Yes, but Democrats spend money on things like the Department of Education, which has done nothing to stop (or has perhaps accelerated) a tremendous decline in the quality of public schooling in the US. This despite a $69.4 billion budget in 2006, with budget that increases faster than inflation every year. Thank you, Jimmy Carter. And the program is politically un-killable, because it is "for the children".
It's amazing, but Democrats can't seem to figure out that throwing money at social problems almost never solves them. Federal Social programs are just as much "pork barrel" as a defense contract, except the beneficaries are bureaucrats and unions instead of Lockheed-Martin. Giving money to a homeless drug addict on the street suprisingly doesn't magically get him into rehab. The addict just buys more crack.
At least the Republicans waste money on cool thinks like submarines and missles. They cost a lot, but they have at least some value (even if they're only sold a decade later on the arms market). Have you ever seen video of an MLRS lauch? As P.J. O'Rourke would say, "Now that's the way you waste taxpayer money!".
Of course, with the rise in popularity of Blended Wing Bodies and Waveriders, there is the question of why Boeing is even sill using the tube-with-wings design. It's inefficient, it's likely more prone to failure, and a good BWB airframe should be able to land more passengers on smaller runways, greatly increasing the number of people who could buy them.
The Boeing X-48 and X-51 are are the forefront of BWB and Waverider research respectively. Boeing is still using the tube-with-wings design because it is a known-safe configuration. BWBs, like the F-117 (sort of) and B-2, can require some fairly serious computer assistance to remain stable in low-speed flight. Boeing is not "behind the curve" as you suggest.
Safety trumps everything else in passenger aviation. Reading suggests the first non-military BWB you see will be used for cargo, not passengers. It will likely be decades before they are in production for passenger use, as some Googling reveals that neither Boeing, Airbus or any other company has a production BWB airliner on the drawing board.
Perhaps some maverick start-up will build a BWB airliner soon. But they better also start their own airline, because airlines are so risk-averse that they would never buy such a thing until it had decades of safe service in military or other markets.
Almost every processor is bechmarked using SPECint 2006 and (for multi-core systems) SPECint_rate 2006. Yes, it ignores some I/O effects, but it's not too shabby of a "universal" benchmark. We frequently use SPECint_rate scores to help us decide what sort of application and datbase servers to buy in a particular year.
Doing realistic benchmarks on all that disparate hardware with our own apps would be way too expensive and time consuming, so we just rely on SPEC to make the benchmarks "fair".
Wouldn't it be nice if the SPECint_rate score was the model number of a chip? You'd have the intel 2-180 or the AMD b193. You'd know the AMD was marginally faster than the 180 in total multi-threaded throughput.
So can you explain me why I can go around in all the countries of Europe just showing my passport (if requested) and if i even try to travel to the USA, i *must* give my fingerprints to the US cops?
Because you are not a US citizen, so you don't have any rights under the US Consitution. Visiting the USA is a privilege for you, not a right. I can travel freely inside the USA without any documentation whatsoever. Unless, of course, I get pulled over for speeding, in which case I'd better have a drivers license. Or if I need to get through security in the airport, in whcih case a Driver's License, State ID, or passport is required.
Really, the only way to travel freely in the USA these days is by train. But passenger train service in the US is almost non-existent.
In Trinity and Beyond: The Atomic Bomb Movie, film clips of several high-altitude detonations are shown. Carried atop various missiles, all of which are probably going much faster than mach six (vertically), the fireballs from these high-altitude detonations are still almost perfectly spherical until they start drifting inthe wind many seconds after detonation.
As I mentioned, the fireball is formed by radiation effects within about one microsecond, and its radius is several hundred meters at least. Given that in one microsecond, the Mach 6 plane would only travel about two milimeters, I do not think the fireball (and the resulting mushroom cloud if the detonation were near the ground) would be significantly deformed. Everything inside the fireball is essentially superheated plasma, and the blast wave, which creates the mushroom cloud by kicking up debris, radiates outward from the fireball. So the momentum of the vaporized plane would essentially be overwhelmed by the momentum added to all the particles inside the superheated fireball.
But this is all just semi-informed speculation on my part.
What would an H-bomb explosion at Mach 6 look like?
