Why wait for Google. According to the Lightening 0.3.1 Release Notes there's a version you can download with integrated Sun Java System Calendar Server (WCAP) support. And in October you might have another option, given that Apple's Leopard iCal Server supports Sunbird, it should also support Lightening too. Or you could wait for it to be ported to Linux after they've released the source code, which they've promised to do.
I know I should have tested it on a test machine before bringing it into production. (or maybe waited a bit) But this is a small machine in an informal setting. I don't have a test machine. But I do have 20+ users with slow internet. and it's really not asking for too much to expect a thing so blatant.
Man, that's pretty reckless, and you know it. Did you even take a backup first? As for not having a test machine, with Xen and VMware are your disposal these days there's no real excuse for not installing it elsewhere, and at least taking a few days to give it the once over before going near a real server. The truth is that you rushed in without proper forethought and planning and you got burned.
I remember when I was at school there was an episode of the BBC show "Tomorrows World" where for an April Fools day prank they reported on the breakthrough invention of de-hydrated water tablets. You expose them to sunlight and they turned back into water. There was one particularly slow boy in class who had an argument with the chemistry teacher the next day, insisting that these things existed cos he'd seen it on TV.
This is really interesting for me. I didn't realise until reading this thread that the ESX hypervisor worked that way. Do you know what its capabilities are for (disk) volume management? For instance one solution I'm deploying at the moment for a customer is an HP-UX Campus Cluster, where the SAN storage and cluster stretches over two data centers on the same campus. The volumes are mirrored across both data centers using LVM, so loss of an entire data center doesn't take down the storage or the cluster. For future solutions I've been considering Xen VMs on RHEL 5 and I could potentially support that environment with similar disaster tolerant stretched clusters using CLVM.
Both of those are viable solutions because the OS on the metal can do RAID volume management. Do you know if ESX can too ?
The article is right. At best, high-tech CCTV has been used to identify people after the fact, in some cases but has done nothing to deter or prevent crimes
Bullshit. It's well documented and proven in the UK that areas well covered by CCTV have lower crime rates. Sure the crims still commit their crimes, they just do it elsewhere, but not as much in CCTV covered areas.
Also, the 21/7 failed London tube bombers were all identified and found because of CCTV before they could correct their mistake in their bomb design and try again. We got to see their faces all over national TV the very next day, and because of that the police were alerted to their location by a member of the public. There have been so many cases in the UK in recent years where CCTV footage has been pivotal in identifying the perpetrator, leading to their arrest.
Sure, this is all "after the fact", but which is worse, not catching the crim because you haven't a clue who they are, or catching them because you have their face on CCTV? As a victim you're going to feel a lot better knowing the ass wipe who turned you over the first time isn't lurking around the next corner to do it again. Bottom line, the UK public like CCTV and its here to stay. I wish we had it on my street.
That better not be true, because it's impossible. But we know that it it isn't true - Mac OS X and Windows Vista are far from flawless, and yet people still manage to muddle their way through using those systems. In fact, lots of people manage to use Ubuntu right now even with a couple of bugs.
The difference between Mac OS X and Ubuntu in this instance is that a user can use, setup and troubleshoot OS X without ever having to go near the command line. It's been designed that way because the majority of OS X users don't have a clue what to do with a command line shell, and don't want to know. If OS X can do it then so can Ubuntu, and that's what they should be aiming for.
Actually they've already addressed the safety question on their FAQ page where they say:
300 bars of compressed air stored on board the vehicle, Is this dangerous for the passengers?
Compressed air tanks have already been proven safe by one of our partners EADS(AIRBUS). This company's reputation in the aeronautical field is unprecedented, given the reliability of its tanks. What's more, the compressed air does not present any risk of explosion. Countless test have been carried out in the most extreme conditions (gun shoots, resistance to fire...) to guarantee passenger safety in every possible condition. The high pressure tanks have been developed using a similar technology to those used in natural gas vehicles and by firefighters. All are produced with carbon fiber over plastic.
The tanks that MDI puts in its vehicles are similar to those already in use in natural gas busses in Germany and other countries
Actually according to this page they're talking about a tank of 300 litres at 300 bar resulting in 46 MJ and also a 340 litre tank with 52.1 MJ. So you're a bit short on power. In addition they're measuring distance in km not miles.
What most people seem blissfully unaware of is that water vapor is also a 'greenhouse gas'. If all our cars were replaced tomorrow with vehicles that emit water vapor instead of their usual chemical cocktail, we would still be over contributing to the heating up of the planet.
No way in a million years will Dell or any other PC vendor offer Fedora pre-installed on a "noob-friendly" PC. Fedora is a test bed for new and interesting technologies that one day hope to make it into RH's production ready releases. No company in their right mind will risk that on their customers. Ubuntu is probably the only distro that's potentially noob-friendly enough for the job today.
I think this whole conversation is moot anyway. The author of the article hit the nail on the head over the cost of support. It's just too expensive for Dell to do it in-house. The only way they could make it work would be for Dell to outsource the support to another company already geared up to deliver it effectively.
So calling me a jackass and such instead of responding to the technical points I've made isn't ad hominem? I see, thanks for the english lesson (not).
You consistently ignore the technology use examples I give you with respect to their advantages to business because you have no answer. Instead you respond with laughable statements like "VM is still emulation" clearly demonstrating that you don't know the difference between either. Let me spell it out for you. Microsoft VirtualPC and QEMU are emulators because they emulate the x86 chip architecture on non-x86 platforms, as well as the PC hardware environment. Virtual Machines like VMware, Xen and Parallels are not because they run native and don't emulate chip set instructions. Their only purpose is to provide a high performance hardware abstraction layer that multiple operating systems can sit on, that's why they're called Virtual Machines, not Emulators (duh!).
