Thursday, March 26, 2009

Is Operating System Push-Back Inevitable?

I'm afraid Microsoft and the rest of the business world is in for a bit of a surprise.

Yes, the economy is down. So that will affect some sales. But on a grander scale, there are other forces that will keep the sales of some new operating systems lower than expected.

I have recently been in public forums where Microsoft employees are probing to find out why an operating system hasn't been selling very well. No. It's not Vista.

It's EBS.

I contributed some thoughts (a shocker, I know). I believe there's no market for EBS. If a brand new company needs exactly three servers, then it's fine. But if a company is growing from one to two to three servers, it's nearly impossible to justify.

Let's say your first server is a file server and runs exchange. It might be SBS. I hope it is SBS!

When you get to the point where you're adding a server, it's probably SQL or Terminal Server. If you have SBS Premium, then you already have the licensing to just install these once the hardware arrives.

Stop. What does that network look like? Does it look the same as 90% of your network maps from for the last six years?

By the time you get to three servers, you've grown there organically and the easiest thing to do is to add one server, not buy three new servers. Formatting two servers to totally install EBS fresh on existing hardware would be hard to justify.

Two critical opinions determine whether a specific operating system gets sold into this environment:

1) The consultant believes that it's the right operating system and therefore recommends it.

2) The client believes that it's the right operating system based on what they know.

The client may be a former network engineer or know zero about computer-thingies. In either case, the client's knowledge comes from the press, the consultant, friends, neighbors, relatives, and loud-mouth know-nothings who happen to have an opinion.

So, here's the deal with EBS:

There's no market for this operating system. Pretty much the only appropriate client is a brand new company that needs exactly three servers. That population was small until October 1, 2008. Now it's non-existent.

A consultant might recommend it in an attempt to make sales. Other than that, there's no reason for a consultant to recommend this O.S.

As for the client . . . how could they even know about EBS? They can't, so they are in no position to think it's the right O.S. for the job.

- - - - -

Enough about EBS. Microsoft has introduced OS's that flopped before. We all either remember Windows ME and Microsoft Bob or we've heard about them.

No, EBS isn't' the big problem for Microsoft.

Microsoft's real problem is their past ability to build great operating systems.

What was the biggest challenge to the adoption of Windows 2000? NT 4.0.

What was the biggest challenge to the adoption of Server 2003? NT 4.0.

What was the biggest challenge to the adoption of Server 2008? Server 2003.

What is the biggest challenge to the adoption of Windows Vista? XP.

What will be the biggest challenge to the adoption of Windows 7? XP.

With every generation of operating system, there's a larger and larger community that just wants to stay where they are.

This is complicated by a growing community of hardware and software providers that refuse to update their stuff for the new operating systems.

Some vendors, like HP, work very hard to keep up with every O.S. They embrace every product Microsoft introduces (Example: They're the only brand-name provider of an off-the-shelf Home Server box.).

But LOTS of other vendors never upgrade their crap. Most of these lazy so-and-so's wish that we still lived in a Windows 98 world.

They program in .net 1.0.
They use SQL 4 instead of the very modern 5!
They don't really understand ActiveX
. . . or anything created in the last five years.

- - - - -

I am shocked at the laziness of software developers.

I am shocked at the laziness of hardware developers.

I am shocked at the laziness of SOME computer consultants.

I was just shredding an old case that involved this phenomenon. The client showed me the quote from the freakin' moron consultant who came before.

This schmoe recommended -- in writing -- that the client buy seven new computers with Windows 98 and then pay the consultant to remove 98 and install Windows 95. Here's the best part: "We don't know the quality or stability of the new operating system. Windows 95 is stable, reliable, and well understood by the consultant."

In other words . . . "I'm a moron. I refuse to learn the skills required to advance in my so-called profession. I hope this client is stupid enough to spend the next three or four years using an operating that's already obsolete. And I get to charge by the hour to replace new, working stuff with obsolete old stuff. Woo-hoo"

Thank Goodness such Bozos represent a microscopic portion of our profession.

But since our profession has no barriers to entry, these folks will always be around.

- - - - -

The problem is very complex, taken as a whole.