Almost exactly the same as a "traditional" H-Bomb airburst. The thermal pulse travels at nearly the speed of light, and would vaporize the plane almost instantly. So the shockwave and fireball that form would probably radiate from the detonation point at almost exactly the same speeds and shapes as a stationary airburst.
The actual fission-fusion part of a thermonuclear nuclear detonation is so short that the fact that the bomb is moving at high speed makes almost no difference. The added energy of a hurtling piece of machinery is inifitesmally small compared with the detonation energy.
now if those ibm blades only had video/audio-in...
Why in the hell would a blade server have video or audio capabilities? Servers are designed to sit in data center racks and just keep running. They're a bit expensive, noisy and strangely-shaped to use as AV workstations.
You may want a blade server with audio/video for some ungodly reason. The other 99.9995% of the server market does not.
it's been... wow... 15 years since i touched a original ibm made computer.
Considering IBM makes most of their money from software and services these days, IBM probably don't care a bit that the tech support guy from Circuit City is no longer an IBM customer.
isn't the most direct remedy for the American media consumer to listen to internet radio from provider outside the USA?
This isn't practical in an Internet without pervasive global multicast services. Do you think ISPs are really going to let their extraordinarily expensive trans-continental pipes be clogged with Internet radio?
And don't assume BitTorrent-like schemes solve the problem, as your friends and neighbors participating in a swarm would have the RIAA beating on their door for redistributing Internet Radio without paying license fees. BitTorrent is a clever hack to emulate multicast in the unicast Internet, but it has the undesirable property that participating nodes must store and publicly redistribute data blocks, thereby running afoul of the whole copyright regime.
By the way, you also should look a little more closely at the *chart* below the graph in the study I linked to. Linux is mentioned specifically, rather than just "other". "Linux"+"Unknown" = 0.75% market share at most. Vista has 3.74%, or 4.99 times the market share of Linux.
And that assumes all "Unknown" boxes are Linux, which is most definitely not the case. At least some are FreeBSD, Cable Boxes, random mobile phones, and those crazy airport pay-per-minute terminals.
Yes, your indeirect measure of RAM is certainly more relaible than one based on visitors to 40,000 LiveStats customer sites.
Here's a hint: one billion internet USERS != 1 billion WORKSTATIONS. The majority of internet users share a PC, wither with family or and an Internet Cafe.
Vista is not selling AS FAST AS MEMORY MAKERS EXPECTED, but it is by no means not selling. I see Vista laptops galore on the Chicago commuter trains, far more than OSX or Linux machines. In fact, my Ubuntu laptop is the only Linux laptop I've ever seen in a public place (IT conferences being an exception).
Your sample is probably a bit skewed if you're coming from an academinc environment. I was pretty shocked 10 years ago when I came out of college eleven years ago realized nobody used the Macs or Solaris worksations with which I was so familiar in the real world.
This study shows there are already 5 times as many Vista workstations in use versus Linux workstations. Assuming Microsoft has sold ~20M copies through May, that means there are far fewer than 20M linux workstations in active use. Unless you assume that more than 4/5 Vista installs is pirated.
Note my original statement should have been qualified to compare Vista with Linux workstations, and not all "Linux machines". Comparing the population of a workstation OS to a server install base doesn't make much sense.
What makes you think people are going to use Vista? There's no evidence of that to date. Vista has other larger issues than IPv6 that keep people away from it.
Twenty million licenses sold already is "no evidence"? That's more than all the active Linux and MacOS machines out there in just a few months.
I have Vista on my laptop, and it is quite an improvement over XP. The very few rough edges I've encounetered have been with support of oddball hardware, like my Sprint Wireless Broadband card, which didn't have good Vista drivers until recently.
There is a huge pain in the ass with corporate licensing and activation that will be a real barrier to business migrating existing machines to Vista, though. Most likely they will just wait and acquire pre-activated Vista with new hardware. Microsoft really kicked themselves in the balls with activation for enterprise customers, it is just horribly broken. We have not had one machine activate properly, and a phone call to the automated activation system is always needed as a result. Completely unwordkable for a company our size, and we only have a few hundreds of machines. So our ten guinnea pig machines will be the only Vista we run until the next hardware refresh.