They are so important to the future of this industry that both Intel and AMD have changed the design of their chips to accommodate them, implementing concepts e.g. VT technology (Intel) and in future, Extended Page Tables (Intel) a.k.a Nested Page Tables (AMD) and loads of other improvements are in the pipe line. Even the PCI Special Interest Group is working on changes to the PCI standard so that for example a network card's capacity can be split among different virtual machines.
According to you, all these pillars of industry are "silly" because VM's are "inefficient", "undesirable" and "to be avoided when possible". Maybe you should put your money where your mouth is and write a column article about this and tell them publicly why they're wasting their millions in development costs. Please do... I'd love to see the public kicking you'd get afterwards for spouting such garbage.
As for continuing here on Slashdot, you're right, you shouldn't bother. You're clearly unqualified to comment on what I'm talking about, as most of it seems to be going straight over your head. Why don't you give your Ego a rest and admit when you don't understand something instead of pretending to look clever by responding with bullshit. You think that access to data is the only thing that matters. LOL, you're an idiot. Do you know anything about running a business and why people pay for computers in the first place? Do you know anything about why businesses buy the systems they do and what technologies drive those decisions? Clearly not. You need to give up writing that column, go get yourself a real job in this industry and get some real world experience.
Well you posted under AC this time, but the style of writing is the same so I'll assume it's you.
I'm just really amazed at how out of touch you are with what enterprise customers want from their systems. "Thin Client" has nothing to do with access to data. I've already said that, repeatedly. It's about taming the beast when you have thousands of desktops to try and manage.
You throw phrases and terminologies around and make technical references to bolster your credibility, but people familiar with the limitations of Active Directory versus not only eDirectory but even versus free software implementations, read your raves over Active Directory and just shake their heads
I mentioned Active Directory once, just to give you a clue that centrally controlled user account administration is part of the demo I saw. Replace that with LDAP or Redhat's Directory server if you like, makes no difference to me. Given a choice I'd propose one of those first anyway.
As for virtual machines, VM is just a buzzy way of saying emulation and emulation is inefficient, always. A pragmatic person would prefer native implementations of software on his or her main platform and would see emulation as, at best, a necessary evil. And running a billion instances each of a hundred different emulated environments is real gee-whizzy-bang and all that, but you're best off not getting yourself into that expensive, inefficient, and really rather quite silly situation in the first place. So once again on your part, bad judgment.
LOL, funniest thing I've read in ages, you need to read more about VMs. You've completely missed what I've tried to tell you about average server utilisation in the industry and how customers are using VMs to consolidate their compute power to improve their ROI (Return on investment). That's what's its all about and what's driving it, plus the flexibility they get when they are able to live migrate VMs from one lump of hardware in a cluster to another. It has a massive impact on:
Resource allocation -- come end of month you can move your development VMs off the more powerful servers, freeing up compute resources so they can be added (on the fly) to the production VMs.
Hardware maintenance -- one of your systems starts reporting correctable memory errors. You know its just a matter of time before it fails and needs an engineer to replace the defective DIMM. So you live migrate the VMs off that system so the engineer can work on it. No interruption to your business.
Change control -- it's a big headache. You need to take a production service offline for maintenance etc and you don't use VMs, so a change control request has to be made and all affected business managers need to get involved. It has to be timed to happen when it's going to have the least impact (no batch jobs) and no one is using the system. That will mean either late at night or over the weekend (and you wanted to go fishing, tough). It's expensive, disruptive and takes a lot of time. I have loads of personal experience of this! Live migration of your VMs to other hardware will significantly reduce the number of times this has to happen, consequently improving application uptime, availability and saving money.
Upgrading the system -- you don't have enough compute resources to run your business, so you add new systems into the cluster, then redistribute your VMs across them. Job done, no downtime.
OS compatibility -- Your app is old and either the vendor doesn't exist any more or its a legacy app and there are no funds for a massive data migration (with all the UAC testing around that) to something newer. It's also not qualified to run on newer OS versions, but the old OS version its qualified for doesn't support the new hardware you have to run on. No problem. Your VM hypervisor runs happily on the new hardware and your old OS + app runs happily in a VM.
For those who are interested here are the release notes. The technology preview is particularly mouth watering. Personally I'm especially looking forward to GFS2.
Ok, gloves off. I've been very civil to you through out this dialogue, but now you're getting unnecessarily fucking rude. A few things you should know about this "jackass" before you go any further.
1) I started my career using VMS for about 4 years before switching to Unix and I've been a solid Unix man for the past 16 years. Today I specialise in designing, implementing, integrating, tuning and troubleshooting Unix cluster solutions based on Tru64, HP-UX and Linux. I've only ever owned a Windows laptop once, for about 2 years, and its was the smallest little VAIO I could get away with for use with MS Office and because it was before the days when Open Office was really that useful.
2) I switched from that little laptop to my first PowerBook in 2002, primarily because OS X ran on top of a flavour of Unix, and because it allowed me to ditch my work MS laptop and home Linux workstation and combine both functions on one platform. At that point I was self employed, though now I work with 3 other guys and I've used nothing but Mac OS X on my desktop (exclusively) for the past 5 years.
3) I'm a terrible Apple evangelist to all my family and friends and have converted a modest number in the last couple of years. I hope to do better in the future.
You, in your rabid MS bashing frenzy for anyone who doesn't share your OTT hatred of Gates and his spawn, can't see the fucking wood for the trees. And you think I'm some kind of MS fanboy! You're way off base pal.
