Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
No, its axing an Enterprise solution without giving clients any warning or any roadmap.

It's being axed specifically because said "clients" are nearly non-existent. Apple never made a pretense for a single second of being, say Sun.

Jobs explicitly said at the XServe launch that its focus was for administering OS X clients in places like .edus and graphics houses. They didn't actively disabuse anyone of the notion that it could be a general-purpose server--they were willing to take anyone's money, of course. But anyone who thought that was a good idea was clearly not qualified to be making such decisions. I'm a little amazed out the outcry, because I haven't encountered XServes in the wild other than as rough parallels to the AD client-management infrastructure at a site. That is, exactly what their stated purpose was.
 
And you know the scope of my CRM from ... what exactly ? Let me tell you, there are laws in place that dictate the uptime my CRM needs to have unfortunately, it has to do with the very specific field my company operates in.

Anything less than fully redundant would be suicide for us.

You misunderstand my point. I'm saying that there are many forms of redundancy, and not one single route that is always right.

Redundancy in itself is not a luxury at many sites. It's a necessity, clearly, you're right. But your described way is only one approach to that end, so you shouldn't be so quick to talk about it like it's a religion.

I oversee around 200+ servers that have numerous times had failed power supplies - multiple vendors, multinational data centers - Sun, IBM, HP, data centers in NA, EU, AP - if it wasn't for redundancy that would have meant huge losses. (For instance if a server with a single power supply was to fail in the middle of a distributed DB transaction leaving locks on the DB - end of story, hundreds of thousands lost.)

Yes, but I'm going to call that a strawman. Nobody is saying that for that particular use case a redundant power supply was unnecessary. The frame of the debate was:

"Hot-swappable drives and redundant power are always an absolute necessity if you care about your business."

and this is simply too broad a statement to be true. It is true of SOME PARTICULAR ELEMENTS of a business, but not as a generality. Clear?

And don't get me started on Sun and power supplies. Or fans. Ugh.
 
Last edited by a moderator:
I'm saying that he should not respond to customer emails. CEO's have "Yes Men" to do those things. Yeah, I'm sure other CEO's respond to customer emails but I'm referring to large corporations, not mom and pop entities.

I get frustrated at this forum on how some people here get annoyed at Job's responses. Do you have any idea how many emails he probably gets from customers? I certainly don't, however I'm sure it's a plenty. I'm sure he has other things to do than sit back and respond to the most likely hundreds of emails he gets a week from customers that are most likely nonsense and rants. Do people actually expect him to write beautiful long winded personal emails with love attached to them?

Where did I say it was "worthy of admiration" that he responds?


Sorry, I confused you with another poster.

I agree, no response from Jobs would be preferable.
 
You gotta love his honesty. Some CEOs couch the reasons in non-speak, here Jobs calls it like it is.
 
While I understand them not wanting to sell something that doesn't sell very well, but I don't think Apple really did much to sell it. I feel that the xServe didn't provide everything that an enterprise might need. Some enterprises/data centers probably need blade servers, or servers with more the 2 CPU sockets, more RAM, etc. The current xServe might be good for medium sized businesses or schools, but not big, nation, let alone world, wide enterprises. Heck, even Apple didn't use them in its data centers. Maybe Apple should take a cue from that and make xServes more like the servers they need.

Also, I feel that how much profit you can get off of a product and how much you sell aren't the only important factors in why to sell something; it's how important it is. I'm sure not very many stop lights are sold, but they still are important, hence why they're being sold. It's just sad how obsessed the US has become with profit and greed, it totally ignores the basic infrastructure. I've heard about how other countries have 100 Mbps internet access and whatever, yet we, the most powerful country in the world, doesn't. Sure, there may not be much profit in upgrading our infrastructure, but believe me, having a great infrastructure makes a whole lot of difference!
 
While I understand them not wanting to sell something that doesn't sell very well, but I don't think Apple really did much to sell it. I feel that the xServe didn't provide everything that an enterprise might need. Some enterprises/data centers probably need blade servers, or servers with more the 2 CPU sockets, more RAM, etc. The current xServe might be good for medium sized businesses or schools, but not big, nation, let alone world, wide enterprises. Heck, even Apple didn't use them in its data centers. Maybe Apple should take a cue from that and make xServes more like the servers they need.

