The problem is that the technology hasn't caught up yet with the demand... but it's getting there.
The simple fact is that I agree with GP; converged devices are the future. Currently I have both an iPod and a Windows Mobile device (that I am using more and more like a laptop all the time). I'd rather carry one device, but Windows Media Player sucks... but it sucks TODAY. Who knows what tomorrow may bring?
The reason converged devices become a monstrosity are that there is too much going on at once and the processors in current devices just aren't really powerful enough to support it. That's changing though, per Moore's Law. Of course, they're WAY behind a desktop or laptop PC, but that's mostly because of the performance per watt thing... and that's because battery technology dictates how much power you can carry in your pocket. These things will change, they will improve.
I really love my WinMo device... warts and all. Hopefully later this month or early next month I'll buy myself an HTC Touch Pro to replace my aging TyTN and that'll get me one step closer to that converged device I really want. I doubt it'll replace my iPod any time soon, but it might relegate my laptop to less frequent duty.
At the end of the day, it's a phone and PDA, not a notebook.
Which is an idea that I reject absolutely. What we're dealing with her is neither a phone and PDA, nor a notebook. What we're dealing with her is a portable computing platform the exceeds both the capabilities and portability of a laptop while also exceeding the functional capabilities of either a phone or PDA.
What we're seeing here is a transition to truly converged devices that do everything. The technology isn't really there yet to make the truly converged devices I can imagine, but it will come. Like it or not, people want to carry one device around, not five.
I have been a WinMo geek for years... I know the platform has horrible warts but I still can't bring myself to move to anything else... even Android. It just gives me everything I need today... and besides I can see the POTENTIAL of the platform despite its warts.
Now, (returning a little to the subject at hand) even I considered recently buying an iPhone because I think the platform's slick, and I considered an Android phone as well... yet I find myself going back to Windows Mobile because I still feel it's closer to the leading edge of the truly converged devices than either of these other two platforms. Of course, Symbian has been even closer in recent years, but has fallen behind because of its rather crappy support of 3rd party developers and a rather confusing SDK. Here, WinMo beats it out, as do the other two big platforms.
I digress... I feel that I see the true potential of the platforms out there because I USE them as such. In fact, I use my laptop less and less these days particularly since I bought a Redfly recently that I have been using tethered to my original TyTN. In many ways, it sucks because there's not really enough memory in the TyTN to support the applications at 800x480, but I can see where the platform can go here. I can do 95% of what I need to do on my laptop, on my phone! And there are things I can do with the phone I can't with the laptop (like make calls). Here is the converged device I've always wanted.
I grant you, this is two devices, but both are small and the Redfly appears like a Netbook to the unitiated... I've used it in meetings to take notes and have been asked if it's a competitor to the EePC... only to have people stunned by the instant "boot" to the desktop when I power it on. Of course, the phone's already booted and on my hip but that amount of functionality is really key in my work.
My point really is that what we're seeing in the iPhone, Android and in the upcoming WinMo devices is the beginning of a new breed of truly converged devices. Sure, we've been seeing them coming for years, but I think only now are we starting to see the signs of maturity.
As processing power increases per watt, and battery life improves we're going to see this more and more... and I love the concept of devices like the Redfly (though not totally convinced on the price). Things like this and third party applications are rendering laptops less and less relevant except for working with large databases and/or graphical applications. Even this I think will change over time as the technology improves and I think we'll see a day soon when we'll all carry one device that does everything... maybe with one or more external devices that hook up wirelessly to that device to aid in working with larger files, databases or graphical apps. The Redfly and WinMo are just the beginning...... and back on topic... that's what's going to destroy the iPod eventually. It's a single-use device and its position will be usurped by the converged device. And yes, there's plenty of room in the market for Windows Mobile, iPhone, Android and Symbian... they all appeal to different markets. Yes, I know there are others... and there's room for those too. However, I do see Android as being more of a risk to OpenMoko than anything else... the iPhone will probably for the forseeable future be a consumer device / executive toy and Windows Mobile will be targeted primarily at the professional market. How those markets will converge... that's also going to be interesting.
I can't believe I'm going to bite on this... but that would be a water hypothesis... not a theory. It's not provable or disprovable (or even testable) and therefore is not a theory.
Plate tectonics is a theory... we test it often. For example, we're using sensors to "watch" California trying to escape to Alaska.
If you run a Datacenter off of Hardware you bought at Fry's, I don't want to be near it when it blows up. x86 hardware isn't all cheap, especially if you're thinking of a solid storage solution. Think stuff like HP XP arrays. Disks are the most fragile things, we swap at least a few per week where I work, there's no way we're running 1 SATA drive off the local controller for anything.
This is a fair comment, but in fairness to the GP, I think he was just reaching to make a point.
The disparity between the cost of an IBM solution as opposed to a VMware solution for similar needs is quite large.
Even you bring up the disks. Yes, they're fragile... but nothing a decent array can't fix. We've fired up virtual machines on stand-alone hardware with a nice array controller and some SAS drives... and we've set them up on big, meaty SANs. The cost/performance and reliability ratio for even a decent SAN (HP EVA) beats the hell out of a similar solution for a mainframe. I think that was GP's point:)
Of course, the advantage to the x86 way of doing things is that you can buy off-the-shelf hardware (or re-purpose old hardware) for the purposes of doing this kind of high availability. For all intents and purposes the result is a very stable and available system that costs less than what IBM are selling as a packaged solution.
