Tuesday, July 31, 2012

Microsoft gets snared by "embrace and extend"

The irony, she is rich.  The Internet Explorer 6 web browser was introduced with great fanfare in 2001, and when Microsoft Windows XP shipped two years later it was the default browser - in fact, the only browser shipping on Microsoft desktops.  By then, it had almost 90% market share.

Ignoring the almost decade long record of terrible security problems, companies soon found that their business applications had mostly been written to only work with IE, instead of other browsers.  Netscape stumbled and died the death, and Firefox struggled for years against this.  Sure, you could use Firefox, but there were certain critical corporate business apps that just didn't work with it.  This "embrace and extend" standards strategy had served Microsoft well in the past, and this was no different.  It was simply too expensive for most companies to rewrite their critical business apps to work with other browsers.

Well, fast forward ten years, to where Microsoft wants its customers to upgrade to Windows 7 (and soon, Windows 8).  The problem?  IE5 doesn't run on Windows 8, and the business apps don't work without IE6.  And so companies are staying on Windows XP:
“We continually hear that legacy web applications are the number one blocker to migration. When it costs millions of dollars to rewrite or replace a critical business application, migration projects invariably stall until a cost-effective solution can be found," the blog coninued.

Browsium offers a web browser plugin called Ion, which runs IE6 and IE7-only apps in IE8 and IE9 on Windows 7. It does this by recreating the IE6 environment, including configuration files and security settings, within newer browsers.

Microsoft’s own advice on moving apps off of IE6 and IE7 isn’t particularly helpful – in fact it’s probably compounding the problem: developers are told to rewrite old apps, which will cost time and money.
Now you need to take some tech stories with a grain of salt - the Press Release doesn't make for good technology any more than it makes for good science.  But the cost of Microsoft lock in is real, and the irony that Microsoft is getting hit by that (intentional) installed base inertia is rich indeed.

8 comments:

George said...

We used to describe it as "ordering pizza for 250 million."

Upgrading is very tough for any large installed base without a lot of focus. Which is why Windows 8 is going to be a big flop, I think. The new UI is pretty and all, but corporate adoption is going to be slow.

Rick C said...

Have people bashing the new Start screen even tried it? I've hated the Start menu since Vista, when MS fixed the size of it at "too small." The Start screen allows far more stuff to be on screen with much bigger hit targets.

There's things to complain about with the changes to the shell--like the way the Sleep command was hidden--but the Start screen ain't it.

George said...

The challenge in that any change that forces end user training is a very, very expensive proposition. Without very specific, definable business value it's really hard to force a change. That's one reason why corporations tend to be laggards (app compat is another).

Assume you are a 10,000 person company. And MSFT makes a change to the start menu. And say it takes one hour to train users on the new system. That's 10,000 hours of lost productivity in year one. Assume the average salary is$50 per hour, and you have to be able to demonstrate $500,000 worth of savings just to keep even. And that assumes you are on a licensing agreement where you already own the software.

Jake (formerly Riposte3) said...

"The challenge in that any change that forces end user training is a very, very expensive proposition. Without very specific, definable business value it's really hard to force a change."

This.

This is why my office - with just 7 people - is still running Office 2003. I dread the upcoming ordeal of getting the attorneys and other staff up to speed on the Ribbon idiocy when we can't avoid it any longer, and it's going to absolutely kill everyone's productivity, including mine.

Which raises another point: your example doesn't even get into the added cost of paying the trainers, whether it's a current employee pulled from his regular job (which is how it would work for my office) or an outside trainer hired just for that purpose. It also doesn't cover lost productivity while people get used to the new software after training, if there's a significant difference between the old and the new.

The compatibility issue is why only one computer (mine) has Windows 7. Our client management software doesn't play nice with Win7, and the upgrade is very close to prohibitively expensive for a business our size, but it will have to happen eventually, since XP is no longer an option for us when computers get replaced.

And Rick C: I have used the new Start screen. I will continue to bash it, because no matter how well it might work for touch screen systems, it sucks for regular desktops and laptops.

Old NFO said...

LOL, yep, coming back to bite em in the butt! And I'm still running 2003 also!

George said...

Ah the Ribbon. An almost verbatim conversation with the Office team:

OT: "The new ribbon interface is way cool!"

George: "Yeah, I know. How does my (Fortune 50) customer disable it?"

OT: "You can't!"

George: (Long dissertation per the above, explaining the economics of end user training to the guy who has never been out of Redmond nor met a customer.)

OT: "You don't understand. The Ribbon interface is better!"

George: "I know! But it will cost Fortune 50 almost a million dollars to adopt it. Can you tell me how they are going to save *more* than a million dollars to move to the new Office?

OT: "You don't understand..."

George: *Facepalm*

Anonymous said...

Well to get round the problem you fire the older employees and hire younger ones saves a fortune on salaries and training.

kx59 said...

Windows 7 is just beginning to be deployed in our office.
With no 64 bit printer drivers for our large format plotters, we've been hanging on to Windows Xp for dear life. Plunking down $60 or $70 grand for two new plotters just to get print drivers that work with Win 7 just wasn't an attractive option.