Ok we use smartphones using fingers on touchscreens quite a bit, but for a device that boasts that it's the evolution of mouse they should have thought that 2-3 minutes (smartphone quick use, on the road, whatever) is very different from desktop/laptop use (could be hours).
This most certainly is an evolution of the mouse. It's every bit as much as the addition of extra buttons, or a scroll wheel, or an optical pickup. You have to remember that evolution is unguided, and completely random. Fortunately, in our current computing environment, it will be heavily selected against, and die off quickly.
The only place I could see something like a virtual keyboard or mouse being of use would be a kitchen, or other similar location where your hands are likely to get messy. At home, you could use it for pulling up recipes and such. In a business, it could be an interface for tracking orders, cook timers, or inventory management. Have an all-in-one unit mounted underneath a cabinet, and the keyboard/mouse projecting down on the counter top below it. A counter is a lot easier to clean off than computer equipment, and the monitor could be easily folded out of the way if needed.
and the fact that regulatory approvals take a shitload of time for EACH reactor.
and you need 1200 of them to even come close to meeting demand.
That hundreds of days of paperwork and review are for large, full size plants, generating gigawatts of power, custom built for each individual installation. The idea is that these smaller units will effectively be made on an assembly line, with a plant producing hundreds of these per year. With sufficiently high production quantity, and sufficiently reliable production methods, you won't have to do the full lengthy review of each and every unit. You review the design, and then confirm each unit has conformed to the design. The article specifically talks about how the regulatory commission is discussing how to revise this review process to better handle such 'cookie cutter' reactors.
The nuclear reactors in Fukushima are boiling water reactors. It uses water for coolant, which boils as it flows through the reactor chamber, goes through a heat exchanger, and is recirculated. Since the coolant systems are not functioning properly, they are dumping saltwater into the coolant lines, letting it boil off, and vent out into the atmosphere through pressure release valves. This is releasing radiation, however it is a small amount, and containing elements with short half-lives that will decay rapidly and cease to be a danger. This has been happening for several days
This steam release is very energetic. It is so energetic that the water is spontaneously disassociating to hydrogen and oxygen. When you get large volumes of hydrogen and oxygen, along with a high temperature source, you're going to have an explosion. There is no way around that, but it is not an indicator that the containment vessel has breached and the core is exposed.
The reports of a breach in reactor #2 appear to be part of the coolant system. The suppression chamber has developed a crack, which lead to an uncontrolled release of coolant, as the system depressurized to atmospheric. This resulted in a large venting of radiation as it depressurized, but now, the situation is no different than at the controlled steam releases at the other reactors. The containment vessel is still intact. Corium is not flowing out of the containment vessel. There is not currently any risk of it being released and contaminating the ground water.
The comment wasn't intended to be derogatory against the ARM. The ARM was just designed from the ground up with low power consumption in mind, not performance. The Cortex A9 has an 8-stage pipeline, 2.5 instructions per clock, around 13M transistors per core, runs at 800MHz to 1.5GHz, and has up to 512KB of L2 cache. The Pentium 3 has a 10-stage pipeline, 2.5 instructions per clock, around 10M transistors, runs at 500MHz-1.4GHz, and has up to 512KB of L2 cache. They're fairly comparable processors, with the ARM probably having a better instruction dispatcher and branch predictor, and the P3 having better floating point performance.
While it doesn't have a lot of power comparable to modern x86 chips, it absolutely blows them away in performance per watt. It's a much better prospect for low power systems than the Atom, where Intel effectively tossed out 15 years of microprocessor design ripping out parts to cut power consumption.
The MythTV guys have completely different needs than an underutilized server operator. We have to deal with a very complex scheduler, which if it takes too long to run can cause problems, and with HD video that typically can only be decoded single threaded. Single threaded performance, and a lot of it, is a must, meaning our minimum recommendation is 2.5GHz Core 2 or Athlon II, or better.
That's not to say you can't be low power while you're at it. Tom's Hardware did an article last year where with not considerable effort, they put together a 3.33GHz dual core i5 that idled under 25W. Even better, one of the Mac Mini XServes would idle at less power than your existing Atom. It's always nice to have the headroom available should you want it in the future, and at 25W, it's only going to consume maybe $50 more electricity over a 5yr life than that Atom system.
