Hype cycle needs to slow down a little

Good opinion piece, and spot on.

I work in one of the IT backwaters you mention. My company is a service provider for a very mature, staid industry sector that has very little use for cloud, big data, etc. What they do care about is reliable, always-up basic information processing. The last few years in our company have been spent keeping the marketing guys and CIO from falling all over themselves and getting caught up in the hype machine, lest we alienate the customers who've come to rely on us.

Big Data is my favorite -- it's great if you're Amazon.com and have petabytes of information about what people buy and what they click on in your vast online retail empire. It's great if you're Facebook and want to sell businesses the eyeballs of 21 year old Asian iPhone owners who "like" Starbucks, skydiving and fly fishing. It's not so great for a small business selling to a small audience. And it doesn't help most vertical markets.

Cloud is another. VMs + network + storage, all more flexible than before, either in your place or someone else's. Plus software to tie it all together. That's it. End of story. When IT tries to explain this, the marketing guys say "That's it?" Yup, that's it.

IT has always been in perpetual hype mode, but it has really gotten out of hand lately. Hopefully this is just another dotcom style bubble and things will calm down.

DS3500

Isn't the DS3500 just an OEM'd model from NetApp/LSI Logic anyway? The GUIs to manage them are slightly different, but the DS3512 and 3524 report their controllers as LSI Logic, and you use LSI's multipathing software on Linux to talk to them...

That said, the storage line does need to be simplified a little bit. Especially in the middle tier, they have a lot of slightly different models that could be combined to save costs and improve focus. Hopefully storage doesn't get sold off like the servers and PCs...

Not happy about that, but I guess it was inevitable.

Lenovo has been selling a "ThinkServer" line for quite a while, which are basically clones of the very low end System x towers and rack systems. I think it was branded as some kind of cross licensing deal, but I guess we know what was really going on.

Hopefully Lenovo will keep the design and engineering teams in North Carolina -- they're still working on most of the business-grade hardware designs for the PC business, and the System x/BladeCenter guys are there too, so it's a good fit.

I guess the only good thing is that even with the recent cheapening of the designs, the ThinkPads and ThinkCentres do retain a good chunk of the original IBM design...they're still my favorite laptop vendor. So hopefully, if Lenovo is smart, they won't mess around with quality too much. IBM's System x gear is (was?) top quality, easily as good as or better than the HP ProLiant. And at the high end of the range, their hardware designs are really interesting. The downside is that it's very expensive compared to Dell, slightly more than HP. For that price difference though, you get US-based tech support who really understands their stuff. I've never had a problem with IBM System x and BladeCenter support (It's a dedicated group sitting in Atlanta if you're in the US,) but have wanted to reach through the phone and strangle HP and Dell "support" representatives.

The fact that IBM is keeping FlexSystem, System z and System p in house seems to me like they're betting that everything will collapse into the cloud, either public or private, and no one will need physical x86 boxes anymore. FlexSystem is basically BladeCenter 2.0 + Cloud-in-a-box, and competes with the VCE stuff. It looks neat, but just like the VCE stuff, I can't afford it to play around with in the lab. However, I wonder what's going to happen with OpenStack gaining popularity now -- people won't be locked into VMWare or IBM and they'll actually want cheap x86 boxes that they can turn into a loose collection of VM containers. Plus, you have places like the one I work in, with highly distributed branches and crappy network connectivity all over the universe that can't do the consolidation thing cost-effectively who will be buying physical boxes for some time to come.

This is kind of a bummer to me though -- I live in New York, and IBM still has a fairly big presence here, since their HQ and main research lab are in Westchester and Dutchess counties respectively. Just like the rest of the US though, that presence keeps dropping with every physical product line they sell off. I don't wish ill on other countries, but part of me is hoping for a Chinese economic collapse just so MBA management consultants will stop telling executives that you can't manufacture physical products, for any price. Lack of physical products or selling that off to the lowest bidder means no long-term hardware innovation, and we're going to end up stuck in a rut forever. (I was a rust belt kid and watched all the manufacturing move to the South, then overseas. Not fun.)

Bye vCenter / vCloud Director...

With Microsoft basically giving away Hyper-V + enough tools to manage basic deployments without System Center, and all the traditional Linux users/vendors (Red Hat, IBM, etc.) moving towards OpenStack, VMWare is going to have to change their model a bit. I guess it was only a matter of time -- I've said in the past that all the OpenStack guys needed was a big vendor to herd the cats and a decent front end to everything that doesn't require a specialized staff just to maintain.