Gee, it's almost as if qualified professionals should have evaluated what it does and doesn't do, and only used it when it really made sense. Like, say, in the places that Apple said were the reason for bringing it into existence.

Shockingly, like most entities in this mercilessly ontological world of ours, it isn't what it isn't and wasn't what it wasn't and isn't what it wasn't and won't be anything at all.
 
What makes you think that you need Xserves to support a network of Macs? You do not. You could use just about any server for that purpose. I bet that Apple has way more data on this subject-matter than anyone on this discussion has. And I bet that data shows that

Supporting a large deployment of Macs using Open Directory means you *need* an Apple Mac Server. Managing MCX without OSX Server (and therefore Apple Hardware) is incredibly complicated and would likely require a full-time xml guy just to sit there maintaining users et al..

So that means you can only use Apple Hardware... Apple Servers.

You suggest 3 x Mac Mini is as good as 1 x Xserve. But for most services, Apple state a Mac Mini can only handle around 10% of the clients that a XServe or Mac Pro can handle (read the deprication PDF for more info). You can't just split a service like iCal server or Mail across three discrete servers, wave a magic wand and call it 'redundancy'. First off it wouldn't work and even if it did, a single failure would kill your service.

So that leaves the Mac Pro. Apple are selling it as 'hey, its cheaper and more powerful!'

But that's because it isn't a server, so costs less to manufacturer, and it's not just the LOM and PSU that people keep talking about: -

- No redundant PSU.
- No LOM.
- Inefficient Form Factor is unsuitable for server rooms.
- Less Fan redundancy and air flow.
- Less monitoring capabilities.
- Uses economy hard drives (Xserve is supplied with WD RE3/4 drives with double MTBF figures).

I sympathise that Apple don't sell many Xserves, and that they make a loss on each and everyone I wouldn't be surprised. But it's a little like HP selling printers; They sell them below cost and make their money on the toner. In this case, Apple make money on the Macs which are supported in the Enterprise and .Edu environment by their Xserves.

Guess we need to go buy an iServer-Room for our mac-pro servers.

:-( unhappy-mac
 
It's just sad how obsessed the US has become with profit and greed, it totally ignores the basic infrastructure. I've heard about how other countries have 100 Mbps internet access and whatever, yet we, the most powerful country in the world, doesn't. Sure, there may not be much profit in upgrading our infrastructure, but believe me, having a great infrastructure makes a whole lot of difference!

Infrastructure spending isn't sexy... Everyone just expects it, but nobody wants to really Pay for it... And there's a LOT of it needed to cover the USA.. Most of the countries you find that have those 100Mbit links are fairly small, and most of the population is clustered in just a few cities... The USA is 1) very large, 2) very distributed population (while many cities, LOTS of suburbs and rural population as well), and 3) the infrastructure is mostly privately owned by numerous companies, not a single government controlled entity that can just decide to do something and have it happen.... Here we have to give incentives for companies to build out infrastructure... It has to be in their best interest to do it....
 
Was anyone at Apple hardly selling them?

I can recall the inundation of ads on the tele for iPod, MacBookAir, iPod touch, iPhone, etc., but never for their higher end MacPro, MBP, or X-Serve. Could that little fact have added to the cause that nobody is buying them Steve?
Have you ever seen an add on TV for any blade server? Or any dedicated RAID hardware?

Your post is a great example that combining two correct statements on their own, no ads for Xserves on TV and them not selling well, do not make a cogent argument.
 
Ironically, the news could prompt people to grab their new Xserve before the 1st Jan deadline and send the sales through the roof... making Apple reconsider the decision and finally update it.:D
 
Gee, it's almost as if qualified professionals should have evaluated what it does and doesn't do, and only used it when it really made sense. Like, say, in the places that Apple said were the reason for bringing it into existence.