Day 1321: Software stuck on version 0.4.00, almost no documentation available, will only work on a very specific version of Gentoo x86 without major code hacking. The only project developer has decided to spend time with family and work to make some money, so support forum posts go unanswered. There have been no commits for two years.
none of these holes work (or so people here claim, if you've got evidence to the contrary I'd be interested to hear it).
I think you meant "If so then a Linux distro is required for Firefox none of these exploits work". The holes (vulnerabilities) are almost certainly still there in the code base for FF on Linux, but the exploit code needs to be crafted differently for the non-Windows OS. It very likely that creating exploit code for the Linux versions of these holes is trivial.
I'll bet you the same exploit code wouln't work on the Itanium versions of Windows, either. But the holes are still there.
I think your solution is ingenious, in that it spreads disk seeks evenly. But I suspect it is probably bad idea for performance, because of how the partitions are physically allocated on disk.
Wouldn't the the disk heads be hopping across 1/2 of the disk the disk quite a bit, since they have to read from partition to partition? It might not matter with your workload, but I would be hesitant run a database server on such an array.
In my employer's environment, we usually do one partition per physical disk. Why? We benchmarked for SQL Server using MSFT's IOstress utility when setting up a new DB server. Having the database files spread over two arrays over 8 disks (with each disk partitioned into 2 parts, one for each array) performed significantly worse than simply using one big partition that occupied the whole array on the same 8 disks. Our theory was that while the same number of individual seeks had to happen in both cases, the seeks were much "longer" when the individual disks were partitioned.
Of course, that was about 5 years ago, so with all the changes in disk density, cache sizes, etc., it may not matter much anymore. Our DB servers now have 16GB or more of RAM, and keep everything "hot" in memory.
We had a VAX at work and the halt button was positioned such that just about ANYBODY working in the server room would manage to bump it. The thing probably cost more than a car and nobody thought to put a guard over the front-panel switches or at least recess them...
We have a number of pieces of equipment with that issue in our server room. Cleaning crews would hit the big red button on our AC flow panel about once a year. Our solution was to take a small plastic cap from an aeorsol can and tape it over the switch with gaffer's tape (no residue). Label it of course. Seems to work, and costs all of a few cents.
Unless you are mirroring it off-site, it is a single point of failure.
We do block-level replication off-site with our SAN, which is not an EMC, but a Left Hand Networks cluster. The reason we bought in was scalability. Our LeftHand units come in as small as 2 TB chunks, each with it own mobo and two Gigabit Ethernet adapters. As you buy more chunks, your storage pool automatically grows (and volumes automatically re-balance themselves amongst spindles). But even better, the aggregate throughput also increases, as each chunk adds its 2 GB/sec to the total SAN bandwidth available. Using iSCSI and and multi-patch IO, there are no single points of failure in the SAN at this site. In case of fire, we have our replicated data off-site (also handled by the SAN in a bandwidth-efficient manner), and also traditional backup-media off-site.
You just can't do that sort of stuff with a roll-your-own solution. If you don't have those sorts of needs, by all means, roll your own cheap solution.
What this should illustrate is that the majority of the value of a SAN is in the management software, not just the hardware. Which is probably why LeftHand has now partnered with HP to sell their iSCSI SAN software running on DL300-series machines.
Note, I do not work for or invest in LeftHand, I am just a satisfied customer. LeftHand's SAN/IQ operating system is Linux-based, which is neat. We were also thrilled with demos of EqualLogic's gear, which is quite similar.
The advertising stating that they get unlimited transfer, probably.
Except I have never seen "unlimited" service advertised to consumers by an ISP in Chicago, USA. I've seen things like "NO PER-MINUTE CHARGE", "24 hours a day", all with asterisks to read the fine print. But never "unlimited". And if you actually read the contract before you sign, it clearly states that the committed information rate is basically zero and all service is best-effort.
Where is all this "unlimited" false advertising? Can someone please post a link? Because to me it just sounds like a bunch of whining by kids that want free movies, free music, and almost-free internet access, too.
Do you know why Beta lost? Sony's licensing fees. They were extortionistic in cost, so JVC et.al. went out and created VHS.