What I am is a pragmatist. I would love it if I could pitch Apple solutions to my Enterprise customers, but as I've tried to tell you over and over, Apple do not tailor their products to sell into the Enterprise. You may be happy sitting there in your little bubble world, writing articles about how great Apple is and how crap MS are, but what you do not have is a finger on the Enterprise pulse and half a clue about the problems they are trying to solve?
Answer me this riddle then oh great and wise Oracle.
Enterprise businesses are turning to Virtual Machines in a big way. Everyone's doing it. MS is working on a VM server; VMware is one of the fastest growing software companies in the world right now; Novel SuSE ship and support Xen; Redhat will do the same in RHEL 5; HP have a VM for HP-UX; Sun have Solaris Domains, and there are other solutions out there like Virtuosso (aka OpenVZ) and KVM is in development and included in the Linux kernel source tree.
Most of them can run multiple OSes from different vendors (apart from Solaris Domains and OpenVZ) but they can all run multiple copies of their own native OS. Why? Because it's what Enterprise customers want.
Oh, hang on a minute. There's one exception to that rule. I wonder who that would be? Why (shock horror) it's Apple.
Apple stands alone on this. They have one VM implementation in production today (Parallels) and one in beta (VMware) yet they are to my knowledge the only VM capable OS vendor that don't allow their own OS to run inside VMs on their own platform.
Tell us then smart ass. Who in their right mind will buy an Xserve to run a single instance of OS X that's tied to that one box, when they can spend similar money to buy a similar box from HP, Dell or IBM and run a dozen specifically purposed VMs with Linux and Windows in them. VMs that can be shunted elsewhere on the fly to balance compute resources or free up hardware for maintenance courtesy of VMware's VMotion or Xen's Live Migration?
The fact that HP is stretching Windows instances beyond absurdity to create a solution that orbits around enriching Microsoft rather than solving a solution doesn't help your case.
Bullshit. It's got nothing to do with enriching MS, and it very much solves a problem (not a solution, duh!). I like thin client solutions. I've worked with them in the past (X windows based) and they are a much smarter way of running a corporate desktop than the fat client for everyone abortion that MS push
Ok, gloves off. I've been very civil to you through out this dialogue, but now you're getting unnecessarily fucking rude. A few things you should know about this "jackass" before you go any further.
1) I started my career using VMS for about 4 years before switching to Unix and I've been a solid Unix man for the past 16 years. Today I specialise in designing, implementing, integrating, tuning and troubleshooting Unix cluster solutions based on Tru64, HP-UX and Linux. I've only ever owned a Windows laptop once, for about 2 years, and its was the smallest little VAIO I could get away with for use with MS Office and because it was before the days when Open Office was really that useful.
2) I switched from that little laptop to my first PowerBook in 2002, primarily because OS X ran on top of a flavour of Unix, and because it allowed me to ditch my work MS laptop and home Linux workstation and combine both functions on one platform. At that point I was self employed, though now I work with 3 other guys and I've used nothing but Mac OS X on my desktop (exclusively) for the past 5 years.
3) I'm a terrible Apple evangelist to all my family and friends and have converted a modest number in the last couple of years. I hope to do better in the future.
You, in your rabid MS bashing frenzy for anyone who doesn't share your OTT hatred of Gates and his spawn, can't see the fucking wood for the trees. And you think I'm some kind of MS fanboy! You're way off base pal.
What I am is a pragmatist. I would love it if I could pitch Apple solutions to my Enterprise customers, but as I've tried to tell you over and over, Apple do not tailor their products to sell into the Enterprise. You may be happy sitting there in your little bubble world, writing articles about how great Apple is and how crap MS are, but what you do not have is a finger on the Enterprise pulse and half a clue about the problems they are trying to solve?
Answer me this riddle then oh great and wise Oracle.
Enterprise businesses are turning to Virtual Machines in a big way. Everyone's doing it. MS is working on a VM server; VMware is one of the fastest growing software companies in the world right now; Novel SuSE ship and support Xen; Redhat will do the same in RHEL 5; HP have a VM for HP-UX; Sun have Solaris Domains, and there are other solutions out there like Virtuosso (aka OpenVZ) and KVM is in development and included in the Linux kernel source tree.
Most of them can run multiple OSes from different vendors (apart from Solaris Domains and OpenVZ) but they can all run multiple copies of their own native OS. Why? Because it's what Enterprise customers want.
Oh, hang on a minute. There's one exception to that rule. I wonder who that would be? Why (shock horror) it's Apple.
Apple stands alone on this. They have one VM implementation in production today (Parallels) and one in beta (VMware) yet they are to my knowledge the only VM capable OS vendor that don't allow their own OS to run inside VMs on their own platform.
Tell us then smart ass. Who in their right mind will buy an Xserve to run a single instance of OS X that's tied to that one box, when they can spend similar money to buy a similar box from HP, Dell or IBM and run a dozen specifically purposed VMs with Linux and Windows in them. VMs that can be shunted elsewhere on the fly to balance compute resources or free up hardware for maintenance courtesy of VMware's VMotion or Xen's Live Migration?
The fact that HP is stretching Windows instances beyond absurdity to create a solution that orbits around enriching Microsoft rather than solving a solution doesn't help your case.
Bullshit. It's got nothing to do with enriching MS, and it very much solves a problem (not a solution, duh!). I like thin client solutions. I've worked with them in the past (X windows based) and they are a much smarter way of running a corporate desktop than the fat client for every
I don't like it on Windows, Linux and I certainly don't like it on OS X
So we're agreed that smoothing shouldn't be on small fonts because it makes them all blurry. I don't think anyone will argue about that. But if you don't have it on larger fonts then they look all blocky and untidy round the edges. Well you're entitled to like it. Personally I like my text nice and legible and no one does it better than Apple.