Shockingly, like most entities in this mercilessly ontological world of ours, it isn't what it isn't and wasn't what it wasn't and isn't what it wasn't and won't be anything at all.

most people don't like to have more than one vendor in their data center unless they absolutely have to. with servers you avoid this like the plague. reason is that dell/ibm/hp all sell branded accessories that just work with their servers and everyone buys them. they also offer a free server app to monitor their brand of servers

introducing different server vendors brings in management headaches. and xserves don't do anything that other server vendors already do much better and cheaper
 
Have you ever seen an add on TV for any blade server? Or any dedicated RAID hardware?

Your post is a great example that combining two correct statements on their own, no ads for Xserves on TV and them not selling well, do not make a cogent argument.

IBM used to have some really cool server TV ads
 
On the contrary, breaking into the enterprise market with the iPhone and then dropping the primary finisher into the enterprise market - the server - is a senseless move.

What on earth are you talking about?

There is nothing on an iPhone that relies on, or particularly has anything much to do with, an XServe.

You're just parroting some received wisdom about "what people do" in "the enterprise". Apparently you think this is a conversation about, I dunno, RIM or something?
 
most people don't like to have more than one vendor in their data center unless they absolutely have to. with servers you avoid this like the plague. reason is that dell/ibm/hp all sell branded accessories that just work with their servers and everyone buys them. they also offer a free server app to monitor their brand of servers

introducing different server vendors brings in management headaches. and xserves don't do anything that other server vendors already do much better and cheaper

I've never seen a data center that was a monoculture. I'm sure they exist, but it seems like a really bad idea for everyone but the executive who's getting the kickbacks. Ask Sun customers how they feel about it right now, especially since Oracle has now dumped all of the vendors and moved to direct sales. It's a living nightmare that I wouldn't wish on anyone.

Edit to clarify: subsections of a datacenter are usually monocultural for exactly the reasons you state, but I haven't seen one of any size that shoehorns a single vendor into handling every aspect of the business' software needs. Some tools are for some things, others... are for others. It depends a lot on how commodified your software needs are, obviously, and the management culture. I've seen places invest millions in switching mail over to expensive Exchange installations because they were revving up AD for client administration. This is clearly silly from a business as well as a tech perspective, as any theoretical "savings" from being all-MS instead of some-MS-some-*nix will not begin to compare to either the transition or recurring costs, but it's one of those things that people do because someone told them in business school that it's "enterprise".
 
Last edited:
Wirelessly posted (Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_1 like Mac OS X; en-us) AppleWebKit/532.9 (KHTML, like Gecko) Version/4.0.5 Mobile/8B117 Safari/6531.22.7)

hahaha "the end of Xserve is a knight in our back" classic :)
 
What on earth are you talking about?

There is nothing on an iPhone that relies on, or particularly has anything much to do with, an XServe.

You're just parroting some received wisdom about "what people do" in "the enterprise". Apparently you think this is a conversation about, I dunno, RIM or something?

On the contrary, I think his point made sense.

On the one hand, iPhones (and now iPads) are super popular with consumers, and making inroads into corporate settings. It's a viable contender with Blackberries now. So on the one hand, executives are saying "Hey, we should maybe give Apple a shot. We could deploy iPhones across the organization instead of Blackberries". Perhaps this could have been a halo effect. Let's get a Mac for our secretaries. Hey, these Macs could make good workstations and dev machines. Hey, maybe even servers too.

To which the IT guys will say "Nuh uh. Apple clearly aren't showing any support for us, look at what they just did to the XServe. We're not investing in Apple if they're willing to pull the plug entirely with just a few months notice." Sale: lost.
 
My first thought was why not release what people often clamor for?

The mythical mid-range smaller but upgradable Mac tower.

One of the reasons I've held onto my G5s is that, well #1, they still work fine and are sufficiently fast and expandable and the Mac Pro is way too overpriced to replace a lot of Macs cost-efficiently.

And the Mac Mini is not upgradable, so not an option for some people.

Such a machine could have a server option and fall in between in price between the Mac Mini and the Mac Pro. It wouldn't be a rack mount, but could be sufficiently small enough, say smaller than the average PC tower, such that it could be useful.

