In practice, I have found that you are more or less forced to go back to the CLI to do anything beyond what Apple's Serveradmin GUI offers you. It's also at that point that you discover (in 10.5 at least. I see that Apple has changed the mail server yet again for 10.6, third time now!) that most of Apple's GUI and tools are wrappers around various Open Source technologies (Cyrus, Postfix, OpenLDAP etc), and about the same time where you start asking yourself exactly why you left Linux in the first place.
It's quite simple. Apple has no chance in server segment, simply because it can't compete with the low priced hardware and free Linux based servers that are frankly better and all you really need to serve your solution.
It's fine to make a market a desktop computer for technophobes (look how easy to use it is, hiding that it's really UNIX under the covers), but when it comes to servers that doesn't matter. Servers are maintained by people who know what they are doing and are not afraid of the shell and command line.
The only time someone will buy OS X server is if they want to manage an office full of macs.
Another point I would like to make is the danger of app store for Mac OS X. The temptation for Apple to start controlling what can or can not be installed (look at iPhone) on my general purpose computer is just too great. And that would be the death of OS X eco system and not a revival.
Right now AT&T has a huge influence over Apple. AT&T doesn?t like people using VOIP apps on iPhone. No problem, just tell Apple to disable such apps and not make them available in the store.
You have a cheap alternative to Photoshop. Well, Adobe doesn?t like it, and since Apple would be making a cut of the profits on all sales (after all it makes sense since they provide distribution service) it is in their interest to listen to Adobe when they say you should not be allowed to distribute your app through the app store.
Another problem with this is barrier to entry. Right now if I want to make a ?script? and charge $1 for it I can. But if I have to sign for developer program with Apple and pay a yearly $150 fee, I would think twice about signing up just to sell my $1 script.
This is just so wrong and such a bad idea and I?m speaking here as a user and software developer.
It really would be the beginning of the end of general computing on Apple hardware.
Apple should give up on servers ASAP. They are good on the desktop, and certainly on the iPhone, but servers are a dead end. Why would anyone run Mac OS on a server, when they can run Linux for free? They are flogging a dead horse here.
Apple's problem in the server space is credibility. The problem is that they have changed their server strategy so many times, nobody is willing to risk going Apple for fear that the hardware/software they rely on will be discontinued.
Unlike the ever changing consumer gadget/desktop market, corporations don't like changes to the engine that drives the entire business. It's just too risky.
Apple was starting to rebuild some credibility with the XServe, but then they foolishly discontinued the XServe RAID overnight and instantly lost all the confidence they had started to rebuild in their server products in that single bone-headed move. Who cares if it was or wasn't wildly profitable? It was confidence building, and in the server space that long term cred is worth more than gold.
If Apple wants to be taken seriously in the server market, they need to reintroduce the RAID array, and create a 2U server product to go along with the 1U XServe. XSan needs to be tuned to support other platforms without relying on a 3rd party file system format. In addition to all this, they need to release a *gasp* ROAD MAP, and do some basic marketing in the server space, and stick to the plan. Never, ever discontinue any of these three core server products! Improve them, sure, but never ever drop/restructure the server strategy ever again!
Entry level servers based on (1) mini/iMac hardware wouldn't be a bad idea either.
Quote:
Originally Posted by MacTel
To summarize what others have stated, if Apple did the following then their server marketshare would grow through the roof.
1) Give a roadmap. example: every six months we'll roll-out new server hardware
2) Mirror Dell, HP, and IBM is support contracts with on-site service and 4 hour turnaround options.
3) Fix the bugs and rework the networking and file system stacks to make OSX Server a competitive server OS.
4) Offer more dense hardware configurations. example: a 4U 4 socket system and/or blades.
5) Provide support for multiple direct access storage vendors.
6) XSan and XGrid are nice but Apple needs to make XVirtual (a.k.a. robust Apple in-house made server virtualization software) and change the OSX Server licensing to allow virtualization.
7) Build and fully support a VAR network with the iPhone ecosystem as an example.
8) Rebuild XServe to match HPs features and quality for 1U servers.