That's exactly what Red Hat did for Linux -- set a sane enterprise release schedule, cut back the thousands of choices open source gives you to a manageable number, put up a sane and nice enough front end for all the OSS stuff under the hood, and provide decent support.

I definitely see OpenStack gaining ground in medium sized installations where Windows isn't necessarily the technology of choice and the pockets aren't deep enough to buy VMWare, or worse, VCE vBlock stuff.

Next bubble -- public clouds?

Wow, another vendor jumping into public IaaS. Amazon, Microsoft, VMWare, Rackspace, IBM buying its way in, now Google... I wonder if we're going to have a public cloud bubble on our hands.

With all these vendors spinning up huge amounts of compute capacity, and turning it into a commodity, there's going to be an interesting race to the bottom on pricing. I can't wait until one vendor starts going below their own costs to "make it up in volume" the same way the dotcom people did in the 90s.

Actually, the way out for these guys is to get out of truly public cloud and get back into the old-fashioned hosting business. After all, they have massive data centers now, all they would have to do is carve them up into private instances the way they did before the cloud became the new marketing hype. I think that if customers could be reasonably assured that their data was separate from others' data, and there was an SLA beyond "oops, sorry about that." they might begin trusting the cloud vendors to host their workloads.

So yeah, just like the dotcom boom was good for expanding the Internet massively, the public cloud boom is going to leave us with lots and lots of cheap compute, storage and network capacity for some time to come.

Great, more volatility

R11's comment about people not being able to see information until the WSJ publishes it kind of highlights a problem we have today -- extreme market volatility that gets worse every year.

HFT has a lot to do with this, but another factor is just the fact that it's so easy for Joe Investor to see something on CNBC (or Facebook/Twitter now) and instantly log in to their brokerage account and trade. I don't want to go back to a world without the Internet, but it sure was a much more predictable stock market back then. Even very wealthy people had to get their information from old-fashioned sources, and in most cases only had a few hours' jump on everyone else. Think about it -- Sir Thurston Picketts III picks up the WSJ, sees an article predicting the imminent collapse of Company X. He has to call his broker, put in a sell order, and hope he can get out before the world burns. So does *everyone* else. Now, that same story pops up on Facebook, and 25 million investors immediately put in the same sell order, all at once, with zero delay, dragging down the price, causing more people to sell, and creating a negative feedback loop.

This is just going to contribute to the ongoing casino-ization of the markets. Too bad everyone's retirement savings is in there (in the US at least...)

Not what I wanted to see.

It's good that Windows 8 is getting some usability fixes, but it's still missing the one key feature I care about.

And it's not the Start menu. I'd like that back but if Microsoft wants to take it away, I'll deal. What I NEED back in the next SP for Windows 8 is the Aero Glass theme and the theme-able desktop in general.

I kind of understand why Microsoft took away the themed desktop (force everyone to think of RT and apps first) -- but I can't stand working in the flat, featureless user interface. Things like Server Manager and Visual Studio are a huge pain to use for me...acres of monochrome text and icons with very little standing out. The other thing I'm not a fan of is the flat window controls; on a busy screen it can be hard to find what's actually the active window (and yes, I know the color changes, but Aero to me was much easier to navigate.)

I'm disappointed, because Win8 and Server 2012 make huge improvements (Client Hyper-V, neat storage stuff in Server 2012) but it's held back by a user interface I don't like using.

Lots of companies doing this

I think this is the latest MBA/accounting fad from the last 20 years or so. Not sure why, but apparently it's bad to have assets like real estate on your books. Why that is is beyond me - don't they get to knock a huge amount off their profit number for the taxes and maintenance of property they own?

But I guess that's why I'm not an MBA or accountant.

It's not just limited to companies either - here in the US, state governments are doing the same thing, selling and leasing back facilities, roads, bridges, etc.

Check the HCL first

One piece of advice I'd give everyone is to check whether your hardware is "officially" supported on VMWare's compatibility list. All of it. Otherwise, your upgrade will take more than "a pot of coffee, a shut door, google and a big pile of PDFs printed out" as david64 so eloquently put it.