And yet the holy white bar at the top of the scren still stays the same among certain other UIs in applications
Depends on the size of the font on the menu bar. The smoothing preference only goes up to 12 points (though I bet there's a command line way of pushing it higher) and the menu bar font is usually bigger than that.
Says it right in the article you linked:
Did you actually take a look before posting this? "emlx" is only the file extension part of the filename. They're just plain old ascii text inside with a tiny bit of XML appended to the end. You can vi them to your hearts content. I had to do that once a few years ago when someone sent me an email containing some weird HTML in a MIME attachment that crashed mail.app. Its only purpose was to try and format the text nicely so I just edited the offending mail, deleted the HTML crap and everything was fine again. It doesn't get much easier or accessible than that !
I wonder what Apple will do when VMware on OS X goes into production later on this year and we're in a situation where misc flavors of Windows and Linux can happily run on an OS X server, but OS X itself can't.
Part of VMware's appeal to the corporate masses is its ROI message. The average server load across the x86 industry is just 5%. Shocking huh. So consolidating multiple under used systems and encapsulating them on a single platform is a compelling proposition, and it's why everyone and his dog is getting into the VM business today. Throw in VMotion (Live Migration in Xen terms) where you can move your VMs around in a hardware farm according to load and demand and you have something businesses are very keen to adopt.
Apple however stands alone on this. They have one VM implementation in production today (Parallels) and one in beta (VMware) yet they are to my knowledge the only VM capable OS vendor that don't allow their own OS to run on inside VMs on their own platform.
As time goes on and the x86 architecture continues to get more powerful this is going to work against them more and more. Who in their right mind will buy an Xserve to run a single instance of OS X that's tied to that one box, when they can spend similar money to buy a similar box from HP, Dell or IBM and run a dozen specifically purposed VMs with Linux and Windows in them. VMs that can be shunted elsewhere on the fly to balance compute resources or free up hardware for maintenance.
Apple are going to be forced to bite the bullet sooner or later, and when they do it opens the door to running OS X inside VMs on non-Apple hardware. But that won't matter, because the VM presents a known and controlled hardware picture to any guest OS, which does away with the instability OS X would have to face if allowed to run natively on bare non-Apple boxen.
but it just ain't gonna happen in a million years
I think I've just proved that it is going to happen and soon, albeit in the comfy and safe environment of a VM. Either that or Apple can kiss goodbye to their server business.
Personal choice. Many others love it (including me).
The lack of ability to set simple things (turning off anti-aliasing, shadows?)
Rubbish. That's only useful when the font size gets too small. Go into System Preferences, select "Appearance" in the "Personal" row and look at the bottom. You can select the size of font at which you want to turn off text smoothing. The default is 8.
It uses strange unknown formats (ie: mail.app etc.) which provides various ways to lock people into those apps
Pure FUD, mail.app stores emails in plain text. A 2 second google search would reveal just how easy it is to get your mail out of mail.app and for example into Thunderbird
Microsoft invented the "Virtualized Desktop"? Perhaps you forgot about the webserver
I'm not really sure where to start replying to this. Not only have you completely misunderstood what I was talking about, but you don't seem to know anything about virtualisation or thin clients. Thin clients (effectively a display server) + a virtualised OS is not the same as a webserver serving web based applications. They are completely different technologies.
It was invented on NeXTSTEP by a guy named Tim
Yes I know who Tim Berners-Lee is. I was already using the internet for email and ftp access to remote files before the Tim did that work, and remember the birth of the web as we know it today very well. Not that there was much internet access from Corporate networks back then. We had to run TCP/IP over DECnet on a MIPS box running Ultrix to hook up to the wider internet.
Seriously, how can you be so ignorant? You're impressed by old technology and think the world is doing well under the thumb of an incompetent monopoly?
You know you really should read up on things before you dis 'em. You have no clue what the problem is that Athena was created to tackle, and 16 years later HP have done the same with VDI. It's about management overhead. Not access to information/apps.
You seem to think the worlds IT problems can be fixed by putting everything onto web servers providing anywhere access. All well and good, but what that doesn't address is the thousands of desktops in an organisation that have to be installed, patched and tailored to the configuration the user of that PC needs, i.e. fat clients. The big problem with this model is that it requires a huge investment in manpower and time to make it work. Plus in many cases it ties people to just one PC/workstation, or a small group of PCs/workstations with the same config. There's no flexibility and it's an expensive management headache.
Project Athena fixed this by having the workstation do an extremely fast network boot that loaded a root f/s and a skinny OS, with the usr and var filesystems remotely mounted on centrally controlled servers. Data and applications (if I remember right) were provided using AFS (the Andrews Filesystem) and bundled into containers. The whole thing was centrally managed and at its height there were 20,000 campus workstations supported by just 6 members of staff. Any student could use any one of the 20,000 workstations and guarantee their environment would be the same. This should have been a smash hit, but it was the 15K to 5K price difference between RISC workstations.vs. cheaper PCs that killed it.
HP's VDI solves the same problem. The thin desktop clients are cheap and dumb. They connect to generic Windows instances running under VMware and the user environment is layered on top at login time. The thin desktop clients can be deployed en-mass as throw away items, which means you don't need a small army of support staff in your remote offices to manage them. It solves a very real management problem in the Enterprise today, one that's not solved by just converting all your apps to web based apps.
