That'd work for people who fast forward through commercials, but with my DVR I "jump" 45 seconds ahead, so I don't see the commercial at all. For a commercial break, I hit the button 5 times, if the show isn't back on, I'll hit a 6th time, and rarely I hit it a 7th time. I do get to see a 1 second portion of the commercial between the 5th-6th presses though, just long enough for me to recognize it's not the show I'm watching.
The EU asked for API Documentation. Microsoft gave away the entire source code, and that STILL isn't enough for them. They produced some 11,000 page document to fully document the APIs, but it wasn't enough. Believe what you want, but anything Microsoft does isn't going to be enough to satisfy the EU.
Incase you missed it, Microsoft just got sued in court for trying to support PDF in office. If you wanted Microsoft to use PDF as a viable alternative, go complain to adobe who initiated the lawsuit.
If you get sued for trying to support an open standard like PDF, what alternative do they have?
It'd help games like: Sony Everquest 2 Tom Clancy's Ghost Recon Advanced Warfighter Microsoft Rise of Nations 2: Rise of Legends Oblivion with all the settings on high
Any of those games trying to run with the eye candy on, at my monitors native resolution (1920x1200) will cause the framerates to drop into the sub-teen FPS on even high end graphics cards. Now if they can just make it so that you don't need a SLI chipset, that'd rock. I'll never get quad-sli, but it sure would be nice to be able to drop one of these in a non-sli machine. So far, I've heard that they only run in SLI-capable motherboards, which is a shame.
* PC2-6400: DDR2-SDRAM memory stick specified to run at 400 MHz using DDR2-800 chips, 6.400 GB/s bandwidth
this initial offering of XDR has half the throughput of already available PC2-6400!
The initial offering of XDR happened long ago. And it was (and still is) available in 800Mhz clock rate, dual bank (also called 16 lanes) configurations giving 12.8 GB/s bandwidth. That was over a year ago. OCZ just released thier PC2-6400, at half the speed of what was available to XDR1 a year ago. On top of the raw speed differences, since XDR uses less traces than DDR and the trace lengths don't need to be all of the same length, you can put 2 sticks of these as easily as you can route a single bank of DDR, or 4 interleaved sticks as easier than you can 2 interleaved banks of DDR, for another doubling of throughput in real world scenarios. BTW, XDR1 is what is powering the Playstation 3.
"DDR SDRAM, introduced in 2000, operated at an effective clockspeed of 266 MHz and delivered 2100 MB/s over a 64-bit bus using a 184 pin DIMM form factor."
It would have been squashed if Dual-bank RDRAM 800 wasn't already mainstream delivering 3200 MB/s at the time. While RDRAM was more expensive (About 5%, not the over "double" some people quote), motherboards were less expensive to make and produce because there wasn't a need to have nearly as many traces running to and from the ram banks, and they didn't have to be all the same length. Many times after purchasing an entire system, the RDRAM based solution was $10 more than the SDR/DDR based one (Out of $1400), and they ran significantly faster in most applications (10-20% faster).
You can say what you want about Rambus, because the company did do some unethical things, but RDRAM was and still is a vastly superior product technically to SDR/DDR. The current RDRAM is called XDR2, and has over 10 MB/s bandwidth per bank. Over 20 MB/s if you have dual banks. DDR isn't even in the same ballpark. Never was.
Rambus was never easily twice the price as comparable SDR ram, except in really strange sizes that noone would have bought (I forget if it was the really low low end, or the really high high end of which SDR didn't even have a solution for at the time). I bought many systems with RDRAM 800, and a couple based on SDR where the only difference in the systems were the ram used. There was a HUGE performance difference between the two systems, a very noticable one.
The article makes a mistake when it mentions that DDR was available, but they neglect to mention that at the time DDR came out (With 2100MB/s bandwidth), double-pumped, 1066Mhz RDRAM was already available (4264MB/s) and quad-pumped double banked sticks were working in prototype (17056MB/s). Unfortunately by this time Rambus was already pretty much dead in the PC market, after all the bad publicity and never had a mainstream chipset support the advances beyond single pumped 1066Mhz.