No, Beta lost because it had NO PORN. Sony would not license to adult video types, byt JVC et. all would. Porn was the killer app for the VCR. Men buy electronics for the home, and VHS meant porn at home - no more worrying about being seen at one of the sticky-floored "adult film theaters" on the bad side of town (they were very common in the late 70s and early 80s, especially around Times Square in NYC).
MS tried with Alpha for a while, but noone bought it. So it looks like it's really our fault, not theirs.
Amen. Microsoft supported x86, MIPS, PowerPC, and Alpha with the first release of NT 4.0. Nobody bought PowerPC and MIPS, and very very few people bought Alpha. So by the time Win2k came around, Windows was x86 only. I really hoped Alpha or PowerPC would succeed and get us off the multilayered hack that is x86, but the masses did not agree with me.
The good thing about this debacle for MSFT is that today, almost evertyhing in the Windows kernel is platform-agnostic except for HAL.DLL and drivers. This fact enables the Itanium versions of Windows. It also means that Microsoft can produce a version for, say, POWER6 processors with a modest incremental investment, if they see market for it. One can only hope.
Not to mention The Tudors, which is a 10+ hour historical drama about a young Henry VII on Showtime that kicks ass.
Of course, I am still smarting over Jericho getting canceled. I liked it a lot, but my wife and her sisters didn't, so it probably didn't have enough market appeal to justify its production costs in the eyes of the network.
The tape/HD size ratio is getting so ridiculous that at work we're seriously considering using Hard Drives as removable tape-like media for backup. Any other solution for backing up terabytes of data is too expensive.
We've been doing this for a few years actually, as a "roll your own" solution. We currently use removable drive carriers from DataStor, and 500 GB Seagate disks (first ATA, now SATA). We also use foam-padded locking carriers that are take off-site every day. We do ~1.2 TB of backups every night, and have had exactly one drive failure in 4 years.
We actually back up to a large fixed disk array using various backup software from multiple systems, and then make an encrypted copy all of those files to the removable disks using 7-ZIP. We then pad the empty space on the removable drives with Par2 data if possible (we compute as much data as our window allows, currently about 20% redundancy). We also store unencrypted binaries (and source if possible) of all the software needed for recovery with each backup.
We have not evaluated any of the commerial "disk to disk" backup solutions, as ours has proven simple and reliable, and all we need is an SATA- or USB2-equipped machine to begin recovery. Most of the commercial hardware solutions require you to have a similar (proprietary) unit at the recovery site.
or more esoteric hardware that you can get yourself into trouble
A high-performance transaction processing system is likely to require "esoteric" hardware. Like extreme processor counts, high-throughput I/O subsystems, TCP/IP offload, InfiniBand clustering interconnects, hot-swap memory and CPUs... the sort of things hardware vendors support only on Z/OS, AIX or Solaris (and maybe Windows).
Extremisim in any form is pretty tough to distinguish from satire. For instance, it's hard to tell if the thousands of the inane "OMG Linux+OOo+Beryl rocks M$ is the sux0r!" posts here are satire or not. I hope at least some of them are.
Slashdot Mirror
Yes, but Democrats spend money on things like the Department of Education, which has done nothing to stop (or has perhaps accelerated) a tremendous decline in the quality of public schooling in the US. This despite a $69.4 billion budget in 2006, with budget that increases faster than inflation every year. Thank you, Jimmy Carter. And the program is politically un-killable, because it is "for the children".
It's amazing, but Democrats can't seem to figure out that throwing money at social problems almost never solves them. Federal Social programs are just as much "pork barrel" as a defense contract, except the beneficaries are bureaucrats and unions instead of Lockheed-Martin. Giving money to a homeless drug addict on the street suprisingly doesn't magically get him into rehab. The addict just buys more crack.
At least the Republicans waste money on cool thinks like submarines and missles. They cost a lot, but they have at least some value (even if they're only sold a decade later on the arms market). Have you ever seen video of an MLRS lauch? As P.J. O'Rourke would say, "Now that's the way you waste taxpayer money!".
The Boeing X-48 and X-51 are are the forefront of BWB and Waverider research respectively. Boeing is still using the tube-with-wings design because it is a known-safe configuration. BWBs, like the F-117 (sort of) and B-2, can require some fairly serious computer assistance to remain stable in low-speed flight. Boeing is not "behind the curve" as you suggest.