A little disclaimer; I'm an old UNIX geek, certified in Solaris and AIX, but since I live in a Windows world at work these days I find myself as the SME for VMware at my company (just got back from VMworld as a matter of fact). As a result yes I do have a vested interest in x86 virtualization.
However, even given that I do find what VMware is doing today to be highly impressive. Yes, it's what IBM, DEC and others have been doing for decades, but this is a nicer solution for companies who don't want to be beholden to a single vendor, who want to have the availability and dynamic provisioning that have been around for years in higher end systems but don't want to pay through the nose.
I am still a firm believer in IBM's products... the mainframe still has its place and I see no reason that virtualized x86 systems will replace that any time soon... however I also see areas where these x86 systems excel that a mainframe has trouble competing in; namely the off-the-shelf applications that many SMB's (small-medium businesses) run.
As an example of how the VMware model works today; we have a smallish farm of VMware hosts that I manage. Recently we decided we had a need to build out a VM "stack" at our collo facility in order to remove about 30 physical machines... both to save space (and therefore money) and because all of them were coming due for maintenance. Instead of having to go out to IBM and price out a midrange or mainframe system to virtualize the applications (which was pointless anyway since these were all Windows based systems) we re-purposed three HP DL360 G5's with quad-core CPU's... added quad-port NICs and then just purchased an iSCSI-based HP MSA2000 (plus a couple of gig switches). Add some VMware licenses to the price tag and we were able to P2V all those 30 systems in the space of a week.
This kind of flexibility and ability to re-purpose hardware we already had keeps our costs down, keeps us nimble and allows us to be more dynamic to the business needs. Yes, a mainframe would allow us to do that, too... but having been an AIX guy (and worked a lot with mainframe guys) I know how tricky it can be to work with IBM hardware... particularly because if something breaks you need to go to IBM to fix it.
VMware Infrastructure is their management suite; VirtualCenter. That's their cash cow, not the hypervisor.
They give away the hypervisor for free... have done for a while. They know they can't compete just selling the hypervisor because everyone and their mother has one these days.
And FYI, the advanced technologies such as VMotion are part of the whole "Virtual Infrastructure" suite... the VI suite is what you buy, ESX is part of that suite. The simple fact is that VMware have already toyed with releasing ESX itself for free and may well do so.
Also, having used all sorts of virtualization technologies, I have to say that VMware still has the best solution even just with the standalone ESX. It's far more scalable than the competition, but add in the rest of the VI suite (Vmotion, VirtualCenter, HA, DRS and so forth) and then invest in the wealth of third party tools (free and non-free) and you have a kick-ass scalable, manageable and dependable environment.
I know at least one very large company that has invested a ton of money in a true virtual infrastructre of massive scale... and they're now reaping the rewards in scalability for minimal cost, and management.
But don't you think the very fact that the BBC is government funded shows its bias on its sleeve?
Besides, my experience with the BBC (having lived in the UK all of my life up until my 20's when I moved to the US) was that the bias you cite only occurred in a small minority of BBC programming. Most of the programming is fairly unbiased and tends to annoy the opposition party at least as much as the incumbent.
The BBC news is pretty much one of the best sources for relatively unbiased news in the world, as sad as that state of affairs may seem. Even the more ad-and-corporate-friendly BBC News America is a better news source than just about anything else on TV.
Thing is about #3 is that I don't really think it's relevant to this situation. If he had millions of dollars stashed in a non-extradition country, then he would have had to have been stupid (or #1, or #2) not to just take the 2 year stint in Club Fed, then run. Now he's going to have a lot more problems.
And right there we have the beauty of an OSS router OS that's under development by people who really understand what geeks want from their router! The ability to modify these sorts of things.
Though just out of curiosity when I saw this post I just had to go check what I had mine set to. I never had a problem, but I figured it wouldn't hurt to try. Apparently when I configured it, I maxed it out at 4096... never had an issue:)
I bought a WRT54GS and kept it stock for a while. I think as I recall it was a version 4... the one with more memory. Anyway, I had some problems with it requiring occasional resets. More often than not it was the wireless that crapped, not the router itself... at least in my case. It was like the WAP would just suddenly vanish and I'd have to power-cycle to get it back.
Eventually, because I got Vonage I wanted to implement some QoS rules which the stock firmware on the GS at the time didn't support. I hunted around and eventually put DD-WRT on the device... and voila! Currently, having just logged onto it for the first time in a while I see that my uptime is showing as 381 days. Pretty much once I'd configured it, I haven't had to mess with it since. Hmm... wonder if I should reboot it for old time's sake;)
Although I agree that bikes are cheaper to buy than cars, and take less space to "operate" & park, they are far from an ideal solution IMHO.
As someone who rides a bike most days, let me tell you my opinion. Now, it's an opinion so take it with a requisite grain (cup?) of salt...
* consumption isn't all THAT low from what I hear from my 2-wheeled-colleagues
Well, I ride a big bike; a Kawasaki Concours 14. The thing's huge and has a 1.4L engine. It produces around 155hp at the higher end of the tach. I get around 45mpg in mixed riding just commuting back and forth to work. Worst gas mileage I've seen was around 32mpg (but I was also carving up some twisties in low gears).