Do Apple have anything like this today? No. Do I wish they did? Yes. But Apple are not targeting the Enterprise (yet) and you should know that in your position. You don't do Apple any favors by pretending otherwise.
My own personal experience of being uprooted at 14 and moved to a school at the other end of the country (north to south UK) resulted in chaos. While the national curriculum was the same, the order of teaching was different. So I ended up learning some subjects twice and completely missing out on others. I received no support and had to learn what I was missing on my own time and on my own initiative.
My head didn't explode but there were times when I felt like it might.
Apple have the potential to take on Microsoft in the consumer space. In many ways they have an advantage here in that Apple customers (currently) don't have to worry about security problems like viruses. That may change in the future but right now it's not an issue. There's very little your average Apple consumer can't do on OS X that they explicitly need a Windows PC for.
Leopard server (when it ships) offers a lot more to the SMB crowd that Tiger currently doesn't, and Apple will be able to leverage some of this new strength to gain further traction into the SMB space.
Where Apple stands no chance at all is in the Enterprise. The majority of Enterprise desktops have too much invested in MS workstations, plus Apple is not producing products targeting the Enterprise that would allow them to mass deploy OS X on the desktop with any advantage over MS Windows. Quite the opposite in fact. I'll give you an example. I was at a VMware presentation/seminar very recently and one products I saw demoed was HP's Virtual Desktop Infrastructure. This allows you to have Active Directory controlled logins, a set of application servers and a suite of VMware servers virtualising the desktop OS with HP thin clients at the front end. The thin client selects a virtual desktop OS to connect to based on load balanced availability, which is then personalised at login time with the (served) apps and data that match the users profile. It's pretty impressive stuff.
I'm not under any kind of NDA so I can quote a specific usage case given (in production today) as Prudential, who in the UK have moved their call center ops to somewhere in India. Only the thin client exists in the Indian call centers, all the virtual desktops, data and applications are in datacentres in the UK. Access to data and applications is centrally controlled on a per account basis and can be updated and (forcibly) refreshed at any time.
The benefits to the Pru are obvious. The security of their data (SAN storage) virtual operating system instances, user accounts and app servers remain in their protected UK datacentres. And the thin remote client architecture means that implementing a remote desktop pretty much any where in the world is cheap, quick and flexible. If in future they want to move their call center ops to somewhere else in india, or eastern europe, or China or even back to the UK, they have the flexibility to do this cheaply, without disrupting their datacentres at all.
Is this possible with Apple desktops? No! Hell, you can't even do it with any of the Linux desktop solutions. The only technology in Unix history that could have matched this solution was Project Athena from MIT, and that was officially retired 16 years ago in 1991 !!
My point is that current *ix desktops (including Apple) are all about glitz and glamor and capturing the hearts and minds of the consumer, and the small footprint of academia. In the mean time, MS and its partners are listening to the Enterprise and building innovative solutions like virtualising desktops for remote, cheap, flexible access.
Lucky you. You obviously have the luxury where you live of being able to choose which school your kid goes to, and have a wealth of choices available so you can move him/her from school to school at a whim.
I'm not sure either that your kid would thank you for flipping his/her learning and social life on it's head so quickly.
On the other hand, Apple is also operating under the presumption that many people will want to - and in fact do - replace their phones when the subsidy contract period is up. Therefore, the number of people who actually do need a battery replacement while the device is in service as a phone (as opposed to keeping it as an iPod) will be small.
I wonder how often the majority of people replace their phones, compared to how often they replace their iPods? If the buying patterns are wildly different then this could be a big customer satisfaction problem for Apple. We all know that Apple are not shy about refreshing their range, sometimes very quickly (e.g. MBP Core Duo to MBP Core 2 Duo). And the mobile industry moves at such a pace that annual subscription renewals are often accompanied with a hardware refresh to keep up with the Joneses. I would speculate that the price of the latter (low) compared to the price of an iPod refresh (high) produces two very different purchase profiles, neither of which are compatible with the other. Yet the iPhone is a combined unit so it has to marry both of them. This is where I think the high price of the iPhone could cause it to come unstuck.
Slashdot Mirror
Why wait for Google. According to the Lightening 0.3.1 Release Notes there's a version you can download with integrated Sun Java System Calendar Server (WCAP) support. And in October you might have another option, given that Apple's Leopard iCal Server supports Sunbird, it should also support Lightening too. Or you could wait for it to be ported to Linux after they've released the source code, which they've promised to do.
I remember when I was at school there was an episode of the BBC show "Tomorrows World" where for an April Fools day prank they reported on the breakthrough invention of de-hydrated water tablets. You expose them to sunlight and they turned back into water. There was one particularly slow boy in class who had an argument with the chemistry teacher the next day, insisting that these things existed cos he'd seen it on TV.
This is really interesting for me. I didn't realise until reading this thread that the ESX hypervisor worked that way. Do you know what its capabilities are for (disk) volume management? For instance one solution I'm deploying at the moment for a customer is an HP-UX Campus Cluster, where the SAN storage and cluster stretches over two data centers on the same campus. The volumes are mirrored across both data centers using LVM, so loss of an entire data center doesn't take down the storage or the cluster. For future solutions I've been considering Xen VMs on RHEL 5 and I could potentially support that environment with similar disaster tolerant stretched clusters using CLVM.
Both of those are viable solutions because the OS on the metal can do RAID volume management. Do you know if ESX can too ?
Also, the 21/7 failed London tube bombers were all identified and found because of CCTV before they could correct their mistake in their bomb design and try again. We got to see their faces all over national TV the very next day, and because of that the police were alerted to their location by a member of the public. There have been so many cases in the UK in recent years where CCTV footage has been pivotal in identifying the perpetrator, leading to their arrest.