Actually, most people I know don't consider "Total Cost of Ownership." That's a term made up by Microsoft in an attempt to make FOSS proponents look like they're narrow-minded and that their conclusions were incomplete and "irrelevant to business." Everybody I know looks at "cost" - period. "Cost", by definition, without any modifiers, *must* mean total cost. "Partial cost" or "license cost" may mean something other than Cost, capital C.
Actually, the term TCO was made up by the Gartner Group in 1987. Cost by definition, without any modifiers, *must* mean cost. Any other assumption is wrong, and most likely just used to try and prove a point that you believe. Either you had no better proof, or you made the poor choice of using a false idea as a backup to your supposid truth.
You are of course, welcome to your own opinion. Everyone has one, but stating things that are either opinions or wrong as undeniable truths/must-be/can't-be-anything-buts doesn't help prove anything other than how bad your logic is, and why what you say can't be trusted.
That's not the way it works. It's those that want to deviate from the norm/corporate standard that must demonstrate the benefits/cost of an alternative, and in many cases it must VASTLY outweigh the standard for it to even be considered.
Something being "Free" in itself doesn't make a solution the best one (or default). If I offered your company "Free" lunches for all your employees everyday, would they hire my services? What if I charged $5000 per plate I had to wash (and you have to use me for dish washing)? "Free" doesn't mean anything unless you also consider all the costs involved with such a purchase. This is something companies are very used to doing, and something that FOSS doesn't do particularly well on in most cases.
The fact is that linux is NOT ready for the desktop. It's come a long ways, and it will get there eventually, but it isn't. Without it being on the desktop, IT shops then have to decide whether it is worth having a single OS throughout the company (Windows) for all desktops, laptops, servers, and even some PDAs, or would it be more cost beneficient to have some machines linux and some machines windows. This entails having two completley different set of tools (Backup, maintenance, diagnostics, imaging), and atleast one more person in the IT department that is a linux person that can support, configure and maintain the linux machines. Then you have to retrain all the people that might need to use the servers (Network admins, email admins, web server admins, etc). You will either have to move some stuff to OS-agnostic software packages, or deal with the inevitable higher maintenaince of dealing with the interoperability issues. Usually, it does right at the point you say it's FREE (But we have to hire another $40,000-$80,000 IT guy to support it). At $30-$50 a license for Windows that most mid/large companies pay for a copy of Windows, it'll take a LOT of licenses to justify that single additional person. Not to mention that in a time when companies are trying to focus on their core products, it's a rather hard sell to try and explain why you need to expand you IT department to support something that has nothing to do with your product at all. Most accountants/CFO's see this a dilution of ROI, which is not a good thing.
I'm not against linux. It has it's place and it has it's uses, but advocate the right tool for the right job when appropriate. There is a lot more than just what is technologically "better" when deciding on what platform(s) a company uses. All good companies keep an eye on what will make it profitable (or more profitable), not necessarily what costs the least to buy, or even what costs the least in TCO.
Go with the IDE. Unless the purpose of the course is about how to get things to compile, and how to use outdated editors (Like vi), dropping them on the command line will just turn your course from Java/Python into a unix/linux 101 class instead. Overburdoning your students with cryptic command line syntax (For a specific compiler and editor -- of which they will probably use neither of ever again) doesn't really do your students that much good, and lands up detracting from them learning the language itself.
I'm all for teaching basic command line skills to those that actually want to do it for a living, but that's a different course entirely.
TCP/IP isn't Open Source. You may classify it as a standard, but it definately isn't open source. TCP/IP was developed by the US military for DARPAnet (the precursor to the internet), which wasn't open either.
Of course you may be talking about the sockets API. But then again, that was developed by AT&T for Unix (Commercial development). So I'm sure you have a point, but TCP/IP both the protocol, the stack, and the API's are all examples of commerical (or closed source) development.
I would say by the definition that the "best" are the ones that are able to commit their crimes, cover their tracks (either through not allowing anyone to track their activities back to them, or better, to not even allow those who have been attacked be aware of the crime at all).