Is that bad? Not really. My cube-neighbor gives me crap because he says he can get that in his Prius... at least until I point out that he can't out-accelerate a Maserati in his Prius:D
I guess it's a perspective thing. I like getting 45mpg because I ENJOY the process of riding my bike instead of being trapped in a car. I love the freedom, I love the visceral pleasure I get from the bike and I enjoy the social aspects of being a biker. I've owned cars with great communities (my old Subaru SVX had a great community), but there's nothing that matches the community of bikers who don't care what you ride so long as you ride. The pretty damned decent gas-mileage is just a small benefit in my book that helped justify the initial cost of the bike... and the insane amounts of power (which are quite docile at low RPMS where I do much of my riding) are just the icing on the cake when I want it or need it.
* it might be nice in warm / dryish countries, I for one don't look forward to arriving all drenched at work
I've ridden in the rain, and never gotten more than my helmet wet. At least until I stopped and got off the bike to walk into the office. With a decently designed fairing the rain just goes around you.
In a crunch, you can always get a rain suit. I keep one in my panniers all the time just in case the heavens open (I live in the Midwest... you never know around here!) If it starts to get really heavy then I can stop under a bridge and put on the rain suit.
Besides, biking is partly a different state of mind. When I ride, I'm never rushed. If the heavens open, there's nothing stopping me rolling into a coffee shop to wait it out (presuming it'll stop soon), or take a route that allows you to stop the minimum number of times so you keep the air flowing around you with the rain. Just a different way of thinking.
Besides, I actually like the sound of raindrops on my helmet:)
* I for one feel quite a bit more safe being surrounded by a steel cage & airbags-combination
Again, a difference in perspective. Engineering in a car is all about surviving an accident... engineering in a bike is all about avoiding it.
I've had some near-misses on my bike that would've been nasty wrecks in my car (my car's a BMW by the way in case you're wondering... quite a safe vehicle). I've had people lose traction or even have a wreck in front of me... leaning hard over can allow you to go around just about anything... and in a worst case scenario just head off into the grass by the road (done it and lived to tell the tale!) It just requires a lot more attention than driving... but then that's one reason it appeals to me in the first place. I thrive on the attention I have to give riding... I get a kick from focus.
In my life I've had an accident that totaled my car once, and one that totaled a bike once. When I totaled the bike, I stood up, called a friend and asked him to come help me pick up the bits of my bike. When I had a similar accident in a car many years ago, I spent 45 minutes upside down inside my car with a head injury until the fire department cut me out of it; it was also in the country so it took a bit before anyone noticed me. I considered myself lucky in both instan
According to Jobs, PA Semi is for embedded devices... Also, according to Steve Jobs and a number of other people, actually... the iPhone and iPod Touch are running OSX. Right now Apple is maintaining three code-bases for OSX; PPC, x86 and ARM. Doesn't it make sense that if they're planning PA Semi chips for future iterations of embedded devices, they might want to drop the ARM architecture and switch to a PPC arch... particularly since they now have a stake in that market?
I think it's unlikely they'll drop PPC support any time soon... not least of which is the fact that Apple was still shipping PPC based Mac Minis just over 2 years ago, and PPC-based G5's less than two years ago (August 2006 they were discontinued). Not sure when the iMac G5's were discontinued... but it was almost certainly around the same time-frame.
Given Apple's recent moved in the embedded space with PA Semi, and their interest in going the embedded route with their systems (which is why we'll never see a mid-range Mac desktop machine) it seems unlikely they're going to drop PPC support. If they're going to go embedded PPC with OSX, then why stop developing now just to re-start later?
The x86 architecture is just a sop to the fact that the PPC architecture really fell behind in development when it came to the demands of desktop users. Users demanded the speed and horsepower that x86 gave them... so Apple relented and produced an x86 product. We'll still see x86 in the Apple computers for a long time (I think), but on the smaller end of the scale it seems odd to me that a company with a stake in a PPC embedded processor developer wouldn't be interested in leveraging that relationship in their embedded devices.
The reasons not to use x86 in an embedded device are legion, by the way.
No, I don't want to hear that Adobe lost the code or anything stupid like that. That's just inconceivable. I do not think that word means what you think it means...
Have you dealt with Adobe much? I mean, dealt with their coders?
Uhm... the G5 was the last PowerPC... what... two years ago?
Sorry... I don't even own a G5 but I would still call foul if Apple dropped PPC support. There's still a lot of people who run G5's (and G4's, but that's another matter). I'd say that we're still at least three years from the dropping of PPC support. But then again, I'm not Apple. I don't even play Apple on TV:)
Ironically, I may be one of the exceptions to the rule.
I bought a first gen Macbook Pro 15"; 2Ghz Core Duo. It came with the originally shipping Tiger. I didn't have major problems, but I did have plenty of issues with Tiger including rather flaky Bluetooth support and really unreliable tethering to my phone.
I bought Leopard after about two weeks, mostly because toward the end Tiger had become reasonably stable and I was cool with it. It was only after I'd seen Leopard running faster on a friend's identical machine than it had been running Tiger that I went ahead and bit the bullet.
I did an upgrade. No wipe and install, not even an archive and install. You know what? My machine has been more reliable, faster and generally better since I put Leopard on there. Even when I recently upgraded to a 320Gb drive, I used Carbon Copy Cloner to move it over, cruft and all. So far, I have seen the kernel panic crash once since I bought Leopard... I saw it at least every other month in Tiger. It may be because they fixed a lot of bugs in the suspend code (since I just close my laptop lid and almost never reboot unless I want to play games in my Vista partition).