Sure, this is all "after the fact", but which is worse, not catching the crim because you haven't a clue who they are, or catching them because you have their face on CCTV? As a victim you're going to feel a lot better knowing the ass wipe who turned you over the first time isn't lurking around the next corner to do it again. Bottom line, the UK public like CCTV and its here to stay. I wish we had it on my street.
Actually they've already addressed the safety question on their FAQ page where they say:
Actually according to this page they're talking about a tank of 300 litres at 300 bar resulting in 46 MJ and also a 340 litre tank with 52.1 MJ. So you're a bit short on power. In addition they're measuring distance in km not miles.
Where? Prove it.
I see no mention of an ecotaxi running on compressed air anywhere in Mexico on the first 2 pages of a google search.
No way in a million years will Dell or any other PC vendor offer Fedora pre-installed on a "noob-friendly" PC. Fedora is a test bed for new and interesting technologies that one day hope to make it into RH's production ready releases. No company in their right mind will risk that on their customers. Ubuntu is probably the only distro that's potentially noob-friendly enough for the job today.
I think this whole conversation is moot anyway. The author of the article hit the nail on the head over the cost of support. It's just too expensive for Dell to do it in-house. The only way they could make it work would be for Dell to outsource the support to another company already geared up to deliver it effectively.
So calling me a jackass and such instead of responding to the technical points I've made isn't ad hominem? I see, thanks for the english lesson (not).
You consistently ignore the technology use examples I give you with respect to their advantages to business because you have no answer. Instead you respond with laughable statements like "VM is still emulation" clearly demonstrating that you don't know the difference between either. Let me spell it out for you. Microsoft VirtualPC and QEMU are emulators because they emulate the x86 chip architecture on non-x86 platforms, as well as the PC hardware environment. Virtual Machines like VMware, Xen and Parallels are not because they run native and don't emulate chip set instructions. Their only purpose is to provide a high performance hardware abstraction layer that multiple operating systems can sit on, that's why they're called Virtual Machines, not Emulators (duh!).
They are so important to the future of this industry that both Intel and AMD have changed the design of their chips to accommodate them, implementing concepts e.g. VT technology (Intel) and in future, Extended Page Tables (Intel) a.k.a Nested Page Tables (AMD) and loads of other improvements are in the pipe line. Even the PCI Special Interest Group is working on changes to the PCI standard so that for example a network card's capacity can be split among different virtual machines.
According to you, all these pillars of industry are "silly" because VM's are "inefficient", "undesirable" and "to be avoided when possible". Maybe you should put your money where your mouth is and write a column article about this and tell them publicly why they're wasting their millions in development costs. Please do
As for continuing here on Slashdot, you're right, you shouldn't bother. You're clearly unqualified to comment on what I'm talking about, as most of it seems to be going straight over your head. Why don't you give your Ego a rest and admit when you don't understand something instead of pretending to look clever by responding with bullshit. You think that access to data is the only thing that matters. LOL, you're an idiot. Do you know anything about running a business and why people pay for computers in the first place? Do you know anything about why businesses buy the systems they do and what technologies drive those decisions? Clearly not. You need to give up writing that column, go get yourself a real job in this industry and get some real world experience.
Well you posted under AC this time, but the style of writing is the same so I'll assume it's you.
I'm just really amazed at how out of touch you are with what enterprise customers want from their systems. "Thin Client" has nothing to do with access to data. I've already said that, repeatedly. It's about taming the beast when you have thousands of desktops to try and manage.
I mentioned Active Directory once, just to give you a clue that centrally controlled user account administration is part of the demo I saw. Replace that with LDAP or Redhat's Directory server if you like, makes no difference to me. Given a choice I'd propose one of those first anyway.
LOL, funniest thing I've read in ages, you need to read more about VMs. You've completely missed what I've tried to tell you about average server utilisation in the industry and how customers are using VMs to consolidate their compute power to improve their ROI (Return on investment). That's what's its all about and what's driving it, plus the flexibility they get when they are able to live migrate VMs from one lump of hardware in a cluster to another. It has a massive impact on:
For those who are interested here are the release notes. The technology preview is particularly mouth watering. Personally I'm especially looking forward to GFS2.
Ok, gloves off. I've been very civil to you through out this dialogue, but now you're getting unnecessarily fucking rude. A few things you should know about this "jackass" before you go any further.
1) I started my career using VMS for about 4 years before switching to Unix and I've been a solid Unix man for the past 16 years. Today I specialise in designing, implementing, integrating, tuning and troubleshooting Unix cluster solutions based on Tru64, HP-UX and Linux. I've only ever owned a Windows laptop once, for about 2 years, and its was the smallest little VAIO I could get away with for use with MS Office and because it was before the days when Open Office was really that useful.
2) I switched from that little laptop to my first PowerBook in 2002, primarily because OS X ran on top of a flavour of Unix, and because it allowed me to ditch my work MS laptop and home Linux workstation and combine both functions on one platform. At that point I was self employed, though now I work with 3 other guys and I've used nothing but Mac OS X on my desktop (exclusively) for the past 5 years.
3) I'm a terrible Apple evangelist to all my family and friends and have converted a modest number in the last couple of years. I hope to do better in the future.
You, in your rabid MS bashing frenzy for anyone who doesn't share your OTT hatred of Gates and his spawn, can't see the fucking wood for the trees. And you think I'm some kind of MS fanboy! You're way off base pal.