I just finished an upgrade from 4.1 to 5.1 on some IBM stuff we have for our lab. Luckily it's the lab, because this particular combination of very expensive gear isn't on the HCL, and 5.1 isn't tested by IBM. And, it's just just a support issue -- there was a whole host of strange, nested problems between drivers, firmware levels, the defaults IBM picked for their customized ESXi build, ESXi kernel settings, etc, etc, etc. Because it's IBM stuff, the docs on everything couldn't be found from their website and I just happened to find the magic tech note and manual from a lucky Google search after a lengthy quest.

I eventually got it working and everything's fine -- but those who think all upgrades are going to be "Next, next, next, finish" should plan to spend more time. For those who do make it, it works great. The web client is improved, and the whole system just feels a little faster.

Beer because I needed one after that particular waste of my life...

Slimebag meter overload

Wow, that guy really sounds like a dumbass.

Until you see these types in action, you think it's all a stereotype. But I've seen my fair share when working for various IT services and software houses, and this is coming from a guy who just has a normal relationship, isn't a rabid feminist, or a Puritan.

I'm not sure what drives these types, but it's definitely linked to the sales culture. There's plenty of socially maladjusted colleagues of mine in the engineering realm, but that's a whole other kettle of fish. Maybe it's upbringing, maybe it's the fact that these salesy types are outgoing and think they've got extra swagger, the fratboy culture that naturally attracts people to sales, or whatever. But I've definitely witnessed behavior that made my "slimebag meter" jump into the red. The big software houses like Oracle, CA, etc. all have huge sales forces full of these types. You would think that they would know to at least keep their remarks out of earshot of the people they were aimed at. It's even worse when you hear it coming from a balding, slightly overweight middle aged software salesman. Again, just a normal guy like myself looks at those types and thinks, "Wow, you're classy, NOT."

I forget what movie this was from, but one character happened to be this 50 year old dude with a bad dye job hitting on random women in the bar, who the others idolized for his pickup skills. In confidence, he told his best friend that no one should want to be the creepy old guy in the bar hitting on women. This guy should take that advice to heart.

Lax controls always lead to temptation

I would guess a major telco like Verizon would have a Molybdenum Tier Service Contract with Cisco that does advance shipment of replacements. We're a small engineering lab for a major corp and can buy contracts from HP and IBM which would let us "own, but not have to buy" a fully stocked spares pool onsite for our servers if we wanted to. But I can't see even the biggest of a vendor's customers having a service agreement that let them keep, or at least not account for, defective parts. The only explanation would be that Verizon buys such a massive volume of equipment from Cisco that the tracking and returning costs outweigh what is spent...but when you're talking things like $10K switch cards, I can't see Cisco not caring whether they are actually replacing failed kit or not. So, I really wonder what the loophole was; it had to be something like this guy being able to sign an affidavit saying that the equipment wasn't serviceable, and Cisco would accept that.

More importantly, I don't know anything about this guy and his state of mind, but this goes to show you that people will look to find the holes in any process and exploit them for their own gain. I've always been an honest person, but I can see how people get tempted. Especially so, if this guy's girlfriend was the type who needed to be placated with rounds of cosmetic surgery on top of all the other stuff! [1]

I've been in a lot of similar situations, where I've found myself in charge of a process with massive potential for abuse. And I admit, you do think to yourself, "Wow, I could take thousands worth of equipment and service using this loophole, and no one would never know." This case does provide people like me with some vindication...that some auditor, some forensic accountant, somewhere, will eventually find these things out and start digging. Nice guys may finish last, but at least I won't be going to jail like this guy. :-) You may think you're smart, but up against someone whose sole job is to make the books balance, you're playing with fire.

And it's too bad that people like this often cause massive walls of paperwork and bureaucracy to be erected around the process for buying equipment.

[1] Why, oh why, does the IT community constantly reinforce the "Married IT people only have mail order brides and trophy wives" stereotype?? Seriously, some of us are normals with normal family lives.

Oracle mainframe?

"We own it. We manage it. We upgrade it. You only pay for what you use,"

Sounds like the IBM high-end systems model to me! Definitely a different track for Oracle to be going down, but not surprising given their software model. I can definitely see large organizations salivating at the thought of getting rid of those pesky, expensive Oracle and Solaris admins. The IBM System p and System z platforms run like this too -- you have some control since the machines are physically there, but the daily maintenance is handled by IBM and they send out part-swappers when physical tasks need to be performed. The machine even calls in the tickets by itself. And when you want more capacity, they just turn on more processors, which they happily provide you because the cost to use them way outweighs the cost of providing the physical hardware to your site.