Still, I'll probably take the path of least resistance with 10.6 whenever it does hit... whether next year or three years from now: An archive and install for me, I think. I know I may be the exception that proves the rule, but generally I've been very happy with Leopard.
Well said, sir. Having ALSO grown up in Belfast (hey, wonder if we ever met!) I know the exact bridge you're talking about. I lived in a different flashpoint area (Manor Street) where catholics and protestants lived at opposite ends of one street and met in the middle... right where my house used to be! However, I doubt the cameras would have made a huge amount of difference. Throughout most of the year we had a police landrover parked at the end of the street on a rotating shift, and that tended to just move the violence to parallel streets rather than actually stop it. People always find a way to stick it to another, as much as I hate to admit it sometimes.
Having said that, I did see a huge difference when I was at school when they started putting the cameras on the buses. Violence on the buses did seem to dissipate, but didn't entirely stop. The extreme violence (burning the bus) seemed to continue unabated, but the generalized "beat up the opposite sides" kind of violence was lessened visibly... at least while on the bus. Of course, being ~14 we figured out ways to tell when one was real or fake... but that's another matter.
Now, having also lived some time in London I wonder about the efficacy of many of these cameras because most of the ones I saw while there were either obviously non-functioning replicas, or vandalized... or often, both. I remember seeing quite a few of the cameras in London, but particularly after dark it didn't seem to make a significant impact on the crimes being committed. If anything, the criminals just started wearing clothing like hoodies that made it more difficult to identify the criminal... they still did the same thing.
I guess my point is that the advantage of the cameras is somewhat debatable. I've seen some good, but I've seen bad as well. As I mentioned above, the petty crime was somewhat impacted by the cameras that started to become ubiquitous when I was a teenager, but the truly extreme and violent crimes... well, they stayed about the same.
Although not absolute, I did return to the UK in 2005 for the first time in 10 years (been back twice more, since... but that's another story). I did notice a lot more surveillance than there had been when I left... or maybe 10 years living in the Midwest USA had sensitized me more to the existence of the cameras. Either way, crime was still a big problem and that didn't seem to change. I saw the signs of crime everywhere I went, even in the small towns on the back roads that otherwise seemed idyllic.
I take for example the fact that I rode a motorcycle in the UK back in 2005. Everwhere I parked, I had to chain it to something, as did all the other riders around and often I had a tough time finding a space to chain my bike. Here in the Midwest with less to almost non-existent surveillance, I have never needed to chain my bike to anything. I also know extremely few people who do. Of course, far fewer people ride motorbikes here than in the UK (relatively speaking), so it may just be a "target" thing.
In some instances, cameras do help... but generally I think their use as either a crime solving tool or crime deterrent is questionable at best. It "calms the yobbos", but doesn't seem to do much else... and even then it's only until they figure out how to sneak up "behind" the camera and vandalize it before they start beating the snot out of each other.
You barely need to make the leap from bacteria to human. Apes are barely capable of understanding us, either and they're a lot close both in time and evolution to us than bacteria.
Of course, we don't really understand them as well as some people would like to claim, either:)
Given that I was ripping out token ring from banks as recently as 2005... yeah. Big organizations like that don't move quickly, and TR was the default (and at one point only) option available for connecting AS/400 and Mainframe systems to a LAN. It was only in the mid 90's that it started to all switch to Ethernet, to the sound of much wailing and gnashing of teeth from the Mainframe and 400 guys and gals.
Also bear in mind, many mainframes and 400's installed at the end of this TR era may actually still be in use today in many locations... some of them so loaded up with cards that they are incapable of accepting an Ethernet interface. In our client/server-centric world we sometimes forget that there are systems out there with a much longer shelf-life than the 3-5 year x86 system running Windows or Linux.
How did this work with client machines that only had Ethernet? Quite well. TCP/IP is a very versatile protocol, quite happy to go across a set of routers from one architecture to another. As a result, even up until 2005 I touched on Token Ring systems in datacenters that used routers and gateways to get to their Ethernet-connected clients.
Seriously, I went through the same thing as the submitter a few years back. I felt disillusioned with IT and was really ready to pack it all in. I felt depressed, down and to be honest my wife and kids didn't really want to be around me so much any more because I was such a downer to be around. As it was, all it took was an "in-field career change".
I went from an administrator to a systems architect. OK, a bit of administration thrown in, too but minimal. To be honest, the pay increase was marginal (actually, I think I took a slight drop in pay initially), and the hours are and were actually worse than being an admin... but here a few years into this job I'm much happier, earning more than I was as an admin (even relatively speaking), and actually pretty happy.
Sys Architect IS more demanding. It requires an in-depth knowledge of the technologies and their interactions with other technologies. It also requires in-depth knowledge of the business in order to architect solutions that fit the business. But it does have its advantages:
1. I get to have my hands on new technology constantly. 2. I get more involved in business decisions that help drive our department and profitability 3. I get to mentor without having to be a manager... I love to teach 4. I don't get called at 4am because there's a problem to reboot a system. If I get called, there's REALLY a problem that a quick fix won't remedy and the system needs redesigned.
