it's up to engineers and programmers to push back when the objectives are horribly unreasonable. As much as it sucks to us technical people companies don't stay in business by making superior products, they stay in business through superior sales. A buggy half assed product delivered early will yield by far more sales than a complete product will that comes to market a couple quarters later.
For enterprise class storage this seems pretty reasonable. the two high end drives which only produce ~150 IOPS are 146GB FC and 300GB FC these drives typically run in 1k-2k range, although they can be purchased cheaper then that someplaces. So 852GB is ~6 146GB drives, or about 6k on the low end, which put the price on par for a superior performing product. Now add in the power savings from having fewer drives, using a more power efficient technology, using less cooling for the disk. Look at the money saved on drawers, extra FC cables,SFPs, and actually fewer enterprise level disk systems because in all Enterprise level systems the bottle neck is the number of spindles. Big pictures the flash drives will probably be cheaper up front, provide significant power savings( Go Green ), and reduce very expensive data center footprint. I'm actually surprised there isn't a big flash based data enterprise class storage solution out there yet.
How old is your 360. I have an original release 360 and it's a little loud. My friend baught one a couple months ago and it's like night and day on the noise issue.
The only two companies in that list that shouldn't be merged are Microsoft and Google. The other comapnies, Exxon, AT&T, and either Microsoft or Google could merge no problem. Event though they are huge, they don't have that much overlap in business and if the combined Googattxon tried to do something like block other search engines from AT&T network or something else they would shoot themselves in the foot so bad they would start falling apart immediatly. There are very few conglomerates that really opperate efficiently and just because you are the biggest oil company and the biggest phone company you don't really gain that much more leverage then you had before.
IBM didn't decide to switch from a lease model to a sales model they were forced to switch to it as a result of the anti-trust lawsuit that was filed against IBM.
The link you posted lists alternative fuels as sales and use tax free not fuel tax free. The fuel tax in NC was suppose to be used maintance and expansion of the roads. Since NCs fuel tax is part flat fee and part percentage they have actually been running surplus with the high gas prices, which they have been using to pay down some of the education bonds and teacher raises.
I have, most of the time not intentionaly mind you, but there have been several occasions where I mistyped a domain got the page o' ads and then a window from another program pops up I click somewhere and end up accidentally clicking on an ad. There are other times where I'll go to some URL like www.aspecificplace.com the ad screen pops up and the first ad says the official web page of specific place, so I give it try, A good potion of the time 80% plus, it's the right palce. The other thing is how many people do you really need to click to be sucessful. You can get a domain for $9/year, let's say the ad get's you $.50, get 18 people to click on it and you paid the domain registration. So all you need is 2 mistaken or shot in the dark clicks a month and you're making money. If you get a couple a day you're making several hundred a year. I do know people who will go to an ad trap and click on each item, all 20+ of them thinking one click might take them where they want to go.
Trolling for a fight much. I was just stating a couple facts. People need to do what people need to do, but here are a couple of more facts for ya. Dell got 242 million in incentives, wich is estimated to cost the States 72 million over 20 years, so yes the state will get some of it back through new taxes and rest absorbed by the tax payers. What I find interesting in the Dell deal was that the next nearest incentive package totalled 37 million, 6.5 times less then the NC deal. So you might say that over a 30 year window it might pay off, and it might if Dell doesn't skip town the moment there is a better deal, which if you look at past history is pretty much what they do. The other question you have to ask is in this case might 242 million been better spent elsewhere, incentives for smaller businesses maybe, businesses without 6 billion in cash. I'm not opposed to states, towns, counties offering incentives but incentives that cost the taxpayers millions of dollars need to be looked at closely, which I'm sure you would agree with.
I'm sure there other other reasons besides tax breaks, and you make a good point that most everywhere some sort of incentive is going to be offered but NC gives huge incentives, look at the Dell deal I mentioned before. The total incentive package is $242 million, the next closest bid was $37 million by a county in VA, that's 6.5 times the amount of incentive, so while everyone offers incentives NC offers huge ones.