Genetic profiling these days is fairly commonplace. The problem is that it requires a bit of setup, followed by a lot of waiting, and more importantly waiting while tying up limited equipment resources. In addition, when doing forensics work, you're not going to have the large clean sample you have have with clinical tests, meaning you're going to tie up more equipment and more time cultivating a good sample. You can have a genetic profile done in a couple hours (which they show as the writer calls the mayor to get priority for his testing), but that's only if you have dedicated staff and equipment. NCIS, with Abby and her basement lab for four agents simply isn't realistic.
I liked how in one of the early episodes of Castle, the writer is expecting DNA analysis to come back by the end of lunch, only to be told by the detectives it won't be ready for another week or two, due to processing time and a deep backlog at the lab.
That's not at all true. The CM-1 and CM-2 were very strange systems, but the CM-5 displayed in the movie was just a traditional 'fat tree' of SPARC processors. Fully loaded with 16K processors, the maximum allowed for, a CM-5 was capable of an Rpeak around 130GFlop/s, or nearly 150 times that of the Cray X-MP that was used in the book.
The P6 IS a RISC architecture. One of the big changes of the Pentium Pro over the previous x86 processors was the fact that internally, the processor used a reduced set of micro-ops, translated from the much larger x86 instruction set. It made it far simpler to design an instruction dispatcher, and made the superscaler design introduced in the P5 architecture a lot more effective.
NCIS does the same thing, and it's for the same reason that everyone uses 555-xxxx phone numbers. No one can possibly have that phone number, so they can use it without fear of undue harm coming to the owner of that number. Now they could use addresses in 10/8, or 172.16/12, or 192.168/16, or 169.192/16, or the whole 240/4 block. Making all the numbers >255 just has the effect of making the numbers appear more complex and random.
You do realize that diesel fuel has roughly 11% more energy by volume than gasoline, right? That puts your higher priced diesel fuel just about on even footing with standard unleaded from an energy standpoint. The engine with better thermal efficiency should still prevail. If it's still more economical to buy a gasoline powered car, well then that TDI engine is just a piece of junk.
VMs are a tool. It's not really up to one person to say how the tool is used. More tools in the toolbox is a good thing.
A sledgehammer is a perfectly fine tool, but I wouldn't use it to put a nail in a wall to hang a picture from. Use the proper tool that fits the task. Carrying around a 5GB system install that you have to boot up just to run a web browser, and another one for email, and then snapshots of each of those to recover from, and then more images to see if applications cause problems (you can't actually use this for testing of drivers which are far more problematic than individual applications), is absolutely absurd.
Do yourself a favor. Get an email provider that supports IMAP. If you lose your system install, that data is still stored all safe and sound on the server, ready to access from anywhere at any time. Get a file server, or even just an external drive, and configure Windows to start doing periodic incremental backups. It will take a huge amount of space, and will let you recover in full from the backup at any time.
Legacy games would be covered under 'support of old no longer available hardware'. In that case, you would probably want to use something specialized like DosBox, rather than the general purpose VirtualBox. I dont know what 'specialized software' a non-computer-savvy person might want to run that's not otherwise available under Windows, except maybe system recovery tools, which would be run natively off a bootable disk anyway. If it's anything performance or resource demanding, you would also want to run that natively, rather than translated through a VM.
I also develop some software that needs to run properly in both windows and linux and others.
I am not a full time developer but I have a need to Virtualize and it is the only solution that works for me, without having two boxes under my desk.
I did mention cross platform development and testing as a reasonable use for virtualization in my original comment.
No. If I want to have a Mac and my company forces me to use a Windows app, I won't be dual-booting just to run the company's app.
Does that mean your employer doesn't provide you a computer to use? Can you not run this application through some translation layer like WINE or similar derivatives? Surely you can come up with a better solution than to run a fully virtualized instance of another computer, running a complete install of Windows.
Isn't that a convoluted way out of a problem? Why write (and test) things when you can virtualize and run your VMs within minutes.
Mount, chroot, a couple select init scripts, and you have an instance running within seconds of allocation. More advanced tools than chroot could further limit the memory space or network access as needed. It's not like any VM solution could be put into production use without extensive testing.
Gee! MAYBE I'm virtualizing to be ABLE to live migrate when I get new hardware?
If you're that concerned about the downtime caused by hardware upgrades years apart, then perhaps you need software that won't be affected by loss of a single machine.