Only problem with this model? Huge hundreds-of-percent margin for the vendor and massive lock in. Imagine trying to extract something like this private, Oracle-managed cloud from your datacenter. Not that locally run Solaris and Oracle software are any less of a lock-in, but when you own the systems you at least have the option to get rid of it without rebuying things. I can see a couple of customer types for this service -- customers who just want to absolve themselves of any responsibility beyond paying the IT bill, and those without the staffing levels to work through all the crazy Oracle software problems that pop up from day to day.

New baseline unemployment likely

I'm guessing that the author's speculation that a new baseline unemployment level has appeared is the most likely cause of the high number.

My personal experience during this recession has been decent - I'm still getting phone calls about interviewing for jobs, just less frequently. However, I do know a lot of people who aren't so lucky.

The first problem is that people looking for full time employment who have been laid off are having trouble getting interviews; apparently, any gap in your employment record is the kiss of death.

The second is a huge skills mismatch. Within IT, there are a lot of people who fill very specific niches, and employers refuse to look beyond that to see if they're trainable for something else. Outside of IT, there are fewer and fewer jobs for "the rest of the population". In days past, people who weren't cut out for technical or scientific work had their choice of decent-paying factory jobs or various decent-paying mindless paper-shuffling jobs in corporate-land. Automation and outsourcing on both fronts really hurts those in the middle of the IQ scale, and permanently relegates them to low-level service jobs. Even that sector can't absorb all these people, so you get the unemployment levels we're seeing now.

I have no idea how we're going to solve this. Until we do, it will be a very divisive issue and various political rhetoric is going to get less and less reasoned/rational.

Interesting point

"The z12 chip is implemented in a 32 nanometer high-K metal gate process and fabbed by IBM itself at its East Fishkill, New York foundry."

Very interesting that IBM, king of all things outsourcing and offshoring, still makes the chips for its mainframes in-house. They should teach something like this in MBA school -- never give away stuff that makes you money or risk your intellectual property! It might keep those of us in the US and Europe employed for a few more years.

Then again, when you're talking about a 700% ROI, maybe they figure it's worth it to not let a competitor compromise some contract manufacturer. Also, since the mainframe system is highly proprietary and not well known in general IT circles, I'm betting IBM has more than a few hardware nerds on staff that they absolutely must keep, and don't dare let the customers see. :-) I actually work in an industry highly dependent on mainframes, and my experience is that recent grads have no concept of mainframe computing, either because it's a total black box accessed through a web page or (shudder) terminal emulator, or because it's complex.

Oh well, I guess we can't all go back to the glory days of Digital/Sun/IBM/SGI with the order-of-magnitude margins on hardware that let those companies do things like this. Some of the older folks in the circles I travel worked for these places in the 80s/early 90s, and said they never wanted for anything, simply because the companies were so cash-rich. Of course, the flip side of that is those $25000 workstations....

Maybe this is a good sign.

If these 25,000 cuts are actually in "services", they're probably finally getting around to cutting the last of the EDS dead wood. By that, I don't mean the core group of older, smart people who secretly keep things humming behind the scenes. I'm talking about the ones who have managed to hide out in a little corner of the company for years longer than they should have. Almost all "services" companies have trainloads of customer liaisons, analysts, product development specialists, more analysts, multiple layers of operations staff, and sales and support staff that just add overhead. And customers pay for that. EDS has tons of ultra-large enterprise and government outsourcing agreements -- imagine how many staff are covered by those!

Or, we may finally be seeing the shift back from the "send us your problems and we'll get half the population of India to work on them for you" era. I work at an IT provider for a *very* "trailing edge" industry and our management is just starting to see the folly of that, so that phase must definitely be over now...

Or, HP might be getting back to its roots of actually making physical products people want to buy, and selling them for more than it costs to make them. You know, the dirty "M" word.

Nah, who am I kidding? Whitman is going to get rid of all the engineering staff, replace them with marketing drones, negotiate a Lenovo-style deal with some contract manufacturer and collect a huge payday.

No money in hardware?

Why does IBM seem obsessed about getting out of the hardware business? Even the guys at HP figured out that gutting the PC business was a bad idea since they were able to sell into consulting engagements. And those were commodity boxes. This is retail POS hardware, which is hideously expensive and high-margin.

I have no idea what's taught in MBA classes that drills the "physical products = bad" mentality into the latest generation of managers. Someone has to make good quality equipment, and a "smart" consulting company should know that having your own hardware division to directly push boxes to suckers^Uclients is a good thing.