Thankfully, with item 4, I've never been called on one of my designs. It does mean you spend a lot of time with Visio though...;)
Slashdot Mirror
The problem is that the technology hasn't caught up yet with the demand... but it's getting there.
The simple fact is that I agree with GP; converged devices are the future. Currently I have both an iPod and a Windows Mobile device (that I am using more and more like a laptop all the time). I'd rather carry one device, but Windows Media Player sucks... but it sucks TODAY. Who knows what tomorrow may bring?
The reason converged devices become a monstrosity are that there is too much going on at once and the processors in current devices just aren't really powerful enough to support it. That's changing though, per Moore's Law. Of course, they're WAY behind a desktop or laptop PC, but that's mostly because of the performance per watt thing... and that's because battery technology dictates how much power you can carry in your pocket. These things will change, they will improve.
I really love my WinMo device... warts and all. Hopefully later this month or early next month I'll buy myself an HTC Touch Pro to replace my aging TyTN and that'll get me one step closer to that converged device I really want. I doubt it'll replace my iPod any time soon, but it might relegate my laptop to less frequent duty.
At the end of the day, it's a phone and PDA, not a notebook.
Which is an idea that I reject absolutely. What we're dealing with her is neither a phone and PDA, nor a notebook. What we're dealing with her is a portable computing platform the exceeds both the capabilities and portability of a laptop while also exceeding the functional capabilities of either a phone or PDA.
What we're seeing here is a transition to truly converged devices that do everything. The technology isn't really there yet to make the truly converged devices I can imagine, but it will come. Like it or not, people want to carry one device around, not five.
I have been a WinMo geek for years... I know the platform has horrible warts but I still can't bring myself to move to anything else... even Android. It just gives me everything I need today... and besides I can see the POTENTIAL of the platform despite its warts.
Now, (returning a little to the subject at hand) even I considered recently buying an iPhone because I think the platform's slick, and I considered an Android phone as well... yet I find myself going back to Windows Mobile because I still feel it's closer to the leading edge of the truly converged devices than either of these other two platforms. Of course, Symbian has been even closer in recent years, but has fallen behind because of its rather crappy support of 3rd party developers and a rather confusing SDK. Here, WinMo beats it out, as do the other two big platforms.
I digress... I feel that I see the true potential of the platforms out there because I USE them as such. In fact, I use my laptop less and less these days particularly since I bought a Redfly recently that I have been using tethered to my original TyTN. In many ways, it sucks because there's not really enough memory in the TyTN to support the applications at 800x480, but I can see where the platform can go here. I can do 95% of what I need to do on my laptop, on my phone! And there are things I can do with the phone I can't with the laptop (like make calls). Here is the converged device I've always wanted.
I grant you, this is two devices, but both are small and the Redfly appears like a Netbook to the unitiated... I've used it in meetings to take notes and have been asked if it's a competitor to the EePC... only to have people stunned by the instant "boot" to the desktop when I power it on. Of course, the phone's already booted and on my hip but that amount of functionality is really key in my work.
My point really is that what we're seeing in the iPhone, Android and in the upcoming WinMo devices is the beginning of a new breed of truly converged devices. Sure, we've been seeing them coming for years, but I think only now are we starting to see the signs of maturity.
As processing power increases per watt, and battery life improves we're going to see this more and more... and I love the concept of devices like the Redfly (though not totally convinced on the price). Things like this and third party applications are rendering laptops less and less relevant except for working with large databases and/or graphical applications. Even this I think will change over time as the technology improves and I think we'll see a day soon when we'll all carry one device that does everything... maybe with one or more external devices that hook up wirelessly to that device to aid in working with larger files, databases or graphical apps. The Redfly and WinMo are just the beginning... ... and back on topic... that's what's going to destroy the iPod eventually. It's a single-use device and its position will be usurped by the converged device. And yes, there's plenty of room in the market for Windows Mobile, iPhone, Android and Symbian... they all appeal to different markets. Yes, I know there are others... and there's room for those too. However, I do see Android as being more of a risk to OpenMoko than anything else... the iPhone will probably for the forseeable future be a consumer device / executive toy and Windows Mobile will be targeted primarily at the professional market. How those markets will converge... that's also going to be interesting.
I can't believe I'm going to bite on this... but that would be a water hypothesis... not a theory. It's not provable or disprovable (or even testable) and therefore is not a theory.
Plate tectonics is a theory... we test it often. For example, we're using sensors to "watch" California trying to escape to Alaska.
If you run a Datacenter off of Hardware you bought at Fry's, I don't want to be near it when it blows up. x86 hardware isn't all cheap, especially if you're thinking of a solid storage solution. Think stuff like HP XP arrays. Disks are the most fragile things, we swap at least a few per week where I work, there's no way we're running 1 SATA drive off the local controller for anything.
This is a fair comment, but in fairness to the GP, I think he was just reaching to make a point.
The disparity between the cost of an IBM solution as opposed to a VMware solution for similar needs is quite large.
Even you bring up the disks. Yes, they're fragile... but nothing a decent array can't fix. We've fired up virtual machines on stand-alone hardware with a nice array controller and some SAS drives... and we've set them up on big, meaty SANs. The cost/performance and reliability ratio for even a decent SAN (HP EVA) beats the hell out of a similar solution for a mainframe. I think that was GP's point :)
Pity that they're not, isn't it?
They're running multiple OS's... one per computer. They ACT as one because of the management infrastructure that's in place. Kind of like a mainframe.