It would sell as both a low end consumer Mac tower as well as a mid-priced server and would help sell more Apple displays as well, hence filling a gap in two areas Apple doesn't serve well and increasing sales of another Apple product line on the consumer level. And maybe an upstart 3rd party company could offer a service to rack mount it, albeit at a larger size, but still technically possible.

I know what Steve Jobs argument will be... cannibalization of other product lines, but there is clearly a gap now and an iMac obviously does not serve some peoples' needs, the Mac Mini server is underpowered/un-expandable, & the Mac Pro is costly.

HELLO APPLE! ARE YOU LISTENING?
 
Other than a small collection of repetitiously vocal loudmouths, no one is clamoring for this.

And such a machine would be an even worse server than their current (meager) server offerings.


My first thought was why not release what people often clamor for?

The mythical mid-range smaller but upgradable Mac tower.

One of the reasons I've held onto my G5s is that, well #1, they still work fine and are sufficiently fast and expandable and the Mac Pro is way too overpriced to replace a lot of Macs cost-efficiently.

And the Mac Mini is not upgradable, so not an option for some people.

Such a machine could have a server option and fall in between in price between the Mac Mini and the Mac Pro. It wouldn't be a rack mount, but could be sufficiently small enough, say smaller than the average PC tower, such that it could be useful.

It would sell as both a low end consumer Mac tower as well as a mid-priced server and would help sell more Apple displays as well, hence filling a gap in two areas Apple doesn't serve well and increasing sales of another Apple product line on the consumer level. And maybe an upstart 3rd party company could offer a service to rack mount it, albeit at a larger size, but still technically possible.

I know what Steve Jobs argument will be... cannibalization of other product lines, but there is clearly a gap now and an iMac obviously does not serve some peoples' needs, the Mac Mini server is underpowered/un-expandable, & the Mac Pro is costly.

HELLO APPLE! ARE YOU LISTENING?
 
My first thought was why not release what people often clamor for?

The mythical mid-range smaller but upgradable Mac tower.

So your definition of "people" is computer geek, interesting.

My guess is Apple's definition is the mass consumer market, you know, where they are thriving.
 
On the contrary, I think his point made sense.

On the one hand, iPhones (and now iPads) are super popular with consumers, and making inroads into corporate settings. It's a viable contender with Blackberries now. So on the one hand, executives are saying "Hey, we should maybe give Apple a shot. We could deploy iPhones across the organization instead of Blackberries". Perhaps this could have been a halo effect. Let's get a Mac for our secretaries. Hey, these Macs could make good workstations and dev machines. Hey, maybe even servers too.

To which the IT guys will say "Nuh uh. Apple clearly aren't showing any support for us, look at what they just did to the XServe. We're not investing in Apple if they're willing to pull the plug entirely with just a few months notice." Sale: lost.

Can you talk about a time when you've seen an analogous scenario play out in real life?

My experience is that the type of organization that has these archetypal "dictator" IT departments has perfectly matching management, for reasons that should be obvious. So in this scenario the exec would either say "too ********** bad, you're doing it anyway"... or, (s)he never would have suggested it in the first place.

I'm not saying you're totally wrong, the "enterprise" encompasses millions of people and situations so it clearly will happen somewhere, at some time. I'm just wondering how often you've actually *seen* it happen, as then we could extrapolate whether it might be common enough for Apple to care--clearly, they don't agree.
 
HELLO APPLE! ARE YOU LISTENING?


Apple never listens, they only broadcast. If you like what you hear thats fine, if you don't, thats also fine.
 
Guys, will you please stop emailing Steve Jobs every second? He needs to be thinking about the future more than the present.

LOL, we are only allowed to climb the mountain to ask the sage once every 10 years about the meaning of life...
 
Ironically, the news could prompt people to grab their new Xserve before the 1st Jan deadline and send the sales through the roof... making Apple reconsider the decision and finally update it.:D

More likely, it's going to cause cancelled orders, and also hurt vendors. I cancelled an order for a $20k Promise RAID because I have no idea how long it would be supported, since it uses Apple firmware.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.