Those who have been identified, aren't that good. They got caught. Being able to commit a crime is easy. Being able to commit it, and get away with it isn't (I don't consider being hunted by the FBI "getting away with it").
6-18 months? Are you kidding me?
After I left my long term job I had for 10 years, I put my resume up. I had 3 phone calls that night. Two interviews the next day, and the following day I had a short term contract.
When that contract was up, I put my resume up again. That night I had 2 phone calls. Two interviews the next day with consulting companies on two 6-months contract to hires. The next day I interviewed with one of them, and had an offer that evening. I turned the second job offer down.
3 job changes, longest time out of work was about 45 hours.
Sorry, I think most of us figured this out by the time we were 18. It was also covered in my High School economics class, and a vast number of classes I took in college (Economics, Business, Accounting, and Business Ethics).
While I will probably get modded down to hell:
1) I'm not going to comment, as this is just FUD.
2) I fail to see how this is Microsoft's problem. Go yell at the Open source guys to make their stuff disability friendly. If word wasn't around, guess what? OpenOffice, etc STILL wouldn't be disability friendly. Lastly, Microsoft hasn't refused to implement an open standard. Word can read and write in all kinds of open standards just fine and has for years --.TXT (Plain text, arguable the most standard of all),.RTF (Rich text format),.XML (eXtensable Markup Language),.HTML (HyperText Markup Language), and now they are implementing two *MORE* open standard document formats in Office 12. Office 12 will support PDF (Portable Document Format), and MOOX. Clearly the problem isn't what you are complaining about at all, it's that they haven't implemented the one standard YOU want.
That would be relevant, *IF* it was the US. It was a NATO attack, run by england's general, using english pilots. And just to top it off, it was an accident that it happened. That's hardly proof that even Europe/England don't "worry" about such things.
Ok, it's apparent that I might have to clear up a few things first.
Netscape when it was the dominant browser was the one who started the whole free browser thing. It's wasn't Microsoft. Secondly, they were never found guilty of this in court, if you notice, the remedy in both the US and EU cases, neither of them requested, required, or even really inquired (much) about IE needing to be unbundled from the OS. They talked about it for a total of 2 days out of the kazillion days of the trial, and the courts SIDED with Microsoft on that issue.
They got hammered for not allowing Netscape to be pre-installed on computers coming from manufacturers, or the so-called "Start up" experience.
I'd also have to argue that browsing and searching the web are both related and in the same "market". If browsing HTML and being able to search HTML pages isn't the same market then sue adobe for letting their PDF viewer only use their search to search within a PDF file, baring entry to all those other PDF-search plug-ins that got stiffled right out of existance before they even started.
Google also doesn't compete in the "browser integrated search" market. They only offer HTML form based searches.
The above poster obviously doesn't fully understand DHCP either. The IP pool for many scenarios are much smaller than their installed base. Dial-up users and DSL users included. You only have to have as many IP Addresses available in the pool as you can have online at the same time. And the DHCP server can reuse any unallocated address, it does not have to necessarily reuse them in longest since has been allocated strategy.
Besides the technical issues, I agree with the above poster in general. The article is full of crap, as even for those people with dial-up, a single user may (Will most likely) use more than 1 IP, it's also quite possible the opposite is true. A bank of modems (reusing the same IP), where multiple people dial-in during different parts of the day and get the IP reused will only be counted as a "single" unique user. Trying to track unique visitors by IP is only a ballpark figure, and a VERY large ball park at that.
The point of the story is this: Linux doesn't have the problems that Windows has, because it's more secure by design - not by luck.
So if I write a crappy program that won't run on linux unless you are the root user, or have root privs then this proves linux is secure by luck?
Your story proves nothing other than that software was written poorly, and the average user is stupid. I don't think we needed proof of the latter, plenty of good examples around.
They should fine them for everything they have PLUS $100,000. Then employ some of the ruthless collection agencies to try and collect the $100,000. See how they like getting calls starting at 8am until 9pm at home and work.
Heh.