Part of a trend, makes us all look bad

I'd never do anything like this, simply because I'm a professional. I like working in this industry and wouldn't like the idea of basically being blackballed from it for a stupid move like this. But it's an interesting pointer to a trend I've seen -- employees being completely disconnected from their employers' business.

That's not a total shocker given how little loyalty has been shown on the other side of the table. It doesn't excuse stealing proprietary chip designs, but I can see why some individuals might take this approach (on a related note, check out the top 10 IT rogues' gallery for more examples.)

Losing your job is one thing -- stealing a chip design to impress a new employer is a totally different one. If I were AMD, and saw those documents, I would (1) physically get that guy far away from my facility, and (2) scrub my eyeballs and brain of any trace of what I just saw. There would be no way AMD could have used this data without throwing up a million red flags, and it serves as a pretty good gauge of how loyal this guy would be as an employee.

I honestly don't know which way IT should go. One way is for us to become 100% untrusted, hired-gun contractors who only do the job we are authorized to do, and get tossed as soon as we're not needed. To make that a reality, we'd all have to work off-premises or in a secured contractor bubble in the employers' facilities. This model works great for 20-somethings with no family and a one-bedroom apartment -- I've worked with lots of people who travel 300+ days a year, get paid 4x the salary of a regular employee, and actively love it. The only problem is that every one of the older ones I know is unmarried or divorced because there's no way to sustain a family with that kind of life. For those of us who aren't as enthusiastic about our multi-Platinum status on frequent flyer/frequent guest programs, I think a model with more stability could work. Roll back the clock to the 50s through the 70s, but keep current technology in place. Workers could get a job for life, pension, benefits, salary that keeps up with inflation, etc. In return, a company would get workers who were more invested in their employers. I think you'd still have some wrongdoing, but you'd see less incidents of people walking off with company secrets. That's just because people like me would figure, "Hey, this company is paying me well and doesn't treat me like crap...why steal from them?"

Oracle's weird support model strikes again...

I'm sure these guys probably were trying to pull a fast one on Oracle, but resellers could really get burned.

Oracle, Cisco and others have an interesting model. You can go on Oracle's website, and download the latest major release of any of their software, with no licensing. If you were crazy, you could just take their software and use it in production. Or...could you? Our company has legit, paid-for support for Oracle and Solaris and we play the per-core, per-socket licensing game every year with them. And it turns out that you actually need it. Why?

1. No patches without a support contract

2. Oracle doesn't put out point releases on the web for free. Any problems that require you to patch the original media remain unsolved on the "free" side of the website.

3. Oracle also doesn't update the official product manuals, instead relying on "Notes" to tell you the real deal once they find a problem, again, only available on the support site.

Now, adding Solaris and Sun hardware into the mix, there are no more Solaris patches for download, and no more firmware updates for Sun kit without a corresponding contract. So in one shot, Oracle killed the hobbyist SPARC and Solaris ecosystem.

It could be that Oracle is working to keep their consultant base in business, but the simple RAC setup I did in our lab last month took WAY longer than it had to. I ran into spots in the install that I would never be able to overcome without patching the media, reading the 12 or so linked support notes, finding out what in the original install guide was just plain wrong. etc. etc. Compare that with a 45-60 minute install of a SQL Server cluster, a couple hours messing with DB2, etc.

So, given how critical support is to making Oracle products work, I could see a bunch of resellers falling into this trap. It's like Oracle is giving them enough rope to hang themselves, then swooping in when they discover someone has been sharing support contracts.

Not the first, not the last...

I wonder how this is going to affect Groupon's share price...

The funny thing is, stories like this have been very common, albeit not this wildly out of control. Groupon's MO seems to be targeting small businesses whose owners are either desperate for traffic (already a bad sign) or greedy enough to think that all these new customers will empty their wallets once they're in the shop for the item on offer.

I've used Groupon once, and it was for a major national retailer offering a too-good-to-ignore price on a service that's insanely high margin. I didn't feel too bad because (a) the retailer isn't going to notice the difference at scale, and (b) plenty of non-Grouponers pay the full markup, so they're doing fine. Now, doing this to a small business person who is pulling out the last weapon in the armoury before bankruptcy? Not so nice...

It is nice to see that the business owner isn't trying to weasel out of the deal, as I probably would have expected. It takes integrity to admit that you miscalculated and pay for your mistake, as I'm sure this bakery owner isn't happy about wiping out a years' profits.

