Japan-based company Hitachi says it has solved our file woes with a new storage system that the company claims can keep data unscathed for 100 million years. Data is etched onto four layers of a thin sheet of quartz glass using a laser that creates dots that can be read by a standard optical microscope.

I read this article this morning and oscillated between delight and dismay. I'm delighted that people are designing things for the long term, it's great to think that technological solutions could be around in so many years time, rather than designing for replacement in a year or two.

But, as a long term archiving strategy I'm convinced that storage lasting 100 million years isn't the answer and I worry that the focus is in the wrong place. I'm not sure that even having storage that is designed to last more than 100 years is the answer.

To be meaningful, whatever information is worthy of being stored for 100 million years is going to need to be kept up-to-date, to be kept current/relevant. I don't necessarily mean that underlying data needs to be changed, but that the context needs to shift, as does the format. Language evolves. The language used by us, now, will almost certainly be to most people, like ancient Greek is to me, unintelligible. Anything that is really relevant in the long term, like the location of the storage of nuclear waste for instance, needs to be maintained and evolved so that for every generation it's message is understood.

That said I do realise that inventing technological solutions is more fun than discussing archival strategies :-)