Comcast doesn't use fiber, they use cable. Verizon with their FIOS service uses fiber, which is current 15Mb down/2Mb down. Cable in my area is currently 8Mb down/768k down, and some some areas it was annouced they'll get 16/1. DSL on the other hand is usually 1.5 down/384k up, sometimes 3.0/768 if you live real close to your central office.
1080p isn't even close to equivalent to 35mm film. 1080p widescreen format is 1920x1080 resolution which is approximately 2 Mega-Pixels of information. 35mm film is rated between 8MP and 20MP depending on circumstance.
You are probably confusing the latest HD resolutions with the latest consumer camera resolutions. Consumer cameras have just started to hit the range where they are more accurate than film (12+ MP), which is causing a lot of professional photographers to finally make the switch.
Actually, the original film stock usually does have enough information for "HD Quality". The problem is that going back to the original film stock is extremely expensive, because it hasn't even been editted. No special effects have been applied, the CGI stuff hasn't been rendered into the scene, etc.
When the original films where transferred to the digital editors, they were converted at that point to digital, then the film was editted, etc. All that would need to be completely redone, and it's not cheep to do right (It's also labor intensive -- meaning even if it was cost effective, you're going to be quickly limited by how many a studio can redo quickly, without causing so much extra demand that prices skyrocket). They explained the whole process in the limited HD version of Terminator 2, and why they have to go back all the way to the original film stock in order to get HD-quality output.
The whole thing about "HD cameras" isn't really for a better picture from the camera. It about how fast they can review/edit and the quality of the picture at the very end when the film loses quality in each A/D conversion (And apparently there are a few that have to be done in most studios). The new HD cameras record the movie in digital and it no longer needs to be converted at any step along the way. It's also much faster to go "All digital", but it's more costly. On the technical side, I'd have to say that's not 100% correct. I'm sure they could get new analog equipment that could also process/work with analog film and/or convert it from analog to digital just as quickly, but, why would anyone spend that kind of money when if you have to spend it, it just makes sense to go 100% digital.
No. Most modern disk utils let you choose between 512b, 1k, 2k, 4k, 8k, 16k, 32k, or 64k CLUSTER sizes. All current drives come pre-formatted from the factory (ATA, SATA, and SCSI), and you can't what is called "low level format" the drive any more. What the OS calls "Formatting" for hard disks is no more than writing the file system structure onto the disk (And writing 0's to all unsused portions of the disk, checking for bad areas).
This change isn't really that big of a deal. You'll get a slight bit of performance increase, and a little bit better disk space (Not all that much). Since most people use the default cluster size of 4k now anyhow, NO EXTRA SPACE WILL BE WASTED. And to the guy who complains that he wants to store 2 billion 13-byte files on his drive, well, those 13-byte files won't be affected anyhow since NTFS stores them directly into the directory structure. It's the files that are 128 bytes to 256 bytes that waste the most space. And if you have THAT many small files, you've got other issues. Buy old drives, use a different file system (Like FAT32), or just get more drive space.
Slashdot Mirror
That'd work for people who fast forward through commercials, but with my DVR I "jump" 45 seconds ahead, so I don't see the commercial at all. For a commercial break, I hit the button 5 times, if the show isn't back on, I'll hit a 6th time, and rarely I hit it a 7th time. I do get to see a 1 second portion of the commercial between the 5th-6th presses though, just long enough for me to recognize it's not the show I'm watching.
The EU asked for API Documentation. Microsoft gave away the entire source code, and that STILL isn't enough for them. They produced some 11,000 page document to fully document the APIs, but it wasn't enough. Believe what you want, but anything Microsoft does isn't going to be enough to satisfy the EU.
Incase you missed it, Microsoft just got sued in court for trying to support PDF in office. If you wanted Microsoft to use PDF as a viable alternative, go complain to adobe who initiated the lawsuit.
If you get sued for trying to support an open standard like PDF, what alternative do they have?
