In the course of unpacking my swag in my shiny new home office, I found the oldest baseball that I own. It was given to me by my mother, who received it (autographed) from a Cardinals minor-leaguer named Whitey Koppenhaver. You never heard of him because he never made the show; he played and then went on with his life. He currently runs a farm stand in north-central Pennsylvania, and still plays hardball fifty years later.
This has a lot to do with drum memory, I promise you.
Nobody can predict the future; sports figures certainly have made history when they have done so correctly. Ruth's digital signalling to the fence and Mark Messier promising a Game 7 win to his Rangers fans left their mark on New York. Signing his name on that baseball didn't make Koppenhaver a sure thing at short; but I use it to frequently remind athletes, young and old, that very few of us go on to the bigs, most of us play small ball and revel in the little victories our entire lives.
If the future is cloudy, and no secret incantations, wearing of unwashed heather grey t-shirts with the "lucky stains", or select couch real estate can influence the future (or even help the Jets complete a forward pass), then our positions as technologists must be built on things that we believe are reasonable future-proofed. That is, we understand that technical evolution happens, we encourage it to happen, and we make sure that the cost of evolving doesn't become a tax on running our businesses.
Not to sound like too much of a Unix codger, but
I vaguely remember device driver entries for
Assume you can read whatever physical media your long-lost document is in. Now you have bits. Big difference between bits and information -- can you use the bits with whatever application created them, or its modern equivalent? Can you turn the bits into a map, an image, a story, a piece of your family's or corporate history? If it's more than 10 years old and it's not ASCII email, I'd vote "no". Perhaps a minor problem now, but with increasing digitization of everything from photos to legal records to bank check images, it's an upcoming crisis. Anybody else remember XyWrite, WordMarc, WordPerfect, or VisiCalc? I used them all (I warned you, I'm a techno-codger).
Note that for some technologies -- TCP/IP, for example -- this isn't such a big deal. The standards are open, easily and freely implemented, and interoperability is done through plugging the new and old into the same network. If the transports work like this, shouldn't the data flowing on the transports have the same future-proofing? If it's not free to be viewed in the culturally hip app-du-jour, then converting it (or retrieving it) represents an implicit IT tax on the data. Somebody has to pay for making your data fungible.
The question posed to me in Garmish was about Massachusetts and its proposed requirements for adopting the OpenDocument format. This may be the first time Massachusetts, a state famous for taxation, has found a way around the concept. They aren't looking to exclude vendors; they're trying to future-proof the digital records of the Commonwealth. Earlier this week, Sun made a bold step in protecting implementations of OpenDocument format by making a statement to OASIS effectively assuring patent peace. Our Chief Open Source Guy, Simon Phipps captures the essence of the protections.
I see it as future-proofing my personal and professional data. My bits -- all of which are now created and edited with OpenOffice and persisted in OpenDocument format -- are just as secure as my atoms, including one dirt-stained Whitey Koppenhaver baseball.