My bet it has more to do with states that offer massive tax breaks to businesses for moving in. I know North Carolina is famous for it, especially with the new Dell facility in Greensboro and Lenovo in RTP. The Dell deal was so good the state could have employed everyone hired by Dell for 11 years with the tax breaks and loans.
Is comparing MTBF correct when saying different drive types are of the same reliability? By looking at an existing system you really aren't looking at idependant variables. Let's say I have two servers one that hits the drives alot and another that barely touches the drives so using what I'll call "the old rules of thumb" I would put the FC drives on the intensive server and SATA drives on the less intensive server. So after 1 year I get my first drive failure on each. One could conclude that SATA is as reliable as FC, but is it really? I setup my environment to more heavily hit the FC drives, when would the SATA drive have failed if I had placed SATAs where the FCs were? The only way to really compare drives would be to hit a large number of each different drive type with the same workload. If you look at the way most places do tiered storage they'll put highly accessed data on an FC tier and then migrate less used data to a SATA tier, this might be one reason why the failure rates of the two drives look the same.
BlueGene runs on PPC440 cores not PPC405 cores. To the best of my knowledge 405 cores were never used in mobile phones. Most mobile phone software is designed to run on ARM processors and PPC and ARM code tend not to translate back and forth to each other very well. Never mind the fact that most of the mobile phone peripherils are designed to work with the AMBA bus and not the embedded PPC's(ePPC) PLB bus. Maybe you are referring to ePPC cores being used in chips for cell phone base stations.
IBM's older super computers were based on Power5 Technology, so IBM did use it in some of their most advanced computer systems.
While power and heat are very important chips like the Power5 are very important even though clusters of lower performance chips can get massive parallelization. Some application can be parallelized so your performance ultimatly becomes that of your fastest processing unit. So Power5 based systems work on entirely different problem sets then BlueGene.
One of the definitions of pod is "Something resembling a pod, as in compactness. " IANAL but I did recently take a class on patents, trademarks, and copyrights and one of the things we discussed was that trademarks that include generic terms had trouble being enforced. We specifically talked about Krispy Kreme. Krispy Kreme's trademark is specifically on Krispy Kreme with K's. They have tried to sue several people for selling products, even donuts with the name crispy cream, but when business have fought back the courts have always pretty much ruled that it wasn't a violation of trademark and the only way it would be was if the company specifically called it a Krispy Kreme something or other.
That being said Apple is probably just trying to cast a wide net in trying protect it's iPod trademarks. I think that the precedent set by other products( I have a Game Pod sitting on my desk that I got back in 98 that has 40 different card games) that pod generically defines any product that is compact in nature.
Depending on the situation, yes I would trust the engineer if it's SOP. I have a processor manual in front of me, and for some reason it has more commands then shift, and, or, and xor. It has an int mult command, that only take two clock cycles, so any mult done by any other means except a bit shift power of two mult is going to take longer then just executing the mult command. You don't need to know how every low level function works, that's how we make progress. We use the work that other people have done and build off of it. Granted there are specialty positions where in depth knowledge of the use of processor level instructions is necessary, and these people should know how to do that but they learned those skills through special classes, such as a compilers class, an embedded programming class, or etc etc etc. You could also take this to the next level, you're not a computer programmer unless you know how the underlying logic works to execute an add or shift, or you're not a computer programmer unless you know how to design a CMOS circuit in a particular technology to implement a shift register. To say anyone that doesn't know how to do assembly level add-shift multiplies or basic logical operations is just a secretary pushing buttons is wrong. I know many a gifted programmer who write awesome code that does awesome things, and have no idea how this works. I write device drivers on occasion and I need to do things like this on occasion for performance, but I do it so other people don't need to think about how the real interface works. I write the low level code so some other person can write the higher level stuff.