Safety trumps everything else in passenger aviation. Reading suggests the first non-military BWB you see will be used for cargo, not passengers. It will likely be decades before they are in production for passenger use, as some Googling reveals that neither Boeing, Airbus or any other company has a production BWB airliner on the drawing board.
Perhaps some maverick start-up will build a BWB airliner soon. But they better also start their own airline, because airlines are so risk-averse that they would never buy such a thing until it had decades of safe service in military or other markets.
Almost every processor is bechmarked using SPECint 2006 and (for multi-core systems) SPECint_rate 2006. Yes, it ignores some I/O effects, but it's not too shabby of a "universal" benchmark. We frequently use SPECint_rate scores to help us decide what sort of application and datbase servers to buy in a particular year.
Doing realistic benchmarks on all that disparate hardware with our own apps would be way too expensive and time consuming, so we just rely on SPEC to make the benchmarks "fair".
Wouldn't it be nice if the SPECint_rate score was the model number of a chip? You'd have the intel 2-180 or the AMD b193. You'd know the AMD was marginally faster than the 180 in total multi-threaded throughput.
Because you are not a US citizen, so you don't have any rights under the US Consitution. Visiting the USA is a privilege for you, not a right. I can travel freely inside the USA without any documentation whatsoever. Unless, of course, I get pulled over for speeding, in which case I'd better have a drivers license. Or if I need to get through security in the airport, in whcih case a Driver's License, State ID, or passport is required.
Really, the only way to travel freely in the USA these days is by train. But passenger train service in the US is almost non-existent.
In Trinity and Beyond: The Atomic Bomb Movie , film clips of several high-altitude detonations are shown. Carried atop various missiles, all of which are probably going much faster than mach six (vertically), the fireballs from these high-altitude detonations are still almost perfectly spherical until they start drifting inthe wind many seconds after detonation.
As I mentioned, the fireball is formed by radiation effects within about one microsecond, and its radius is several hundred meters at least. Given that in one microsecond, the Mach 6 plane would only travel about two milimeters, I do not think the fireball (and the resulting mushroom cloud if the detonation were near the ground) would be significantly deformed. Everything inside the fireball is essentially superheated plasma, and the blast wave, which creates the mushroom cloud by kicking up debris, radiates outward from the fireball. So the momentum of the vaporized plane would essentially be overwhelmed by the momentum added to all the particles inside the superheated fireball.
But this is all just semi-informed speculation on my part.
Almost exactly the same as a "traditional" H-Bomb airburst. The thermal pulse travels at nearly the speed of light, and would vaporize the plane almost instantly. So the shockwave and fireball that form would probably radiate from the detonation point at almost exactly the same speeds and shapes as a stationary airburst.
The actual fission-fusion part of a thermonuclear nuclear detonation is so short that the fact that the bomb is moving at high speed makes almost no difference. The added energy of a hurtling piece of machinery is inifitesmally small compared with the detonation energy.
Why in the hell would a blade server have video or audio capabilities? Servers are designed to sit in data center racks and just keep running. They're a bit expensive, noisy and strangely-shaped to use as AV workstations.
You may want a blade server with audio/video for some ungodly reason. The other 99.9995% of the server market does not.
Considering IBM makes most of their money from software and services these days, IBM probably don't care a bit that the tech support guy from Circuit City is no longer an IBM customer.
This isn't practical in an Internet without pervasive global multicast services. Do you think ISPs are really going to let their extraordinarily expensive trans-continental pipes be clogged with Internet radio?
And don't assume BitTorrent-like schemes solve the problem, as your friends and neighbors participating in a swarm would have the RIAA beating on their door for redistributing Internet Radio without paying license fees. BitTorrent is a clever hack to emulate multicast in the unicast Internet, but it has the undesirable property that participating nodes must store and publicly redistribute data blocks, thereby running afoul of the whole copyright regime.
By the way, you also should look a little more closely at the *chart* below the graph in the study I linked to. Linux is mentioned specifically, rather than just "other". "Linux"+"Unknown" = 0.75% market share at most. Vista has 3.74%, or 4.99 times the market share of Linux.