Yet another example of how the laws of profit and loss aren't suspended once the Internet gets involved.

Hmm...

Not quite sure how to react to this one...an offer of pricing relief from the King of Vendor Lock-In?

That said, customers might want to think about that a bit. It's pretty obvious that Oracle is turning the Sun hardware business into a set of (really expensive) database appliances, and Solaris into a boot loader for the database and middleware.

Everyone I've talked to who has Sun hardware is getting off of it ASAP, because they see where this is going. Almost all their workloads that aren't 100% married to Solaris are being moved to Linux or other OSes. I'm not exactly sure IBM's the right vendor to run to for this. But, it shows you the feelings of the IT industry now about Oracle controlling a huge vertical stack.

Wow.

Is it just me or is this a really strange period for corporate craziness?

- Nokia flushes their entire R&D group to become a WinPhone 7 shop

- Sony has a month-long shutdown of PlayStation Network because it failed to secure its servers

- HP fires Carly, then fires Hurd after, er, "irregularities," then fires Apotheker after failing to pull a Lenovo and flushing a billion on WebOS/Palm

I would like to see Zombie Bill Hewlett and Zombie Dave Packard show up to a board meeting to see what HP turned into in the last 20 years. Maybe feast on the gathered brains...oh, wait...

If I were on the board, I would definitely not have chosen the CEO that oversaw the purchase of Skype and changes to the eBay rules that drove tons of sellers off the site. I know there's pressure to drive quarterly results and get the stock price, but here's a thought guys -- If you can get beyond 3 quarters, and the company actually MAKES a PHYSICAL product people want to buy, that isn't garbage, and your strategy is improving those products rather than all the branding garbage, the company will make money.

No one realizes that you have to put something in to get profit out every now and then. Only IBM, with gobs of money and a huge services division that doesn't stink as badly as EDS, can pull a Lenovo. And a product is not guaranteed to be an overnight success, especially when you fail to price it right. That's no reason to turn around after a month and say "oh well, that was a waste."

My recommendation? Pick a long-time internal executive who has actually been running a profitable product division for a very long time. They're going to be much more invested in HP's success than these celebrity hired-gun CEOs. People like that will realize that enterprises still buy lots of hardware, even if they're building these cloudy things, and work to make the hardware they buy better than anything else out there. That beats flashy marketing, "brand" and printer ink any day in my mind.

I'm a long-time purchaser of HP gear, business and personal, and all I have to say to Whitman is "Please, please please don't sell HP to Oracle - ignore the Hurd." That would be a very ignominious end to a great company.

In the year 2525....

I have a dumb question/observation. In the West, especially in the US, we have a huge "skills mismatch" where tons of factory workers or service-industry people are out of work because of automation and outsourcing. Now, the plan is to put even more people out of work by automating production in the outsourced regions also?? Who in their right mind thinks this will end well?

We have an incredibly difficult time already finding qualified people to fill jobs where I work. What's going to happen when *all* of the manufacturing jobs are gone? Those jobs (used to) provide a semi-stable work environment and wage progression for people who weren't able to do "knowledge worker" kinds of jobs. If you couldn't even make it through high school in the past, you could at least get a job at the local factory, bust your butt and earn something approaching a living wage. Now, I only see a couple of outcomes:

1. French Revolution 2.0, complete with decapitations

2. The non-knowledge-worker types get pushed into education for knowledge worker jobs, pushing down the wages, skill level and driving other knowledge workers insane. (Already happening now...)

3. Some kind of "subsidy for not working" would have to be introduced to quell dissent.

I may be a Luddite, but I say we should bring back huge steel mills/factories employing 20,000 people at a shot. The work may have sucked, but at least it kept people busy. Otherwise, we may have to hide to avoid the guillotine!

Personally, I like the idea of the old system...it doesn't force people to chase promotions they aren't qualified for in order to keep advancing wage-wise. If I was a fantastic Systems Engineer II and never wanted to be a Systems Engineer III or (shudder) Manager of Systems Engineering, I could stay on a decent wage progression without topping out.

ITIL Is Not a Cure-All!

I've done admin and systems design work in a few "ITIL shops". The reason that's in quotes is because, in my experience, being an ITIL shop involves paying 7 figures to an enterprise software vendor for service desk/change management software. The software is then deployed with zero customization, and every single ITIL process, including the ones that don't make sense for that particular org, are put in place.