Go TRY virtualization and then come back. It's clear that your assumptions are based on what you read around the Internet and some experience with VMware/VirtualBox. And one more tip: change your attitude. Stop writing like an elitist douchebag, you certainly sound like one. I could have answered your post just saying "then why does virtualization exist in the first place?". There is a need, there are solutions, and yours are fine for many cases, but not for others.
There certainly are uses for virtualization, but it should not garner nearly the market it has. It's pulling out the 20lb sledge for everything, when most tasks really only need a ball-peen. It comes right back to the old adage of the tool that does everything does nothing well.
A consumer need for virtualization generally indicates an incorrect choice of OS. If you find yourself using OSX or Linux, and having to use a VM to run Windows software, then the better options would be to either run Windows, or find an alternative program on your OS of choice.
A commercial need for virtualization generally indicates system operators taking the convoluted way out of a problem, rather than properly designing a system. If you want to run multiple tasks on a single system, then just run multiple tasks on a single system. You never needed virtual machines unnecessarily complicating things before. If you want isolation so you can move the tasks around easily, use chroot. If you want security between the tasks, use one of the more advanced tools like containers or jails or vservers. If you want want to use rapidly allocatable 'coulds', then just write some simple tools to mount the relevant images, and chroot into it as needed. If you think you need live migration of long running tasks, maybe you really just need to buy some dedicated hardware.
The only reason someone should be running virtual machines would either be for cross platform development and testing, support of old no longer available hardware, or for use on hardware such as the Z-Machines, which have some sort of inherent architectural improvement, such as pervasive redundancy, that are not otherwise available on the architecture of choice.
This cave is near the equator, and as you know, the moon is dark two weeks every month. You'll need some pretty good power storage systems to keep a manned base running for that long on backup. The only place you would have (nearly) perpetual sunlight would be on the poles.
Unfortunately, recuperating the energy while braking has at most a 50% efficiency last time I looked, mainly due to heat dissipation occurring in the process. 85% seems optimistic depending of the real life setup you plan to use but you might be right after all;-)
50% seems awfully horrible for a generator. Induction motors tend to be above 90% efficient over most of their operating range, for both directions. The 85% value was a rough guess for the full power, plus recovery, including battery inefficiency. If I had to guess, I would say the 50% value was from the fact that most existing electric vehicles use pathetically small electric motors. The dynamic braking capacity of the 27kW motor in a Prius is extremely limited, meaning drivers typically fall back to more traditional brakes (heat dissipation). I would bet the 188kW motor in a Tesla fares far better.
Accelerating the mass of the battery doesn't really matter. Induction motors can be run in reverse to recover most (>85%) of the energy put into the vehicle. The chemical storage in the battery is simply exchanged for kinetic storage in the vehicle. The problem is the increase in rolling friction on the tires, which increases with higher load, and is a significant amount of loss until you get well above highway speeds.
Technically, you are correct. However, power companies have to run peak load plants to make up the difference in power draw from what the baseline plants provide. If you can come up with an economical means of storing vast amounts of energy, you would be able to build and operate more baseline plants, and do away with the more expensive, less efficient, peak plants.
Similarly, if you can provide a significant energy buffer, otherwise unreliable power sources like wind and solar become considerably more viable.
Where is the upside for a customer in caring about ipv6? Will they want to decloak when/if ipv6 becomes popular? OMG, my PC is broadcasting an IP address, of course I want your wonderful product to protect me! All ipv6 would do is get every Windows PC pwn3d twenty four hours after deployment and then everyone retreats behind a NAT and dynamic IP again, this time grafted onto ipv6.
How many times do we have to tell you people, NAT is not a security mechanism. All it does is translate packets from one address to another. All of these consumer NAT routers could just as easily become consumer stateful firewalls. Block all traffic unless originating from the internal physical network, or there is a specific rule to allow it. It's not hard. It's really functionally no different. You just lose those restrictions like not being able to run multiple servers on the same port, or not being able to use protocols like SIP which encode the address in each packet.
If you have enough addresses that everyone can have one, there is absolutely no reason why they shouldn't.
Slashdot Mirror
Ok we use smartphones using fingers on touchscreens quite a bit, but for a device that boasts that it's the evolution of mouse they should have thought that 2-3 minutes (smartphone quick use, on the road, whatever) is very different from desktop/laptop use (could be hours).
