Writing in CNET.com, Dennis O'Reilly has published an article about preserving your files for a long, long time. He writes, "It's easier than ever to make sure copies of your most important records, documents, photos, videos, and other personal data will be readable/viewable/playable long after the hardware and software used to create the files have bitten the dust.
However, I found the article to be a mix of great information and some that was not-so-great. For instance, O'Reilly writes, "choose file formats that won't become obsolete." That strikes me as impossible, especially if you wish to preserve information for 50, 100, or more years. He also writes, "use storage media that won't deteriorate or become inaccessible," which is impossible with today's currently available technology. There are promises of some new disks that will last 1,000 years or more but such technology is not yet commonly available. (See my earlier articles at http://goo.gl/kesdk and at http://goo.gl/sgNA8). However, I have to agree completely with O'Reilly's suggestions to "make multiple copies stored apart, and check your archived data regularly to ensure it's still readable."
He also writes, "The general consensus is that CD-Rs should last 30 to 50 years, DVD-Rs less than that, and CD-RWs and DVD-RWs even less. Similarly, tapes and hard disks can be expected to be readable for 10 to 30 years, while portable disks, USB thumb drives, and other solid-state storage devices may survive for half that time, maybe." Then he neglects to mention the simple and obvious solution: in order to preserve digital data for centuries, someone needs to copy the information to new media long before the expiration of the old media's life expectancy.
He then goes on to describe Archivematica's media-type preservation plans to convert .doc, .rtf, and .wpd word processing files to the XML-based Open Document Format (ODF) for preservation and to Adobe's PDF for viewing. The system saves .bmp, .jpg, .jp2, .png, .gif, .psd, .tga, and .tiff raster image files as uncompressed TIFFs for preservation and as JPEGs for viewing.
While I have frequently written about data preservation and generally recommend follow-up reading elsewhere, I would suggest that Dennis O'Reilly's article needs to be evaluated carefully. While generally accurate, it is limited in scope and does not present all the possibilities nor does it mention data archival methods that have already been in use for decades. Nonetheless, he does bring up some good points.
You can read Dennis O'Reilly's article at http://news.cnet.com/8301-13880_3-20026801-68.html
My thanks to Everett Kaser for telling me about this article.