This is the kind of environment where requesting a change involves the *entire* ITIL procedure, including filling out generic change management forms in the software with 150+ mandatory fields, the change management meeting, backout plan approvals, scheduled maintenance windows, post-change reports, etc.

Unfortunately, I wish there was a better way to do what ITIL sets out to do. It seems like you either have a situation where it takes 21 days and the CIO's personal approval to provision a network port, or mad cowboy admins from the local tech degree mill wreaking havoc on live systems 24/7.

The sad thing is, because of these cowboy idiots, those of us who take the time to properly learn the sysadmin trade are reduced to form-fillers and button-pushers. I'm all for planning a change properly and not getting yourself into a situation where you can't recover, but when it gets crazy, that's no fun either. (Example I've heard -- RAID 1 drive fails, change management process prohibits replacing the mirror drive without a million approvals, and the other drive in the mirror dies before the change can be approved! A zero-downtime change becomes a multi-hour outage...)

Ironically, VM technology makes things much more flexible, giving the cowboys even more power if you don't control it. I'm guessing the changes in ITIL for the virtual world are going to involve even further separation of duties (you'll have a "VM provisioner", a network guy, a storage guy, and a security guy all involved with any server build or change, etc.) ITIL is sold to CIOs and corporate boards as a complete cure-all for IT ills -- just have your staff follow this entire set of guidance and you will never have downtime. Reality is (a) the cowboy admins will find a way around it, (b) the processes are so paperwork-intensive that the environment stagnates because no one wants to go through the hassle to fix something, and (c) the better IT staffers will leave to work somewhere that gives them just enough freedom to keep things running.

Unfortunate.

I've been a Mac user for a long time, and my day job is doing systems engineering for a VMWare/Windows/Linux mixed environment. Macs are great machines to have at home for general productivity/goofing off/websurfing.

The funny thing about this is that 90% of all iMac buyers are never going to run into this problem. Apple seems like they desperately want out of the computer business and want to finish becoming a music/movie disributor and phone provider. Part of that strategy seems to be getting people to think of their computers as unserviceable appliances.

The oldies among us will probably remember the days of triple-digit percent margin on computer hardware, and Apple hardware waa near the top of the margin list. (I think the original Mac IIfx sold for $9000 1990 dollars, IIRC, without a monitor or keyboard!) Now that everything's turned out by the millions cheaply, I guess that's the justification for not building in upgrade capability...since you'll just trash it in 2 years and buy a new one. Back "in my day," computer equipment waa a major investment and users demanded that they have the ability to upgrade if needed. Now, not so much - iMacs are still high-margin, but not in the realm of insanity. Apple probably took a decision something like, "the iMacs are for students and consumers, if they want expansion they'll buy a Mac Pro."

Way to alienate the "prosumers" who like the way the iMac looks...

Different but similar bubble this time?

So -- I lived in Bubble Central during the last dotcom boom (NYC, not SF) and very clearly remember some parallels:

- Companies with a shaky (or no) revenue model -- check.

- Gimmicky companies with not much beyond an IPO pitch -- check.

- Metric tons of advertising in every form possible for pets.com, drkoop.com, etc. -- check, but different this time

- Talking heads saying "it's different this time around" -- check.

- Insane tech spend, and spending millions on marketing/parties/etc -- not so much.

Don't get me wrong; I think this social media bubble is going to burn itself out eventually. Investors are going to wake up one day and realize, "hey, there's only a certain percentage of people who will pay to play games on Facebook, or subscribe to "LinkedIn Premium Services" and that'll be that. The difference this time is that there's not the huge tech explosion we saw in the late 90s, simply because everyone's on the Internet already so hardware/comms/infrastructure have already been bought.

LinkedIn does have a revenue model - it's the world's biggest hypertargeted recruiter spam machine. That, and all the freelancing warrior types trying to drum up a living are the paying customers. I like it because it makes it easy for me to keep a work contact list. I don't like the constant friend requests from random sales weasels just because I happen to work for Company X and they're trying to weasel their way in.

A new spin on a famous quote -- "When your local scrap metal dealer is telling you to follow them on Facebook and Twitter, it's time to sell."

Cloudy perception!

These kinds of high-profile outages are what it's going to take to temper everyone's overhyped expectations of cloud computing. The concept is amazing -- ever-expanding, always-on computing power that can be used like water or electricity. The reality behind the marketing stuff falls short.