If you have a machine shop that uses an RS232 interface to control welders, routers, saws, etc., and you're not connected to the Internet, then you can survive for the next 100 years with Windows NT. In fact, your life will be easier with NT4 than with anything that's newer.

If your office receives reel tapes in EBCIDIC and you translate to ANSI, then you need an HP 3000 with MPE/X.

If your primary line of business application was written by a lazy bastard who refuses to keep up with his chosen profession, and you're out $250,000 for the software, then you need whatever operating system and patch level he uses.

If you make decisions about your server, network, desktops, and the rest of your I.T. infrastructure so you don't have to replace a seven year old printer, then you get what you deserve.

But . . .

For 90% of all clients in the SMB space, it will ALWAYS be cheaper and more cost effective in the long run to use the latest operating system and software from the leading providers.

Today (and for the last two years) that means Vista.

Next year that means Windows 7.

Today it means SBS 2008.

And Office 2007.

You can make all kinds of excuses about living in the past.

If your client has no opinion, she is most likely to go with whatever's installed right now. So her uneducated opinion will always look backward instead of forward.

And that leaves your opinion as a consultant, because your opinion and the client's opinion are the only two opinions that matter. If you don't look to the future, your client won't look to the future.

Despite all the bad press, there are about 200 Million legitimate copies of Vista in the wild. So if you forget the B.S. and look at the numbers, it's a hugely successful and spectacular operating system.

So when Windows 7 comes out, it will have to compete with
people who refuse to move away from Vista,
people who refuse to move away from XP,
people who refuse to move away from 2000,
people who refuse to move away from NT4,
people who refuse to move away from 98,
people who refuse to move away from 95,
people who refuse to move away from NT3.x,
people who refuse to move away from Windows 3.1

and the occasional Unix machine.

The point is: Every time Microsoft releases a really great operating system, they create another competitor to their future operating systems.

- - - - -

I am now hearing whispers that SBS 2008 hasn't taken off as Microsoft wished. Again, part of that is the economy. To be honest, part of it is the price jump, especially for Premium.

But part of the resistance is the absolute success of SBS 2003. 2003 changed the landscape of the small business network. It changed the standard practices of the consulting industry.

So the biggest challenge to SBS 2008 is SBS 2003.

Not because we won't love the new technology. But simply because the last operating system kicked so much butt.

This push back against new operating systems is inevitable. As the base of installed systems grows, the base of resistance to new operating systems grows.

New technology will always push operating systems forward. But just like office products, 95% of us only use a limited number of core functions and we need those core functions to work reliably over time. With luck, they will also work the SAME over time. When you get something "right," you don't need to change it for change's sake.

Our company policy is to sell the latest products available unless there is a compelling business reason to do otherwise. That simple rule will keep us on the forefront of knowledge about what's going on in our profession.

But I fear that philosophy is no longer the majority opinion. And even if it is, it won't be for long.

Computing in general has reached a point where resistance from the installed base will exceed the desire for new technology.


Visit for books and more!


  1. An unusually rambling post for you. I certainly agree with most of it. Microsoft is their own biggest competitor which means that they do alot of things right. We should all be so lucky.

    You're missing the boat on EBS though. The point of EBS isn't the former SBS customer. It isn't the magical just launched company that needs 3 severs right now. Every product has a sweet spot and EBS's sweet spot is 100+ users. Those companies have 15 or so servers each performing a specific function, probably running on cobbled together hardware. EBS is a consolidation product add virtualization to the mix and it's story just gets better.

    EBS benefits from the stability of the 2003 product line so it ran out of the gate as a solid product but TA DA! the Internal IT guy working at these companies doesn't want it. (S)He doesn't get that efficiency and a smooth running network won't put them out of a job. But they fear it!

    Like small business, most of the lower medium business don't work with consultants - yet. They have a glorified desktop guy with the title of NETWORK DIRECTOR running the show and he knows his knowledge is out of date and is afraid to change because it might expose him.

    EBS has a market in 100+ user range and it is deperately needed. The trick is convincing those companies that what they are doing now, really isn't working for them.

  2. Thanks, Amy.

    As Pascal would say, "I would have made this post shorter, but I didn't have the time."

    We don't deal with many 100+ clients.