It'd help games like:
Sony Everquest 2
Tom Clancy's Ghost Recon Advanced Warfighter
Microsoft Rise of Nations 2: Rise of Legends
Oblivion with all the settings on high
Any of those games trying to run with the eye candy on, at my monitors native resolution (1920x1200) will cause the framerates to drop into the sub-teen FPS on even high end graphics cards. Now if they can just make it so that you don't need a SLI chipset, that'd rock. I'll never get quad-sli, but it sure would be nice to be able to drop one of these in a non-sli machine. So far, I've heard that they only run in SLI-capable motherboards, which is a shame.
It would have been squashed if Dual-bank RDRAM 800 wasn't already mainstream delivering 3200 MB/s at the time. While RDRAM was more expensive (About 5%, not the over "double" some people quote), motherboards were less expensive to make and produce because there wasn't a need to have nearly as many traces running to and from the ram banks, and they didn't have to be all the same length. Many times after purchasing an entire system, the RDRAM based solution was $10 more than the SDR/DDR based one (Out of $1400), and they ran significantly faster in most applications (10-20% faster).
You can say what you want about Rambus, because the company did do some unethical things, but RDRAM was and still is a vastly superior product technically to SDR/DDR. The current RDRAM is called XDR2, and has over 10 MB/s bandwidth per bank. Over 20 MB/s if you have dual banks. DDR isn't even in the same ballpark. Never was.
Rambus was never easily twice the price as comparable SDR ram, except in really strange sizes that noone would have bought (I forget if it was the really low low end, or the really high high end of which SDR didn't even have a solution for at the time). I bought many systems with RDRAM 800, and a couple based on SDR where the only difference in the systems were the ram used. There was a HUGE performance difference between the two systems, a very noticable one.
The article makes a mistake when it mentions that DDR was available, but they neglect to mention that at the time DDR came out (With 2100MB/s bandwidth), double-pumped, 1066Mhz RDRAM was already available (4264MB/s) and quad-pumped double banked sticks were working in prototype (17056MB/s). Unfortunately by this time Rambus was already pretty much dead in the PC market, after all the bad publicity and never had a mainstream chipset support the advances beyond single pumped 1066Mhz.
Actually, the term TCO was made up by the Gartner Group in 1987. Cost by definition, without any modifiers, *must* mean cost. Any other assumption is wrong, and most likely just used to try and prove a point that you believe. Either you had no better proof, or you made the poor choice of using a false idea as a backup to your supposid truth.
You are of course, welcome to your own opinion. Everyone has one, but stating things that are either opinions or wrong as undeniable truths/must-be/can't-be-anything-buts doesn't help prove anything other than how bad your logic is, and why what you say can't be trusted.
That's not the way it works. It's those that want to deviate from the norm/corporate standard that must demonstrate the benefits/cost of an alternative, and in many cases it must VASTLY outweigh the standard for it to even be considered.
Something being "Free" in itself doesn't make a solution the best one (or default). If I offered your company "Free" lunches for all your employees everyday, would they hire my services? What if I charged $5000 per plate I had to wash (and you have to use me for dish washing)? "Free" doesn't mean anything unless you also consider all the costs involved with such a purchase. This is something companies are very used to doing, and something that FOSS doesn't do particularly well on in most cases.
The fact is that linux is NOT ready for the desktop. It's come a long ways, and it will get there eventually, but it isn't. Without it being on the desktop, IT shops then have to decide whether it is worth having a single OS throughout the company (Windows) for all desktops, laptops, servers, and even some PDAs, or would it be more cost beneficient to have some machines linux and some machines windows. This entails having two completley different set of tools (Backup, maintenance, diagnostics, imaging), and atleast one more person in the IT department that is a linux person that can support, configure and maintain the linux machines. Then you have to retrain all the people that might need to use the servers (Network admins, email admins, web server admins, etc). You will either have to move some stuff to OS-agnostic software packages, or deal with the inevitable higher maintenaince of dealing with the interoperability issues. Usually, it does right at the point you say it's FREE (But we have to hire another $40,000-$80,000 IT guy to support it). At $30-$50 a license for Windows that most mid/large companies pay for a copy of Windows, it'll take a LOT of licenses to justify that single additional person. Not to mention that in a time when companies are trying to focus on their core products, it's a rather hard sell to try and explain why you need to expand you IT department to support something that has nothing to do with your product at all. Most accountants/CFO's see this a dilution of ROI, which is not a good thing.