Don't get me wrong, they're doing some VERY interesting stuff with that, but an OpenMOSIX rip it ain't.
Of course, the advantage to the x86 way of doing things is that you can buy off-the-shelf hardware (or re-purpose old hardware) for the purposes of doing this kind of high availability. For all intents and purposes the result is a very stable and available system that costs less than what IBM are selling as a packaged solution.
A little disclaimer; I'm an old UNIX geek, certified in Solaris and AIX, but since I live in a Windows world at work these days I find myself as the SME for VMware at my company (just got back from VMworld as a matter of fact). As a result yes I do have a vested interest in x86 virtualization.
However, even given that I do find what VMware is doing today to be highly impressive. Yes, it's what IBM, DEC and others have been doing for decades, but this is a nicer solution for companies who don't want to be beholden to a single vendor, who want to have the availability and dynamic provisioning that have been around for years in higher end systems but don't want to pay through the nose.
I am still a firm believer in IBM's products... the mainframe still has its place and I see no reason that virtualized x86 systems will replace that any time soon... however I also see areas where these x86 systems excel that a mainframe has trouble competing in; namely the off-the-shelf applications that many SMB's (small-medium businesses) run.
As an example of how the VMware model works today; we have a smallish farm of VMware hosts that I manage. Recently we decided we had a need to build out a VM "stack" at our collo facility in order to remove about 30 physical machines... both to save space (and therefore money) and because all of them were coming due for maintenance. Instead of having to go out to IBM and price out a midrange or mainframe system to virtualize the applications (which was pointless anyway since these were all Windows based systems) we re-purposed three HP DL360 G5's with quad-core CPU's... added quad-port NICs and then just purchased an iSCSI-based HP MSA2000 (plus a couple of gig switches). Add some VMware licenses to the price tag and we were able to P2V all those 30 systems in the space of a week.
This kind of flexibility and ability to re-purpose hardware we already had keeps our costs down, keeps us nimble and allows us to be more dynamic to the business needs. Yes, a mainframe would allow us to do that, too... but having been an AIX guy (and worked a lot with mainframe guys) I know how tricky it can be to work with IBM hardware... particularly because if something breaks you need to go to IBM to fix it.
VMware Infrastructure is their management suite; VirtualCenter. That's their cash cow, not the hypervisor.
They give away the hypervisor for free... have done for a while. They know they can't compete just selling the hypervisor because everyone and their mother has one these days.
And FYI, the advanced technologies such as VMotion are part of the whole "Virtual Infrastructure" suite... the VI suite is what you buy, ESX is part of that suite. The simple fact is that VMware have already toyed with releasing ESX itself for free and may well do so.
Also, having used all sorts of virtualization technologies, I have to say that VMware still has the best solution even just with the standalone ESX. It's far more scalable than the competition, but add in the rest of the VI suite (Vmotion, VirtualCenter, HA, DRS and so forth) and then invest in the wealth of third party tools (free and non-free) and you have a kick-ass scalable, manageable and dependable environment.
I know at least one very large company that has invested a ton of money in a true virtual infrastructre of massive scale... and they're now reaping the rewards in scalability for minimal cost, and management.
But don't you think the very fact that the BBC is government funded shows its bias on its sleeve?
Besides, my experience with the BBC (having lived in the UK all of my life up until my 20's when I moved to the US) was that the bias you cite only occurred in a small minority of BBC programming. Most of the programming is fairly unbiased and tends to annoy the opposition party at least as much as the incumbent.
The BBC news is pretty much one of the best sources for relatively unbiased news in the world, as sad as that state of affairs may seem. Even the more ad-and-corporate-friendly BBC News America is a better news source than just about anything else on TV.
Thing is about #3 is that I don't really think it's relevant to this situation. If he had millions of dollars stashed in a non-extradition country, then he would have had to have been stupid (or #1, or #2) not to just take the 2 year stint in Club Fed, then run. Now he's going to have a lot more problems.
I vote for #4... stupid.
And right there we have the beauty of an OSS router OS that's under development by people who really understand what geeks want from their router! The ability to modify these sorts of things.
Though just out of curiosity when I saw this post I just had to go check what I had mine set to. I never had a problem, but I figured it wouldn't hurt to try. Apparently when I configured it, I maxed it out at 4096... never had an issue :)
For the record... similar experience here.
I bought a WRT54GS and kept it stock for a while. I think as I recall it was a version 4... the one with more memory. Anyway, I had some problems with it requiring occasional resets. More often than not it was the wireless that crapped, not the router itself... at least in my case. It was like the WAP would just suddenly vanish and I'd have to power-cycle to get it back.
Eventually, because I got Vonage I wanted to implement some QoS rules which the stock firmware on the GS at the time didn't support. I hunted around and eventually put DD-WRT on the device... and voila! Currently, having just logged onto it for the first time in a while I see that my uptime is showing as 381 days. Pretty much once I'd configured it, I haven't had to mess with it since. Hmm... wonder if I should reboot it for old time's sake ;)
Although I agree that bikes are cheaper to buy than cars, and take less space to "operate" & park, they are far from an ideal solution IMHO.
As someone who rides a bike most days, let me tell you my opinion. Now, it's an opinion so take it with a requisite grain (cup?) of salt...