And that assumes all "Unknown" boxes are Linux, which is most definitely not the case. At least some are FreeBSD, Cable Boxes, random mobile phones, and those crazy airport pay-per-minute terminals.
Yes, your indeirect measure of RAM is certainly more relaible than one based on visitors to 40,000 LiveStats customer sites.
Here's a hint: one billion internet USERS != 1 billion WORKSTATIONS. The majority of internet users share a PC, wither with family or and an Internet Cafe.
Vista is not selling AS FAST AS MEMORY MAKERS EXPECTED, but it is by no means not selling. I see Vista laptops galore on the Chicago commuter trains, far more than OSX or Linux machines. In fact, my Ubuntu laptop is the only Linux laptop I've ever seen in a public place (IT conferences being an exception).
Your sample is probably a bit skewed if you're coming from an academinc environment. I was pretty shocked 10 years ago when I came out of college eleven years ago realized nobody used the Macs or Solaris worksations with which I was so familiar in the real world.
This study shows there are already 5 times as many Vista workstations in use versus Linux workstations. Assuming Microsoft has sold ~20M copies through May, that means there are far fewer than 20M linux workstations in active use. Unless you assume that more than 4/5 Vista installs is pirated.
Note my original statement should have been qualified to compare Vista with Linux workstations, and not all "Linux machines". Comparing the population of a workstation OS to a server install base doesn't make much sense.
Twenty million licenses sold already is "no evidence"? That's more than all the active Linux and MacOS machines out there in just a few months.
I have Vista on my laptop, and it is quite an improvement over XP. The very few rough edges I've encounetered have been with support of oddball hardware, like my Sprint Wireless Broadband card, which didn't have good Vista drivers until recently.
There is a huge pain in the ass with corporate licensing and activation that will be a real barrier to business migrating existing machines to Vista, though. Most likely they will just wait and acquire pre-activated Vista with new hardware. Microsoft really kicked themselves in the balls with activation for enterprise customers, it is just horribly broken. We have not had one machine activate properly, and a phone call to the automated activation system is always needed as a result. Completely unwordkable for a company our size, and we only have a few hundreds of machines. So our ten guinnea pig machines will be the only Vista we run until the next hardware refresh.
You forgot:
I think you meant "If so then a Linux distro is required for Firefox none of these exploits work". The holes (vulnerabilities) are almost certainly still there in the code base for FF on Linux, but the exploit code needs to be crafted differently for the non-Windows OS. It very likely that creating exploit code for the Linux versions of these holes is trivial.
I'll bet you the same exploit code wouln't work on the Itanium versions of Windows, either. But the holes are still there.
I think your solution is ingenious, in that it spreads disk seeks evenly. But I suspect it is probably bad idea for performance, because of how the partitions are physically allocated on disk.
Wouldn't the the disk heads be hopping across 1/2 of the disk the disk quite a bit, since they have to read from partition to partition? It might not matter with your workload, but I would be hesitant run a database server on such an array.
In my employer's environment, we usually do one partition per physical disk. Why? We benchmarked for SQL Server using MSFT's IOstress utility when setting up a new DB server. Having the database files spread over two arrays over 8 disks (with each disk partitioned into 2 parts, one for each array) performed significantly worse than simply using one big partition that occupied the whole array on the same 8 disks. Our theory was that while the same number of individual seeks had to happen in both cases, the seeks were much "longer" when the individual disks were partitioned.
Of course, that was about 5 years ago, so with all the changes in disk density, cache sizes, etc., it may not matter much anymore. Our DB servers now have 16GB or more of RAM, and keep everything "hot" in memory.
We have a number of pieces of equipment with that issue in our server room. Cleaning crews would hit the big red button on our AC flow panel about once a year. Our solution was to take a small plastic cap from an aeorsol can and tape it over the switch with gaffer's tape (no residue). Label it of course. Seems to work, and costs all of a few cents.