This most certainly is an evolution of the mouse. It's every bit as much as the addition of extra buttons, or a scroll wheel, or an optical pickup. You have to remember that evolution is unguided, and completely random. Fortunately, in our current computing environment, it will be heavily selected against, and die off quickly.
The only place I could see something like a virtual keyboard or mouse being of use would be a kitchen, or other similar location where your hands are likely to get messy. At home, you could use it for pulling up recipes and such. In a business, it could be an interface for tracking orders, cook timers, or inventory management. Have an all-in-one unit mounted underneath a cabinet, and the keyboard/mouse projecting down on the counter top below it. A counter is a lot easier to clean off than computer equipment, and the monitor could be easily folded out of the way if needed.
and the fact that regulatory approvals take a shitload of time for EACH reactor. and you need 1200 of them to even come close to meeting demand.
That hundreds of days of paperwork and review are for large, full size plants, generating gigawatts of power, custom built for each individual installation. The idea is that these smaller units will effectively be made on an assembly line, with a plant producing hundreds of these per year. With sufficiently high production quantity, and sufficiently reliable production methods, you won't have to do the full lengthy review of each and every unit. You review the design, and then confirm each unit has conformed to the design. The article specifically talks about how the regulatory commission is discussing how to revise this review process to better handle such 'cookie cutter' reactors.
The nuclear reactors in Fukushima are boiling water reactors. It uses water for coolant, which boils as it flows through the reactor chamber, goes through a heat exchanger, and is recirculated. Since the coolant systems are not functioning properly, they are dumping saltwater into the coolant lines, letting it boil off, and vent out into the atmosphere through pressure release valves. This is releasing radiation, however it is a small amount, and containing elements with short half-lives that will decay rapidly and cease to be a danger. This has been happening for several days
This steam release is very energetic. It is so energetic that the water is spontaneously disassociating to hydrogen and oxygen. When you get large volumes of hydrogen and oxygen, along with a high temperature source, you're going to have an explosion. There is no way around that, but it is not an indicator that the containment vessel has breached and the core is exposed.
The reports of a breach in reactor #2 appear to be part of the coolant system. The suppression chamber has developed a crack, which lead to an uncontrolled release of coolant, as the system depressurized to atmospheric. This resulted in a large venting of radiation as it depressurized, but now, the situation is no different than at the controlled steam releases at the other reactors. The containment vessel is still intact. Corium is not flowing out of the containment vessel. There is not currently any risk of it being released and contaminating the ground water.
The comment wasn't intended to be derogatory against the ARM. The ARM was just designed from the ground up with low power consumption in mind, not performance. The Cortex A9 has an 8-stage pipeline, 2.5 instructions per clock, around 13M transistors per core, runs at 800MHz to 1.5GHz, and has up to 512KB of L2 cache. The Pentium 3 has a 10-stage pipeline, 2.5 instructions per clock, around 10M transistors, runs at 500MHz-1.4GHz, and has up to 512KB of L2 cache. They're fairly comparable processors, with the ARM probably having a better instruction dispatcher and branch predictor, and the P3 having better floating point performance.
While it doesn't have a lot of power comparable to modern x86 chips, it absolutely blows them away in performance per watt. It's a much better prospect for low power systems than the Atom, where Intel effectively tossed out 15 years of microprocessor design ripping out parts to cut power consumption.
The MythTV guys have completely different needs than an underutilized server operator. We have to deal with a very complex scheduler, which if it takes too long to run can cause problems, and with HD video that typically can only be decoded single threaded. Single threaded performance, and a lot of it, is a must, meaning our minimum recommendation is 2.5GHz Core 2 or Athlon II, or better.
That's not to say you can't be low power while you're at it. Tom's Hardware did an article last year where with not considerable effort, they put together a 3.33GHz dual core i5 that idled under 25W. Even better, one of the Mac Mini XServes would idle at less power than your existing Atom. It's always nice to have the headroom available should you want it in the future, and at 25W, it's only going to consume maybe $50 more electricity over a 5yr life than that Atom system.
512MB per core really isn't bad at all, when you consider that core has about the same performance of a 10yr old Pentium 3.