* consumption isn't all THAT low from what I hear from my 2-wheeled-colleagues
Well, I ride a big bike; a Kawasaki Concours 14. The thing's huge and has a 1.4L engine. It produces around 155hp at the higher end of the tach. I get around 45mpg in mixed riding just commuting back and forth to work. Worst gas mileage I've seen was around 32mpg (but I was also carving up some twisties in low gears).
:D
Is that bad? Not really. My cube-neighbor gives me crap because he says he can get that in his Prius... at least until I point out that he can't out-accelerate a Maserati in his Prius
I guess it's a perspective thing. I like getting 45mpg because I ENJOY the process of riding my bike instead of being trapped in a car. I love the freedom, I love the visceral pleasure I get from the bike and I enjoy the social aspects of being a biker. I've owned cars with great communities (my old Subaru SVX had a great community), but there's nothing that matches the community of bikers who don't care what you ride so long as you ride. The pretty damned decent gas-mileage is just a small benefit in my book that helped justify the initial cost of the bike... and the insane amounts of power (which are quite docile at low RPMS where I do much of my riding) are just the icing on the cake when I want it or need it.
* it might be nice in warm / dryish countries, I for one don't look forward to arriving all drenched at work
I've ridden in the rain, and never gotten more than my helmet wet. At least until I stopped and got off the bike to walk into the office. With a decently designed fairing the rain just goes around you.
:)
In a crunch, you can always get a rain suit. I keep one in my panniers all the time just in case the heavens open (I live in the Midwest... you never know around here!) If it starts to get really heavy then I can stop under a bridge and put on the rain suit.
Besides, biking is partly a different state of mind. When I ride, I'm never rushed. If the heavens open, there's nothing stopping me rolling into a coffee shop to wait it out (presuming it'll stop soon), or take a route that allows you to stop the minimum number of times so you keep the air flowing around you with the rain. Just a different way of thinking.
Besides, I actually like the sound of raindrops on my helmet
* I for one feel quite a bit more safe being surrounded by a steel cage & airbags-combination
Again, a difference in perspective. Engineering in a car is all about surviving an accident... engineering in a bike is all about avoiding it.
I've had some near-misses on my bike that would've been nasty wrecks in my car (my car's a BMW by the way in case you're wondering... quite a safe vehicle). I've had people lose traction or even have a wreck in front of me... leaning hard over can allow you to go around just about anything... and in a worst case scenario just head off into the grass by the road (done it and lived to tell the tale!) It just requires a lot more attention than driving... but then that's one reason it appeals to me in the first place. I thrive on the attention I have to give riding... I get a kick from focus.
In my life I've had an accident that totaled my car once, and one that totaled a bike once. When I totaled the bike, I stood up, called a friend and asked him to come help me pick up the bits of my bike. When I had a similar accident in a car many years ago, I spent 45 minutes upside down inside my car with a head injury until the fire department cut me out of it; it was also in the country so it took a bit before anyone noticed me. I considered myself lucky in both instan
I think it's unlikely they'll drop PPC support any time soon... not least of which is the fact that Apple was still shipping PPC based Mac Minis just over 2 years ago, and PPC-based G5's less than two years ago (August 2006 they were discontinued). Not sure when the iMac G5's were discontinued... but it was almost certainly around the same time-frame.
Given Apple's recent moved in the embedded space with PA Semi, and their interest in going the embedded route with their systems (which is why we'll never see a mid-range Mac desktop machine) it seems unlikely they're going to drop PPC support. If they're going to go embedded PPC with OSX, then why stop developing now just to re-start later?
The x86 architecture is just a sop to the fact that the PPC architecture really fell behind in development when it came to the demands of desktop users. Users demanded the speed and horsepower that x86 gave them... so Apple relented and produced an x86 product. We'll still see x86 in the Apple computers for a long time (I think), but on the smaller end of the scale it seems odd to me that a company with a stake in a PPC embedded processor developer wouldn't be interested in leveraging that relationship in their embedded devices.
The reasons not to use x86 in an embedded device are legion, by the way.
Have you dealt with Adobe much? I mean, dealt with their coders?
My quote stands...
But... but... the Maus would only have one button!!!
Uhm... the G5 was the last PowerPC... what... two years ago?
:)
Sorry... I don't even own a G5 but I would still call foul if Apple dropped PPC support. There's still a lot of people who run G5's (and G4's, but that's another matter). I'd say that we're still at least three years from the dropping of PPC support. But then again, I'm not Apple. I don't even play Apple on TV
Ironically, I may be one of the exceptions to the rule.
I bought a first gen Macbook Pro 15"; 2Ghz Core Duo. It came with the originally shipping Tiger. I didn't have major problems, but I did have plenty of issues with Tiger including rather flaky Bluetooth support and really unreliable tethering to my phone.
I bought Leopard after about two weeks, mostly because toward the end Tiger had become reasonably stable and I was cool with it. It was only after I'd seen Leopard running faster on a friend's identical machine than it had been running Tiger that I went ahead and bit the bullet.
I did an upgrade. No wipe and install, not even an archive and install. You know what? My machine has been more reliable, faster and generally better since I put Leopard on there. Even when I recently upgraded to a 320Gb drive, I used Carbon Copy Cloner to move it over, cruft and all. So far, I have seen the kernel panic crash once since I bought Leopard... I saw it at least every other month in Tiger. It may be because they fixed a lot of bugs in the suspend code (since I just close my laptop lid and almost never reboot unless I want to play games in my Vista partition).