The shiny PowerPoint stuff fails to mention that your typical IT problems don't disappear; they just become someone else's problem. If your host either is incompetent or just has a really bad day, it doesn't mean that your workload and data are instantly going to migrate themselves to another location. It CAN mean that, but only if the service provider designed it that way, and you paid for that service. Also, if there's a true disaster, SLAs are useless other than ensuring you can extract money or free service from the provider at a later date. You're still down and out till it's fixed.

The reality is that for most big-scale applications, DR is difficult. You're either looking at dedicated LAN-speed network links between datacenters for replication, architecting your app so that it can support failover properly, or both. Problem is, once the CIO sees "monthly charge based on use" and "get rid of your local IT assets and staff" in the same presentation, it's all over.

Beer because it's Friday, and Amazon's sysadmins are going to need it after the 72-hour shift they'll be pulling to fix this.

Where have we seen this before?

I could swear I've heard this story before -- profitable hardware engineering company sells off its hardware business to some cut-rate manufacturer to focus on the wildly profitable world of "services." I think the other company that tried this is now a consulting company that sells mainframes in its spare time.

I see some big problems with this:

(1) If HP thinks EDS is a world-class consulting & services firm, they need to take a hard look at it again.

(2) Where are those of us who run businesses that really aren't "cloud-hostable" going to turn for hardware? HP is (was?) the only vendor that produces/produced a line of non-garbage PCs and servers. I shudder to think Dell is now the go-to provider in this space.

(3) HPUX and OpenVMS customers pour truckloads of money into HP every year, and support an entire ecosystem of applications. It would be pretty foolish to tell them to instantly port everything to commodity Linux or Windows servers -er- cloudmachines, and expect them not to say anything.

I'm telling you - this cloud stuff is just running its course. Eventually companies will figure out that it's just VMs and SANs in someone else's datacenter, and draw some sane boundaries about what they're willing to host and what will stay in-house. And when they do, they're going to need to keep buying hardware from somewhere to host the in-house stuff. If HP doesn't retain a toehold in the real world while trying to sell the cloud, things could get ugly.

They're not really seeing the whole picture.

As an American vwith the attention span to read the entire article, I can see a few grains of truth in it. The evidence is there when my employer tries to hire candidates for IT systems integration work. It really is difficult to find qualified people -- and that's not a ploy to hire H-1B workers. The job description is basically "expert troubleshooter who can learn fast, loves being a lab rat, enjoys the torture^Uchallange of making developers' bug-ridden messes^U^Umasterpieces work in the real world, and deals effectively with a huge variety of different people." (And yes, it's a lot of fun for the right sort of twisted individual.) Most of the people I work with are foriegn-born, and it's very difficult to recruit suitable natives. The pay is decent, and the hours are amazingly flexible most of the time, so I don't think that's it. I think the pool is just low.

So, what do I think the CEO Peanut Gallery has right? I do think the education system is a little messed up, or more accurately, we don't place enough emphasis on education. Developing countries have hugely competitive higher education systems simply because education is the difference between working for a multinational in Bangalore or Beijing, vs. spending your life as a subsistence farmer. I also think that artificially reducing immigration makes a talent shortage worse, but an active giveaway like the H-1B visa just depresses wages for those of us who did the right thing and got educated.

However, here's where I think they miss the boat. In order to incentivize people to get a good education, there needs to be a clear career path laid out for them. You can't say "we need more scientists and engineers" on one day and then close your R&D facilities in the US because you can pay Chinese Ph. D.'s 10% of what your US Ph. D.'s make. Education in the sciences and engineering is tough -- it takes a lot of effort, time and money to get through school. I'm in the unenviable position of getting older, and am dreading the day our corporate overlords decide we're too expensive to house in the US anymore. It's a negative feedback spiral from both ends of the spectrum -- older workers who spent the time educating themselves get tossed out, and tell their children that they should spend their brainpower on finance, project management or some salesy-type job. In addition, getting rid of the entry-level tasks means no one who is interested has a place to start and learn the trade, thus amplifying the tech shortage these CEOs are talking about.

If I had these guys' ears for a few minutes, the one thing I'd tell tham is -- if they want talent, innovation and growth, they need to provide a stable jobs framework. If people aren't constantly worrying about being laid off, or whether the huge investment in their educations will pay off, you'll see a return of interested American science and engineering grads. Keep some entry-level positions onshore with the intention of allowing people to grow their skills. And most importantly, think beyond next quarter and realize that your experienced workers are not 100% liability -- that experience counts for something!