Genetic profiling these days is fairly commonplace. The problem is that it requires a bit of setup, followed by a lot of waiting, and more importantly waiting while tying up limited equipment resources. In addition, when doing forensics work, you're not going to have the large clean sample you have have with clinical tests, meaning you're going to tie up more equipment and more time cultivating a good sample. You can have a genetic profile done in a couple hours (which they show as the writer calls the mayor to get priority for his testing), but that's only if you have dedicated staff and equipment. NCIS, with Abby and her basement lab for four agents simply isn't realistic.
I liked how in one of the early episodes of Castle, the writer is expecting DNA analysis to come back by the end of lunch, only to be told by the detectives it won't be ready for another week or two, due to processing time and a deep backlog at the lab.
That's not at all true. The CM-1 and CM-2 were very strange systems, but the CM-5 displayed in the movie was just a traditional 'fat tree' of SPARC processors. Fully loaded with 16K processors, the maximum allowed for, a CM-5 was capable of an Rpeak around 130GFlop/s, or nearly 150 times that of the Cray X-MP that was used in the book.
The P6 IS a RISC architecture. One of the big changes of the Pentium Pro over the previous x86 processors was the fact that internally, the processor used a reduced set of micro-ops, translated from the much larger x86 instruction set. It made it far simpler to design an instruction dispatcher, and made the superscaler design introduced in the P5 architecture a lot more effective.
NCIS does the same thing, and it's for the same reason that everyone uses 555-xxxx phone numbers. No one can possibly have that phone number, so they can use it without fear of undue harm coming to the owner of that number. Now they could use addresses in 10/8, or 172.16/12, or 192.168/16, or 169.192/16, or the whole 240/4 block. Making all the numbers >255 just has the effect of making the numbers appear more complex and random.
You do realize that diesel fuel has roughly 11% more energy by volume than gasoline, right? That puts your higher priced diesel fuel just about on even footing with standard unleaded from an energy standpoint. The engine with better thermal efficiency should still prevail. If it's still more economical to buy a gasoline powered car, well then that TDI engine is just a piece of junk.
Sounds like you're talking about something like this... http://en.wikipedia.org/wiki/Connection_Machine
but it works 99.99% of the time, all the time
Sex panther. 60% of the time, it works every time. It's made with bits of real panther, so you know its good.
I wonder if Macs have bits of real apple in them.
VMs are a tool. It's not really up to one person to say how the tool is used. More tools in the toolbox is a good thing.
A sledgehammer is a perfectly fine tool, but I wouldn't use it to put a nail in a wall to hang a picture from. Use the proper tool that fits the task. Carrying around a 5GB system install that you have to boot up just to run a web browser, and another one for email, and then snapshots of each of those to recover from, and then more images to see if applications cause problems (you can't actually use this for testing of drivers which are far more problematic than individual applications), is absolutely absurd.
Do yourself a favor. Get an email provider that supports IMAP. If you lose your system install, that data is still stored all safe and sound on the server, ready to access from anywhere at any time. Get a file server, or even just an external drive, and configure Windows to start doing periodic incremental backups. It will take a huge amount of space, and will let you recover in full from the backup at any time.
Legacy games would be covered under 'support of old no longer available hardware'. In that case, you would probably want to use something specialized like DosBox, rather than the general purpose VirtualBox. I dont know what 'specialized software' a non-computer-savvy person might want to run that's not otherwise available under Windows, except maybe system recovery tools, which would be run natively off a bootable disk anyway. If it's anything performance or resource demanding, you would also want to run that natively, rather than translated through a VM.
I also develop some software that needs to run properly in both windows and linux and others. I am not a full time developer but I have a need to Virtualize and it is the only solution that works for me, without having two boxes under my desk.
I did mention cross platform development and testing as a reasonable use for virtualization in my original comment.
No. If I want to have a Mac and my company forces me to use a Windows app, I won't be dual-booting just to run the company's app.
Does that mean your employer doesn't provide you a computer to use? Can you not run this application through some translation layer like WINE or similar derivatives? Surely you can come up with a better solution than to run a fully virtualized instance of another computer, running a complete install of Windows.
Isn't that a convoluted way out of a problem? Why write (and test) things when you can virtualize and run your VMs within minutes.
Mount, chroot, a couple select init scripts, and you have an instance running within seconds of allocation. More advanced tools than chroot could further limit the memory space or network access as needed. It's not like any VM solution could be put into production use without extensive testing.