    But Microsoft has been asking us to sell EBS to the 50+ market.

    In that 50-100 space, I just don't see any prospects.

  3. Hey Karl!

    I couldn't agree more... with 99% of your post.

    The only point I disagree on is this: "For 90% of all clients in the SMB space, it will ALWAYS be cheaper and more cost effective in the long run to use the latest operating system and software from the leading providers. Today (and for the last two years) that means Vista."

    Now, I don't have a problem with Vista. I've been running it as my primary OS both at home and at work since the first release candidate. It's fine, but as far as end users are concerned Vista Business Edition is just a prettier, bloatier version of Windows XP Pro. The only significant feature it offers to my clients over XP Pro is the UAC - which is a tremendous improvement to the OS's security posture and frankly came years too late.

    However, I won't go so far as to say that the UAC is *necessary* in an otherwise well secured and managed network environment.

    Oh, there's also Bitlocker, but since Microsoft bizarrely didn't include it in Business Edition and it therefore requires the added cost of Ultimate, we only recommend it for notebooks containing sensitive data.

    I also believe that there's value to even a small company maintaining a standardized desktop configuration across their organization: Same OS, same Office version, etc.

    Now, I do have an occasional client that wants to do a full replacement of all of their desktops at once or over a short period, and of course we get some startups that need all new stuff as well. In those cases, it's usually Vista all the way. But by far the bulk of our PC sales are to companies that already have 5-35 PC's running Windows XP Pro.

    So the problem is that there's no business case for them to adopt Vista on those existing machines. Heck most of them would have to have their RAM doubled (or more likely quadrupled) just to get somewhere close to the performance the users were used to from XP. Some would have to be replaced entirely, and for what? The UAC? No, I can't look my clients in the eye and make that recommendation.

    So, if my clients want to maintain that standardized desktop I mentioned above then I of course supply the new machines with XP Pro since they're not really losing anything by downgrading.

    Oh, and the machines are downgraded from Vista Business to XP Pro at no extra cost by the OEM, so we're not profiting from any extra work there...

    Anyway, Office 2007 runs fine on XP, all new printers still ship with XP drivers, every new app is XP compatible... In short, Microsoft and the rest of the development community haven't given me a good reason to push Vista - so I don't.

    And by the way, I've seen that 200 million Vista units stat too, but I think it's extremely misleading. Of course they've *sold* that many licenses, but are there 200 million copies of Vista running? Not likely. I mean, my little company alone has sold a few hundred Vista licenses in the past couple of years, but there are only a few dozen that were NOT downgraded to XP Pro. The same is even more true at the enterprise level. Off the top of my head I can name half a dozen large organizations that have collectively purchased tens of thousands of Vista licenses and immediately exercised their downgrade rights. The only place Vista has truly penetrated new PC sales - and stuck - is at the consumer level.

    With all of that being said, I think Windows 7 shows some promise. Already about 25% of my staff are running the beta as their primary OS at work - and loving it. One guy is running it on a 5.5 year old laptop and swears it performs as well or better than XP Pro did. Ironically, our only complaint with Windows 7 so far is that they've weakened the UAC, but there's still time for that to change before RTM and I know Microsoft is getting flack about it.

    So when the time comes, the big question for us will be: Did Microsoft put anything into the RTM of Windows 7 Business Edition that will make it worthwhile for my clients to consider a network-wide upgrade? I hope so.


  4. The funny thing is, Office 2007 is a LOT harder to learn for users of older versions.

    Vista is essentially transparent to the user. Other than the fact that they messed up the windows explorer.

    You click on start and open your program. Done.

    My "90%" comes from the fact that virtually every user out there could start using Vista tomorrow and they wouldn't even know the different.

    It run IE, Word, powerpoint, Excel, InDesign, Outlook, and Quickbooks. That covers 90% of the end users.

    If they're not dealing with outdated hardware and software, there's no reason to not go with Vista.

  5. Anonymous6:49 AM

    If your primary line of business application was written by a lazy bastard who refuses to keep up with his chosen profession ...

    HEY !!! You know that client too? ROFL!



Feedback Welcome

Please note, however, that spam will be deleted, as will abusive posts.

Disagreements welcome!