We do block-level replication off-site with our SAN, which is not an EMC, but a Left Hand Networks cluster. The reason we bought in was scalability. Our LeftHand units come in as small as 2 TB chunks, each with it own mobo and two Gigabit Ethernet adapters. As you buy more chunks, your storage pool automatically grows (and volumes automatically re-balance themselves amongst spindles). But even better, the aggregate throughput also increases, as each chunk adds its 2 GB/sec to the total SAN bandwidth available. Using iSCSI and and multi-patch IO, there are no single points of failure in the SAN at this site. In case of fire, we have our replicated data off-site (also handled by the SAN in a bandwidth-efficient manner), and also traditional backup-media off-site.
You just can't do that sort of stuff with a roll-your-own solution. If you don't have those sorts of needs, by all means, roll your own cheap solution.
What this should illustrate is that the majority of the value of a SAN is in the management software, not just the hardware. Which is probably why LeftHand has now partnered with HP to sell their iSCSI SAN software running on DL300-series machines.
Note, I do not work for or invest in LeftHand, I am just a satisfied customer. LeftHand's SAN/IQ operating system is Linux-based, which is neat. We were also thrilled with demos of EqualLogic's gear, which is quite similar.
Except I have never seen "unlimited" service advertised to consumers by an ISP in Chicago, USA. I've seen things like "NO PER-MINUTE CHARGE", "24 hours a day", all with asterisks to read the fine print. But never "unlimited". And if you actually read the contract before you sign, it clearly states that the committed information rate is basically zero and all service is best-effort.
Where is all this "unlimited" false advertising? Can someone please post a link? Because to me it just sounds like a bunch of whining by kids that want free movies, free music, and almost-free internet access, too.
No, Beta lost because it had NO PORN. Sony would not license to adult video types, byt JVC et. all would. Porn was the killer app for the VCR. Men buy electronics for the home, and VHS meant porn at home - no more worrying about being seen at one of the sticky-floored "adult film theaters" on the bad side of town (they were very common in the late 70s and early 80s, especially around Times Square in NYC).
Amen. Microsoft supported x86, MIPS, PowerPC, and Alpha with the first release of NT 4.0. Nobody bought PowerPC and MIPS, and very very few people bought Alpha. So by the time Win2k came around, Windows was x86 only. I really hoped Alpha or PowerPC would succeed and get us off the multilayered hack that is x86, but the masses did not agree with me.
The good thing about this debacle for MSFT is that today, almost evertyhing in the Windows kernel is platform-agnostic except for HAL.DLL and drivers. This fact enables the Itanium versions of Windows. It also means that Microsoft can produce a version for, say, POWER6 processors with a modest incremental investment, if they see market for it. One can only hope.
Not to mention The Tudors, which is a 10+ hour historical drama about a young Henry VII on Showtime that kicks ass.
Of course, I am still smarting over Jericho getting canceled. I liked it a lot, but my wife and her sisters didn't, so it probably didn't have enough market appeal to justify its production costs in the eyes of the network.
We've been doing this for a few years actually, as a "roll your own" solution. We currently use removable drive carriers from DataStor, and 500 GB Seagate disks (first ATA, now SATA). We also use foam-padded locking carriers that are take off-site every day. We do ~1.2 TB of backups every night, and have had exactly one drive failure in 4 years.
We actually back up to a large fixed disk array using various backup software from multiple systems, and then make an encrypted copy all of those files to the removable disks using 7-ZIP. We then pad the empty space on the removable drives with Par2 data if possible (we compute as much data as our window allows, currently about 20% redundancy). We also store unencrypted binaries (and source if possible) of all the software needed for recovery with each backup.
We have not evaluated any of the commerial "disk to disk" backup solutions, as ours has proven simple and reliable, and all we need is an SATA- or USB2-equipped machine to begin recovery. Most of the commercial hardware solutions require you to have a similar (proprietary) unit at the recovery site.
A high-performance transaction processing system is likely to require "esoteric" hardware. Like extreme processor counts, high-throughput I/O subsystems, TCP/IP offload, InfiniBand clustering interconnects, hot-swap memory and CPUs... the sort of things hardware vendors support only on Z/OS, AIX or Solaris (and maybe Windows).
Extremisim in any form is pretty tough to distinguish from satire. For instance, it's hard to tell if the thousands of the inane "OMG Linux+OOo+Beryl rocks M$ is the sux0r!" posts here are satire or not. I hope at least some of them are.
"Hello, Pot? This is Kettle. You're black!"