Gee! MAYBE I'm virtualizing to be ABLE to live migrate when I get new hardware?
If you're that concerned about the downtime caused by hardware upgrades years apart, then perhaps you need software that won't be affected by loss of a single machine.
Go TRY virtualization and then come back. It's clear that your assumptions are based on what you read around the Internet and some experience with VMware/VirtualBox. And one more tip: change your attitude. Stop writing like an elitist douchebag, you certainly sound like one. I could have answered your post just saying "then why does virtualization exist in the first place?". There is a need, there are solutions, and yours are fine for many cases, but not for others.
There certainly are uses for virtualization, but it should not garner nearly the market it has. It's pulling out the 20lb sledge for everything, when most tasks really only need a ball-peen. It comes right back to the old adage of the tool that does everything does nothing well.
A consumer need for virtualization generally indicates an incorrect choice of OS. If you find yourself using OSX or Linux, and having to use a VM to run Windows software, then the better options would be to either run Windows, or find an alternative program on your OS of choice.
A commercial need for virtualization generally indicates system operators taking the convoluted way out of a problem, rather than properly designing a system. If you want to run multiple tasks on a single system, then just run multiple tasks on a single system. You never needed virtual machines unnecessarily complicating things before. If you want isolation so you can move the tasks around easily, use chroot. If you want security between the tasks, use one of the more advanced tools like containers or jails or vservers. If you want want to use rapidly allocatable 'coulds', then just write some simple tools to mount the relevant images, and chroot into it as needed. If you think you need live migration of long running tasks, maybe you really just need to buy some dedicated hardware.
The only reason someone should be running virtual machines would either be for cross platform development and testing, support of old no longer available hardware, or for use on hardware such as the Z-Machines, which have some sort of inherent architectural improvement, such as pervasive redundancy, that are not otherwise available on the architecture of choice.
This cave is near the equator, and as you know, the moon is dark two weeks every month. You'll need some pretty good power storage systems to keep a manned base running for that long on backup. The only place you would have (nearly) perpetual sunlight would be on the poles.
Unfortunately, recuperating the energy while braking has at most a 50% efficiency last time I looked, mainly due to heat dissipation occurring in the process. 85% seems optimistic depending of the real life setup you plan to use but you might be right after all ;-)
50% seems awfully horrible for a generator. Induction motors tend to be above 90% efficient over most of their operating range, for both directions. The 85% value was a rough guess for the full power, plus recovery, including battery inefficiency. If I had to guess, I would say the 50% value was from the fact that most existing electric vehicles use pathetically small electric motors. The dynamic braking capacity of the 27kW motor in a Prius is extremely limited, meaning drivers typically fall back to more traditional brakes (heat dissipation). I would bet the 188kW motor in a Tesla fares far better.
Accelerating the mass of the battery doesn't really matter. Induction motors can be run in reverse to recover most (>85%) of the energy put into the vehicle. The chemical storage in the battery is simply exchanged for kinetic storage in the vehicle. The problem is the increase in rolling friction on the tires, which increases with higher load, and is a significant amount of loss until you get well above highway speeds.
Technically, you are correct. However, power companies have to run peak load plants to make up the difference in power draw from what the baseline plants provide. If you can come up with an economical means of storing vast amounts of energy, you would be able to build and operate more baseline plants, and do away with the more expensive, less efficient, peak plants.
Similarly, if you can provide a significant energy buffer, otherwise unreliable power sources like wind and solar become considerably more viable.
Where is the upside for a customer in caring about ipv6? Will they want to decloak when/if ipv6 becomes popular? OMG, my PC is broadcasting an IP address, of course I want your wonderful product to protect me! All ipv6 would do is get every Windows PC pwn3d twenty four hours after deployment and then everyone retreats behind a NAT and dynamic IP again, this time grafted onto ipv6.
How many times do we have to tell you people, NAT is not a security mechanism. All it does is translate packets from one address to another. All of these consumer NAT routers could just as easily become consumer stateful firewalls. Block all traffic unless originating from the internal physical network, or there is a specific rule to allow it. It's not hard. It's really functionally no different. You just lose those restrictions like not being able to run multiple servers on the same port, or not being able to use protocols like SIP which encode the address in each packet.
If you have enough addresses that everyone can have one, there is absolutely no reason why they shouldn't.