I see what you're saying and some people need to learn how to do the add-shift multiplier, I believe these are your Computer Engineers today, not your software engineers, they tend to be the ones involved in you're low level voodoo, but think of it like this. 20 years ago when you were in school you spent 4 years learning how to program in C and assembly in order to execute add-shift multipliers. Today students still go to school for 4 years but have so much more to learn about. Where is the time for object oriented programming techniques, advanced data structures, multi-threaded software design, comm. theory, vector programming, and everything else if people are spending their time learning about add-shift multipliers or whatever archaic multi-step register operation you can think of.
Using this logic GMail should be disposed of for Yahoo, which had about 6x the number of results.
The CD will go away eventually, but what doesn't. While iTunes and P2P music sharing may be driving people away from CD's there is something else that drove me and other away, and that was free streaming music. I was thinking about this a couple months ago, when I was at the used CD store buying an album, the first album I had purchased in almost three years. My thoughts went like, wow I haven't purchased a CD in 3 years, ever since I started listening differnt streaming music stations. I pick the station to fit my mood and off I work. Just my 2 cents.
You forgot to add, that after looking at hotmail and discovering that it has anti-virus capabilities MS has clearly ripped off Google again. Why can MS innovate? Stupid copy cats. They have a crappy business model that will be destoryed by Google. Google stops MS for using it's monopoly power to force users from using hotmail anti-virus. Yea Google, MS sucks.
On a realted topic has anyone ever heard or done wikipedia pulluting? I have several friends who claim to purposfuly introduce micro-lies into wikipedia, I'm often sent links to stories that have twists they claim to have inserted. It's always little things such as, when Political Person X goes to Europe they always bring a 32 pack of Snickers because they think European candy is horrible. I've never gone back to check and see if it was removed or not, so I don't know if it's getting cleaned as fast as it's entered. Has anyone here ever done it for kicks, or know of anyone who has?
I did the same thing 6 years ago, with a web cam on a windows 98 box, using Visual C++. So maybe you could say that Linux caught up with 6 year old windows technology, just kidding, although I'm sure I'll get plenty of anti-Microsoft snaps. I've also done many another web cam projects, center a web cam on a moving target, point a web cam at a speaking target, point a laser pointer at a moving target, recognive text on a white board as people write making a virtual white board. All in all this sort of thing isn't that hard or complicated.
SMT isn't the straight up performance gimme it is often marketed to be. In some instances SMT is good, and some aspects of SMT are good most of the time. but SMT introduces issues that can sometimes trash performance. For instance If you have mutliple threads acting in a processor core you suddenly have to worry about register names, this is accomplished either by using register renaming or multiple sets of registers, both solutions adding complexity to the decode time of the chip. Another issue is cache thrashing. If multipe threads exercise the same cache lines you can get into a situation where everytime a thread switches in it invalidates the cache and then switches back out, there are multiple solutions to this but it's just one more thing that needs to be placed into your logic. So now lets say you have addressed every issue with SMT, many of which I haven't bothered to start to address, you have more gates which means longer delays, and if it's in a critical path that means a lower frequency. Accessing different caches or busses might take an extra cycle. So now you plug all this data into your "master performance equation" and you find that overall there isn't that much of a boost if any at all, or at least not enough to justify the cost of a bigger package, lower yield, a higher power consumption, and more heat. Which is why chips like the Power5 have SMT they are a pedal to the floor performance machines so you deal with the drawbacks because it lets you get a 1% more done. Which yields the question if there are so many drawbacks how does Intel do it with HT, well they really don't do it, at least not all of it. HT uses some aspects of SMT in order to get some performance boost while not implementing others, your classic engineering decision.
If you read the article you would have noticed that these two cases are one in the same. The early ruling was one part of the claim this ruling is the second part.
>However, that ruling, issued in December, left open for trial the question of whether the use
>of the trade marked terms in the text of sponsored ads breached GEICO's rights.