The legacy document mountain is a real issue that will not go away because of any theoretical interoperability between Office, Office:Mac and OpenOffice. In many companies, Excel and Access underpin critical business processes that make financial statements that are subject to close scrutiny by aggressive regulators.

You might choose to trust that a spreadsheet compiled from a different code base will give you the same numbers as Excel for Windows, but not many CFOs would be as bold as you. If they cannot explain why their numbers look different this month they face jail.

For a capital markets organisation you also have pricing models in Excel that are trusted to expose you to significant liability. That trust would evaporate if the model were migrated to a different platform.

We know what the answer is: don’t use Excel and Access to underpin critical business processes. But its too late – they already do.

To get yourself out of the mess you can (a) wait for the spreadsheets and Access applications to die a natural death or (b) spend money on converting and testing them. Or probably both.

Did we do this when we upgraded to Office 2003? Not sure, it was before my time and in any case the code base was largely the same as Office XP so the risk was lower.

If we are at cross purposes it is more likely I’m not making myself clear. We are in agreement (I believe) that we need to step out of our walled garden and start using tools that are designed to be interoperable. I’m just not clear how this enables a change of corporate platform any time soon.

Looking forward to your future posts.

Microsoft Office is something that is ALREADY available on Microsoft as well as Apple, and Open Office is moving closer to Microsoft functionality on Linux.

The things that held us up before, things like proprietary extensions to Excel, they are now being built for all three “platforms”, as modern ecosystems emerge on a platform-independent basis.

So Office is no longer a lock-in, not when it’s available across all the “desktops”. And migration from Office will become easier as people get into SaaS, something that even Microsoft is doing.

Google isn’t exactly sitting quietly, and as you know there are a number of new tools out there that are getting traction in institutions. Not just blogs and wikis and 37 signals, there’s a whole new world out there.

When Dell has to take Linux seriously, when PC World shifts Macs, when McDonalds sells healthfood, something’s happening.

About your legacy mountain problem I am less in agreement. Any firm that defines its desktop strategy in the context of its legacy documents alone…..

More later, this is a subject I intend to follow up with a number of posts.

If I understand your point, it is that the functionality we need on the desktop is available from community-generated software and that the underlying platform will become increasingly irrelevant.

Your comment on the low cost of migration applies to this platform of community-generated software built, as it has to be, on de facto standards of interoperability.

It certainly does not apply to today’s platform. The world’s desktops are 95% Windows and of those 95% are using Microsoft Office. This has been true for some time now. This means there is a incomprehensibly massive mountain of documents in the proprietary Microsoft Office format that will prevent any migration in the foreseeable future.

I have done a bit of research in this space as you know. My remarks apply principally to the corporate desktop – the barriers to migration are much lower in the home and the small business.

For corporates the medium-term future is Microsoft, and Microsoft knows this very well. Discounts are shrinking as Microsoft tries to milk its cash cow.

Is this their preparation for a longer-term future when they do not have a desktop monopoly? Undoubtedly. And of course Ray Ozzie has already set the wheels in motion for a different business model, based on supplying services rather than software. He is smarter than me and so are you – I may be wrong about this, but I don’t see his new model becoming the predominant one for Microsoft for some time, and I don’t see the corporate desktop running community-built software for its core activities for some time either.

That isn’t a good thing. We can do a lot to move people into a more sociable way of working today but we have to get everybody working socially before we can even start to run down the backlog of proprietary documents in circulation. Some of them have a surprisingly long lifespan according to my research.

This is a long game.