I'm not against linux. It has it's place and it has it's uses, but advocate the right tool for the right job when appropriate. There is a lot more than just what is technologically "better" when deciding on what platform(s) a company uses. All good companies keep an eye on what will make it profitable (or more profitable), not necessarily what costs the least to buy, or even what costs the least in TCO.
Go with the IDE. Unless the purpose of the course is about how to get things to compile, and how to use outdated editors (Like vi), dropping them on the command line will just turn your course from Java/Python into a unix/linux 101 class instead. Overburdoning your students with cryptic command line syntax (For a specific compiler and editor -- of which they will probably use neither of ever again) doesn't really do your students that much good, and lands up detracting from them learning the language itself.
I'm all for teaching basic command line skills to those that actually want to do it for a living, but that's a different course entirely.
TCP/IP isn't Open Source. You may classify it as a standard, but it definately isn't open source. TCP/IP was developed by the US military for DARPAnet (the precursor to the internet), which wasn't open either.
Of course you may be talking about the sockets API. But then again, that was developed by AT&T for Unix (Commercial development). So I'm sure you have a point, but TCP/IP both the protocol, the stack, and the API's are all examples of commerical (or closed source) development.
I would say by the definition that the "best" are the ones that are able to commit their crimes, cover their tracks (either through not allowing anyone to track their activities back to them, or better, to not even allow those who have been attacked be aware of the crime at all). Those who have been identified, aren't that good. They got caught. Being able to commit a crime is easy. Being able to commit it, and get away with it isn't (I don't consider being hunted by the FBI "getting away with it").
It'd allow the pilot to use the tail rudder to wave good bye.
6-18 months? Are you kidding me? After I left my long term job I had for 10 years, I put my resume up. I had 3 phone calls that night. Two interviews the next day, and the following day I had a short term contract. When that contract was up, I put my resume up again. That night I had 2 phone calls. Two interviews the next day with consulting companies on two 6-months contract to hires. The next day I interviewed with one of them, and had an offer that evening. I turned the second job offer down. 3 job changes, longest time out of work was about 45 hours.
Sorry, I think most of us figured this out by the time we were 18. It was also covered in my High School economics class, and a vast number of classes I took in college (Economics, Business, Accounting, and Business Ethics).
While I will probably get modded down to hell: 1) I'm not going to comment, as this is just FUD. 2) I fail to see how this is Microsoft's problem. Go yell at the Open source guys to make their stuff disability friendly. If word wasn't around, guess what? OpenOffice, etc STILL wouldn't be disability friendly. Lastly, Microsoft hasn't refused to implement an open standard. Word can read and write in all kinds of open standards just fine and has for years -- .TXT (Plain text, arguable the most standard of all), .RTF (Rich text format), .XML (eXtensable Markup Language), .HTML (HyperText Markup Language), and now they are implementing two *MORE* open standard document formats in Office 12. Office 12 will support PDF (Portable Document Format), and MOOX. Clearly the problem isn't what you are complaining about at all, it's that they haven't implemented the one standard YOU want.
That would be relevant, *IF* it was the US. It was a NATO attack, run by england's general, using english pilots. And just to top it off, it was an accident that it happened. That's hardly proof that even Europe/England don't "worry" about such things.
Ok, it's apparent that I might have to clear up a few things first. Netscape when it was the dominant browser was the one who started the whole free browser thing. It's wasn't Microsoft. Secondly, they were never found guilty of this in court, if you notice, the remedy in both the US and EU cases, neither of them requested, required, or even really inquired (much) about IE needing to be unbundled from the OS. They talked about it for a total of 2 days out of the kazillion days of the trial, and the courts SIDED with Microsoft on that issue. They got hammered for not allowing Netscape to be pre-installed on computers coming from manufacturers, or the so-called "Start up" experience. I'd also have to argue that browsing and searching the web are both related and in the same "market". If browsing HTML and being able to search HTML pages isn't the same market then sue adobe for letting their PDF viewer only use their search to search within a PDF file, baring entry to all those other PDF-search plug-ins that got stiffled right out of existance before they even started. Google also doesn't compete in the "browser integrated search" market. They only offer HTML form based searches.