Still, I'll probably take the path of least resistance with 10.6 whenever it does hit... whether next year or three years from now: An archive and install for me, I think. I know I may be the exception that proves the rule, but generally I've been very happy with Leopard.
Yeah, but he's Welsh: All bets are off...
Well said, sir. Having ALSO grown up in Belfast (hey, wonder if we ever met!) I know the exact bridge you're talking about. I lived in a different flashpoint area (Manor Street) where catholics and protestants lived at opposite ends of one street and met in the middle... right where my house used to be! However, I doubt the cameras would have made a huge amount of difference. Throughout most of the year we had a police landrover parked at the end of the street on a rotating shift, and that tended to just move the violence to parallel streets rather than actually stop it. People always find a way to stick it to another, as much as I hate to admit it sometimes.
Having said that, I did see a huge difference when I was at school when they started putting the cameras on the buses. Violence on the buses did seem to dissipate, but didn't entirely stop. The extreme violence (burning the bus) seemed to continue unabated, but the generalized "beat up the opposite sides" kind of violence was lessened visibly... at least while on the bus. Of course, being ~14 we figured out ways to tell when one was real or fake... but that's another matter.
Now, having also lived some time in London I wonder about the efficacy of many of these cameras because most of the ones I saw while there were either obviously non-functioning replicas, or vandalized... or often, both. I remember seeing quite a few of the cameras in London, but particularly after dark it didn't seem to make a significant impact on the crimes being committed. If anything, the criminals just started wearing clothing like hoodies that made it more difficult to identify the criminal... they still did the same thing.
I guess my point is that the advantage of the cameras is somewhat debatable. I've seen some good, but I've seen bad as well. As I mentioned above, the petty crime was somewhat impacted by the cameras that started to become ubiquitous when I was a teenager, but the truly extreme and violent crimes... well, they stayed about the same.
Although not absolute, I did return to the UK in 2005 for the first time in 10 years (been back twice more, since... but that's another story). I did notice a lot more surveillance than there had been when I left... or maybe 10 years living in the Midwest USA had sensitized me more to the existence of the cameras. Either way, crime was still a big problem and that didn't seem to change. I saw the signs of crime everywhere I went, even in the small towns on the back roads that otherwise seemed idyllic.
I take for example the fact that I rode a motorcycle in the UK back in 2005. Everwhere I parked, I had to chain it to something, as did all the other riders around and often I had a tough time finding a space to chain my bike. Here in the Midwest with less to almost non-existent surveillance, I have never needed to chain my bike to anything. I also know extremely few people who do. Of course, far fewer people ride motorbikes here than in the UK (relatively speaking), so it may just be a "target" thing.
In some instances, cameras do help... but generally I think their use as either a crime solving tool or crime deterrent is questionable at best. It "calms the yobbos", but doesn't seem to do much else... and even then it's only until they figure out how to sneak up "behind" the camera and vandalize it before they start beating the snot out of each other.
You barely need to make the leap from bacteria to human. Apes are barely capable of understanding us, either and they're a lot close both in time and evolution to us than bacteria.
:)
Of course, we don't really understand them as well as some people would like to claim, either
Given that I was ripping out token ring from banks as recently as 2005... yeah. Big organizations like that don't move quickly, and TR was the default (and at one point only) option available for connecting AS/400 and Mainframe systems to a LAN. It was only in the mid 90's that it started to all switch to Ethernet, to the sound of much wailing and gnashing of teeth from the Mainframe and 400 guys and gals.
Also bear in mind, many mainframes and 400's installed at the end of this TR era may actually still be in use today in many locations... some of them so loaded up with cards that they are incapable of accepting an Ethernet interface. In our client/server-centric world we sometimes forget that there are systems out there with a much longer shelf-life than the 3-5 year x86 system running Windows or Linux.
How did this work with client machines that only had Ethernet? Quite well. TCP/IP is a very versatile protocol, quite happy to go across a set of routers from one architecture to another. As a result, even up until 2005 I touched on Token Ring systems in datacenters that used routers and gateways to get to their Ethernet-connected clients.
If I had mod points, you'd be getting 'em :)
;)
Seriously, I went through the same thing as the submitter a few years back. I felt disillusioned with IT and was really ready to pack it all in. I felt depressed, down and to be honest my wife and kids didn't really want to be around me so much any more because I was such a downer to be around. As it was, all it took was an "in-field career change".
I went from an administrator to a systems architect. OK, a bit of administration thrown in, too but minimal. To be honest, the pay increase was marginal (actually, I think I took a slight drop in pay initially), and the hours are and were actually worse than being an admin... but here a few years into this job I'm much happier, earning more than I was as an admin (even relatively speaking), and actually pretty happy.
Sys Architect IS more demanding. It requires an in-depth knowledge of the technologies and their interactions with other technologies. It also requires in-depth knowledge of the business in order to architect solutions that fit the business. But it does have its advantages:
1. I get to have my hands on new technology constantly.
2. I get more involved in business decisions that help drive our department and profitability
3. I get to mentor without having to be a manager... I love to teach
4. I don't get called at 4am because there's a problem to reboot a system. If I get called, there's REALLY a problem that a quick fix won't remedy and the system needs redesigned.
Thankfully, with item 4, I've never been called on one of my designs. It does mean you spend a lot of time with Visio though...