What I am is a pragmatist. I would love it if I could pitch Apple solutions to my Enterprise customers, but as I've tried to tell you over and over, Apple do not tailor their products to sell into the Enterprise. You may be happy sitting there in your little bubble world, writing articles about how great Apple is and how crap MS are, but what you do not have is a finger on the Enterprise pulse and half a clue about the problems they are trying to solve?
Answer me this riddle then oh great and wise Oracle.
Enterprise businesses are turning to Virtual Machines in a big way. Everyone's doing it. MS is working on a VM server; VMware is one of the fastest growing software companies in the world right now; Novel SuSE ship and support Xen; Redhat will do the same in RHEL 5; HP have a VM for HP-UX; Sun have Solaris Domains, and there are other solutions out there like Virtuosso (aka OpenVZ) and KVM is in development and included in the Linux kernel source tree.
Most of them can run multiple OSes from different vendors (apart from Solaris Domains and OpenVZ) but they can all run multiple copies of their own native OS. Why? Because it's what Enterprise customers want.
Oh, hang on a minute. There's one exception to that rule. I wonder who that would be? Why (shock horror) it's Apple.
Apple stands alone on this. They have one VM implementation in production today (Parallels) and one in beta (VMware) yet they are to my knowledge the only VM capable OS vendor that don't allow their own OS to run inside VMs on their own platform.
Tell us then smart ass. Who in their right mind will buy an Xserve to run a single instance of OS X that's tied to that one box, when they can spend similar money to buy a similar box from HP, Dell or IBM and run a dozen specifically purposed VMs with Linux and Windows in them. VMs that can be shunted elsewhere on the fly to balance compute resources or free up hardware for maintenance courtesy of VMware's VMotion or Xen's Live Migration?
Bullshit. It's got nothing to do with enriching MS, and it very much solves a problem (not a solution, duh!). I like thin client solutions. I've worked with them in the past (X windows based) and they are a much smarter way of running a corporate desktop than the fat client for everyone abortion that MS push
1) I started my career using VMS for about 4 years before switching to Unix and I've been a solid Unix man for the past 16 years. Today I specialise in designing, implementing, integrating, tuning and troubleshooting Unix cluster solutions based on Tru64, HP-UX and Linux. I've only ever owned a Windows laptop once, for about 2 years, and its was the smallest little VAIO I could get away with for use with MS Office and because it was before the days when Open Office was really that useful.
2) I switched from that little laptop to my first PowerBook in 2002, primarily because OS X ran on top of a flavour of Unix, and because it allowed me to ditch my work MS laptop and home Linux workstation and combine both functions on one platform. At that point I was self employed, though now I work with 3 other guys and I've used nothing but Mac OS X on my desktop (exclusively) for the past 5 years.
3) I'm a terrible Apple evangelist to all my family and friends and have converted a modest number in the last couple of years. I hope to do better in the future.
You, in your rabid MS bashing frenzy for anyone who doesn't share your OTT hatred of Gates and his spawn, can't see the fucking wood for the trees. And you think I'm some kind of MS fanboy! You're way off base pal.
What I am is a pragmatist. I would love it if I could pitch Apple solutions to my Enterprise customers, but as I've tried to tell you over and over, Apple do not tailor their products to sell into the Enterprise. You may be happy sitting there in your little bubble world, writing articles about how great Apple is and how crap MS are, but what you do not have is a finger on the Enterprise pulse and half a clue about the problems they are trying to solve?
Answer me this riddle then oh great and wise Oracle.
Enterprise businesses are turning to Virtual Machines in a big way. Everyone's doing it. MS is working on a VM server; VMware is one of the fastest growing software companies in the world right now; Novel SuSE ship and support Xen; Redhat will do the same in RHEL 5; HP have a VM for HP-UX; Sun have Solaris Domains, and there are other solutions out there like Virtuosso (aka OpenVZ) and KVM is in development and included in the Linux kernel source tree.
Most of them can run multiple OSes from different vendors (apart from Solaris Domains and OpenVZ) but they can all run multiple copies of their own native OS. Why? Because it's what Enterprise customers want.
Oh, hang on a minute. There's one exception to that rule. I wonder who that would be? Why (shock horror) it's Apple.
Apple stands alone on this. They have one VM implementation in production today (Parallels) and one in beta (VMware) yet they are to my knowledge the only VM capable OS vendor that don't allow their own OS to run inside VMs on their own platform.
Tell us then smart ass. Who in their right mind will buy an Xserve to run a single instance of OS X that's tied to that one box, when they can spend similar money to buy a similar box from HP, Dell or IBM and run a dozen specifically purposed VMs with Linux and Windows in them. VMs that can be shunted elsewhere on the fly to balance compute resources or free up hardware for maintenance courtesy of VMware's VMotion or Xen's Live Migration?
Bullshit. It's got nothing to do with enriching MS, and it very much solves a problem (not a solution, duh!). I like thin client solutions. I've worked with them in the past (X windows based) and they are a much smarter way of running a corporate desktop than the fat client for every
I wonder what Apple will do when VMware on OS X goes into production later on this year and we're in a situation where misc flavors of Windows and Linux can happily run on an OS X server, but OS X itself can't.
Part of VMware's appeal to the corporate masses is its ROI message. The average server load across the x86 industry is just 5%. Shocking huh. So consolidating multiple under used systems and encapsulating them on a single platform is a compelling proposition, and it's why everyone and his dog is getting into the VM business today. Throw in VMotion (Live Migration in Xen terms) where you can move your VMs around in a hardware farm according to load and demand and you have something businesses are very keen to adopt.