The above poster obviously doesn't fully understand DHCP either. The IP pool for many scenarios are much smaller than their installed base. Dial-up users and DSL users included. You only have to have as many IP Addresses available in the pool as you can have online at the same time. And the DHCP server can reuse any unallocated address, it does not have to necessarily reuse them in longest since has been allocated strategy. Besides the technical issues, I agree with the above poster in general. The article is full of crap, as even for those people with dial-up, a single user may (Will most likely) use more than 1 IP, it's also quite possible the opposite is true. A bank of modems (reusing the same IP), where multiple people dial-in during different parts of the day and get the IP reused will only be counted as a "single" unique user. Trying to track unique visitors by IP is only a ballpark figure, and a VERY large ball park at that.
They should fine them for everything they have PLUS $100,000. Then employ some of the ruthless collection agencies to try and collect the $100,000. See how they like getting calls starting at 8am until 9pm at home and work. Heh.
Comcast doesn't use fiber, they use cable. Verizon with their FIOS service uses fiber, which is current 15Mb down/2Mb down. Cable in my area is currently 8Mb down/768k down, and some some areas it was annouced they'll get 16/1. DSL on the other hand is usually 1.5 down/384k up, sometimes 3.0/768 if you live real close to your central office.
1080p isn't even close to equivalent to 35mm film. 1080p widescreen format is 1920x1080 resolution which is approximately 2 Mega-Pixels of information. 35mm film is rated between 8MP and 20MP depending on circumstance.
You are probably confusing the latest HD resolutions with the latest consumer camera resolutions. Consumer cameras have just started to hit the range where they are more accurate than film (12+ MP), which is causing a lot of professional photographers to finally make the switch.
Actually, the original film stock usually does have enough information for "HD Quality". The problem is that going back to the original film stock is extremely expensive, because it hasn't even been editted. No special effects have been applied, the CGI stuff hasn't been rendered into the scene, etc. When the original films where transferred to the digital editors, they were converted at that point to digital, then the film was editted, etc. All that would need to be completely redone, and it's not cheep to do right (It's also labor intensive -- meaning even if it was cost effective, you're going to be quickly limited by how many a studio can redo quickly, without causing so much extra demand that prices skyrocket). They explained the whole process in the limited HD version of Terminator 2, and why they have to go back all the way to the original film stock in order to get HD-quality output. The whole thing about "HD cameras" isn't really for a better picture from the camera. It about how fast they can review/edit and the quality of the picture at the very end when the film loses quality in each A/D conversion (And apparently there are a few that have to be done in most studios). The new HD cameras record the movie in digital and it no longer needs to be converted at any step along the way. It's also much faster to go "All digital", but it's more costly. On the technical side, I'd have to say that's not 100% correct. I'm sure they could get new analog equipment that could also process/work with analog film and/or convert it from analog to digital just as quickly, but, why would anyone spend that kind of money when if you have to spend it, it just makes sense to go 100% digital.
No. Most modern disk utils let you choose between 512b, 1k, 2k, 4k, 8k, 16k, 32k, or 64k CLUSTER sizes. All current drives come pre-formatted from the factory (ATA, SATA, and SCSI), and you can't what is called "low level format" the drive any more. What the OS calls "Formatting" for hard disks is no more than writing the file system structure onto the disk (And writing 0's to all unsused portions of the disk, checking for bad areas). This change isn't really that big of a deal. You'll get a slight bit of performance increase, and a little bit better disk space (Not all that much). Since most people use the default cluster size of 4k now anyhow, NO EXTRA SPACE WILL BE WASTED. And to the guy who complains that he wants to store 2 billion 13-byte files on his drive, well, those 13-byte files won't be affected anyhow since NTFS stores them directly into the directory structure. It's the files that are 128 bytes to 256 bytes that waste the most space. And if you have THAT many small files, you've got other issues. Buy old drives, use a different file system (Like FAT32), or just get more drive space.