So this week I’ll chat a bit on what it takes to make two very different products interoperate as seamlessly as a common platform does. I’ll explain further why I think Microsoft’s only real choice may be to Open Source their platform, and I’ll reiterate why that could work better for them long term.
I’ll also point out how the EU, through this judgment and their behavior, are actually ensuring that nothing close to the interoperability they want will actually be achieved. Not that a government working cross-purposes is unique – the current U.S. administration seems to make an art form out of it. Be aware that I’m typically skeptical of government actions in free markets. They tend to move too slowly, are more interested in looking like they are accomplishing something than actually accomplishing it, and often do more damage than good as a result.
|Smart IT Columns|
Integration is Hard
Getting two different products to work together seamlessly is hard. As anyone who has worked in a diverse IBM shop and used System View knows, the promise of high interoperability was never truly met across IBM’s diverse platforms. Some may remember AD/Cycle as another initiative that was to bring commonality in IBM; it died a painful death after making almost no actual progress.
Eventually IBM bought and implemented a third party product (Tivoli) which did a far better job of allowing IBM’s platforms to be centrally managed, but natively IBM’s various platforms still lack the level of interoperability that the EU seems to want between Microsoft and its competitors.
Inside Microsoft, to solve this problem between Microsoft’s own products (and they supposedly had one core code base) they first had to create and implement a broad program called .NET and then had to hire the granddaddy of all collaboration experts and give him Bill Gate’s old title to make progress. They are making progress but it has taken years, a massive amount of management oversight, penalties and incentives, and it still has a long way to go.
If we take companies that have been created though merger, like CA, interoperability problems have been incredibly nasty, making everything I just mentioned look like child’s play. The turf wars in just getting stuff done were likely legendary in that company and go to why Oracle, when they bought PeopleSoft, just decided it was easier to shut down the acquired company than to get a high level of interoperability between the two company’s products.
Having one side of an interoperability problem be forced to interoperate without putting some effort into ensuring the other side will work toward the middle is like asking someone to clap loudly with one hand (without using another surface, for you whacky engineers out there).
Next page: Interoperability Goes Beyond Product
Interoperability Goes Beyond Product
Now let’s say they were successful and that two products from vastly different code basis could be made to work together in much the same way that two products from the same code base can work together (and that’s one heck of an assumption). What about support? If there’s a problem isn’t it likely one side will point to the source of the problem as the other side? Has anyone out there not run into this problem all the time between vendors?
What happens when either side patches? If the two sides don’t have a sustaining relationship, what’s the incentive for either to ensure that key parts of the resulting solution continue to function? When you are trying to match the integration between similar products this is very difficult because similar products have an inherent advantage when it comes to support and patching. They tend to get patched at the same time and, since the platforms are so close (or even identical) support staff is trained on the solution.
|Smart IT Columns|
Once you introduce a completely different product into the mix, the opportunity for breakage goes through the roof. You can have conflicting patches on both sides, conflicting support practices where the process you use to fix something on one platform may bring the other one down depending on the dependency. The opportunity for a truly catastrophic problem goes up strongly.
This is exactly the wrong time to have two companies or groups pointing at each other screaming “it’s your fault!” The IT folks just want the stuff to work.
This is why the Novell, Sun, and HP partnerships are so critical. By agreeing to cooperate and link development and support they deal with the hard part up front. And even if the products don’t interoperate seamlessly, the knowledge base that the joint support group acquires, over time, and the capability of actually creating better solutions and workarounds is greater because they are more often working together.
Why the EU is Forcing Failure
Now I’ve pointed out that even with a lot of incentives and executives running around with whips and chains on both sides, integration is hard. But it can be made harder. Let’s say you have two people working on a project and you tell one of them that when he or she is done you’re going to cut or reduce their salary by about half. I’m willing to bet – regardless of your threats – they will drag their feet forever.
The EU has been doing just that. Instead of trying to provide incentives to get the parties to work together, EU leaders say their goal in all of this is to dramatically reduce Microsoft’s market share. And then they wonder why no progress is being made. You almost want to wonder how they find the light switch in the morning.
Who in their right mind asks for a behavior and then vocally creates massive disincentives in the way of getting it? Politicians and bureaucrats evidently…
Next page: How this could work
Getting This to Work
First, if you want to change a company’s behavior you need to be able to hold individuals accountable who created the problem in the first place. If I know that I’ll be retired before anyone can actually do anything, or in another job, there is little incentive to behave.
Second, interoperability between two separate products requires movement on both sides to get done. Having one side do it is like trying to clap with one hand, it just won’t happen.
Third, the problem wasn’t created by Microsoft’s market share, it was partially created by the lack of adequate and timely law enforcement. The EU shares the blame for the problem, and accountability should likely flow both ways so that the core problem will not repeat with another company and/or another industry. Microsoft isn’t the sustaining problem, a process that takes 5 to 10 years or more to complete is.
|Smart IT Columns|
Fourth, the punitive action has to fit within a framework of things that actually will work. Completely different products won’t perfectly interoperate ever, but with a mix of services and technology you can get really close. You have to find a way to drive alliances that will work. And you typically do that with a combination of incentives and penalties. Or you can set and enforce standards as a government but that will slow significantly advancements in your geography (when was the last time AC plugs changed? Or electrical power in general changed, or gas systems changed, or…).
Fifth, while fines can be effective some portion of the fine should go to help fix the problem. Having all the money go into a general EU pot, which appears to be the case, makes it look like the EU is just looking for excuses to milk a U.S. company, creating natural animosity between nations. Should nationalism kick in, the cost to the EU could vastly exceed the benefit.
Sixth, the experts used in cases like this should have some perspective of history. For instance, AOL put forth a complaint similar to the one regarding the Windows Media Player when Windows 95 launched. History proved, much like it has done with the Windows Media Player, that the complaint was groundless given alternative delivery methods (and we all got free floppy disks, and later CDs, from AOL for awhile).
Seventh, whether in government or business every organization needs someone that, from time to time, raises their hand and says “this is stupid let’s not do it.” It would keep a lot of companies from getting into trouble and it would keep a lot of governments from losing the confidence of their constituents.