Apple however stands alone on this. They have one VM implementation in production today (Parallels) and one in beta (VMware) yet they are to my knowledge the only VM capable OS vendor that don't allow their own OS to run on inside VMs on their own platform.
As time goes on and the x86 architecture continues to get more powerful this is going to work against them more and more. Who in their right mind will buy an Xserve to run a single instance of OS X that's tied to that one box, when they can spend similar money to buy a similar box from HP, Dell or IBM and run a dozen specifically purposed VMs with Linux and Windows in them. VMs that can be shunted elsewhere on the fly to balance compute resources or free up hardware for maintenance.
Apple are going to be forced to bite the bullet sooner or later, and when they do it opens the door to running OS X inside VMs on non-Apple hardware. But that won't matter, because the VM presents a known and controlled hardware picture to any guest OS, which does away with the instability OS X would have to face if allowed to run natively on bare non-Apple boxen.I think I've just proved that it is going to happen and soon, albeit in the comfy and safe environment of a VM. Either that or Apple can kiss goodbye to their server business.
You seem to think the worlds IT problems can be fixed by putting everything onto web servers providing anywhere access. All well and good, but what that doesn't address is the thousands of desktops in an organisation that have to be installed, patched and tailored to the configuration the user of that PC needs, i.e. fat clients. The big problem with this model is that it requires a huge investment in manpower and time to make it work. Plus in many cases it ties people to just one PC/workstation, or a small group of PCs/workstations with the same config. There's no flexibility and it's an expensive management headache.
Project Athena fixed this by having the workstation do an extremely fast network boot that loaded a root f/s and a skinny OS, with the usr and var filesystems remotely mounted on centrally controlled servers. Data and applications (if I remember right) were provided using AFS (the Andrews Filesystem) and bundled into containers. The whole thing was centrally managed and at its height there were 20,000 campus workstations supported by just 6 members of staff. Any student could use any one of the 20,000 workstations and guarantee their environment would be the same. This should have been a smash hit, but it was the 15K to 5K price difference between RISC workstations
HP's VDI solves the same problem. The thin desktop clients are cheap and dumb. They connect to generic Windows instances running under VMware and the user environment is layered on top at login time. The thin desktop clients can be deployed en-mass as throw away items, which means you don't need a small army of support staff in your remote offices to manage them. It solves a very real management problem in the Enterprise today, one that's not solved by just converting all your apps to web based apps.
Do Apple have anything like this today? No. Do I wish they did? Yes. But Apple are not targeting the Enterprise (yet) and you should know that in your position. You don't do Apple any favors by pretending otherwise.
People write things, change their mind, edit and then make mistakes. This is slashdot and it happens all the time time. Get over it!
My own personal experience of being uprooted at 14 and moved to a school at the other end of the country (north to south UK) resulted in chaos. While the national curriculum was the same, the order of teaching was different. So I ended up learning some subjects twice and completely missing out on others. I received no support and had to learn what I was missing on my own time and on my own initiative.
My head didn't explode but there were times when I felt like it might.
Apple have the potential to take on Microsoft in the consumer space. In many ways they have an advantage here in that Apple customers (currently) don't have to worry about security problems like viruses. That may change in the future but right now it's not an issue. There's very little your average Apple consumer can't do on OS X that they explicitly need a Windows PC for.
Leopard server (when it ships) offers a lot more to the SMB crowd that Tiger currently doesn't, and Apple will be able to leverage some of this new strength to gain further traction into the SMB space.
Where Apple stands no chance at all is in the Enterprise. The majority of Enterprise desktops have too much invested in MS workstations, plus Apple is not producing products targeting the Enterprise that would allow them to mass deploy OS X on the desktop with any advantage over MS Windows. Quite the opposite in fact. I'll give you an example. I was at a VMware presentation/seminar very recently and one products I saw demoed was HP's Virtual Desktop Infrastructure. This allows you to have Active Directory controlled logins, a set of application servers and a suite of VMware servers virtualising the desktop OS with HP thin clients at the front end. The thin client selects a virtual desktop OS to connect to based on load balanced availability, which is then personalised at login time with the (served) apps and data that match the users profile. It's pretty impressive stuff.
I'm not under any kind of NDA so I can quote a specific usage case given (in production today) as Prudential, who in the UK have moved their call center ops to somewhere in India. Only the thin client exists in the Indian call centers, all the virtual desktops, data and applications are in datacentres in the UK. Access to data and applications is centrally controlled on a per account basis and can be updated and (forcibly) refreshed at any time.
The benefits to the Pru are obvious. The security of their data (SAN storage) virtual operating system instances, user accounts and app servers remain in their protected UK datacentres. And the thin remote client architecture means that implementing a remote desktop pretty much any where in the world is cheap, quick and flexible. If in future they want to move their call center ops to somewhere else in india, or eastern europe, or China or even back to the UK, they have the flexibility to do this cheaply, without disrupting their datacentres at all.
Is this possible with Apple desktops? No! Hell, you can't even do it with any of the Linux desktop solutions. The only technology in Unix history that could have matched this solution was Project Athena from MIT, and that was officially retired 16 years ago in 1991 !!
My point is that current *ix desktops (including Apple) are all about glitz and glamor and capturing the hearts and minds of the consumer, and the small footprint of academia. In the mean time, MS and its partners are listening to the Enterprise and building innovative solutions like virtualising desktops for remote, cheap, flexible access.
Lucky you. You obviously have the luxury where you live of being able to choose which school your kid goes to, and have a wealth of choices available so you can move him/her from school to school at a whim.
I'm not sure either that your kid would thank you for flipping his/her learning and social life on it's head so quickly.