9) License OSX Server to server OEMs
10) Create a world class server support group
11) Offer a competing cloud computing service
I think that's about it. If Apple is serious then they'll go after Dell's SOHO business.
I think Apple is serious, they are just plain incapable of understanding the enterprise user. They think what what works with consumers and content creation types will work in the office.
Comments
It's fine to make a market a desktop computer for technophobes (look how easy to use it is, hiding that it's really UNIX under the covers), but when it comes to servers that doesn't matter. Servers are maintained by people who know what they are doing and are not afraid of the shell and command line.
The only time someone will buy OS X server is if they want to manage an office full of macs.
Another point I would like to make is the danger of app store for Mac OS X. The temptation for Apple to start controlling what can or can not be installed (look at iPhone) on my general purpose computer is just too great. And that would be the death of OS X eco system and not a revival.
Right now AT&T has a huge influence over Apple. AT&T doesn?t like people using VOIP apps on iPhone. No problem, just tell Apple to disable such apps and not make them available in the store.
You have a cheap alternative to Photoshop. Well, Adobe doesn?t like it, and since Apple would be making a cut of the profits on all sales (after all it makes sense since they provide distribution service) it is in their interest to listen to Adobe when they say you should not be allowed to distribute your app through the app store.
Another problem with this is barrier to entry. Right now if I want to make a ?script? and charge $1 for it I can. But if I have to sign for developer program with Apple and pay a yearly $150 fee, I would think twice about signing up just to sell my $1 script.
This is just so wrong and such a bad idea and I?m speaking here as a user and software developer.
It really would be the beginning of the end of general computing on Apple hardware.
Apple's problem in the server space is credibility. The problem is that they have changed their server strategy so many times, nobody is willing to risk going Apple for fear that the hardware/software they rely on will be discontinued.
Unlike the ever changing consumer gadget/desktop market, corporations don't like changes to the engine that drives the entire business. It's just too risky.
Apple was starting to rebuild some credibility with the XServe, but then they foolishly discontinued the XServe RAID overnight and instantly lost all the confidence they had started to rebuild in their server products in that single bone-headed move. Who cares if it was or wasn't wildly profitable? It was confidence building, and in the server space that long term cred is worth more than gold.
If Apple wants to be taken seriously in the server market, they need to reintroduce the RAID array, and create a 2U server product to go along with the 1U XServe. XSan needs to be tuned to support other platforms without relying on a 3rd party file system format. In addition to all this, they need to release a *gasp* ROAD MAP, and do some basic marketing in the server space, and stick to the plan. Never, ever discontinue any of these three core server products! Improve them, sure, but never ever drop/restructure the server strategy ever again!
Entry level servers based on (1) mini/iMac hardware wouldn't be a bad idea either.
To summarize what others have stated, if Apple did the following then their server marketshare would grow through the roof.
1) Give a roadmap. example: every six months we'll roll-out new server hardware
2) Mirror Dell, HP, and IBM is support contracts with on-site service and 4 hour turnaround options.
3) Fix the bugs and rework the networking and file system stacks to make OSX Server a competitive server OS.
4) Offer more dense hardware configurations. example: a 4U 4 socket system and/or blades.
5) Provide support for multiple direct access storage vendors.
6) XSan and XGrid are nice but Apple needs to make XVirtual (a.k.a. robust Apple in-house made server virtualization software) and change the OSX Server licensing to allow virtualization.
7) Build and fully support a VAR network with the iPhone ecosystem as an example.
8) Rebuild XServe to match HPs features and quality for 1U servers.
9) License OSX Server to server OEMs
10) Create a world class server support group
11) Offer a competing cloud computing service
I think that's about it. If Apple is serious then they'll go after Dell's SOHO business.
I think Apple is serious, they are just plain incapable of understanding the enterprise user. They think what what works with consumers and content creation types will work in the office.
Maybe it's just your incompetence? Because my experience is 180 degress different.
Mine too, we even used netboot from the xServes to the computer labs and literally removed the hard disks from every computer in the lab.