As best I can determine, the 100 and 200 year life expectancy of CD and DVD media report by NIST was published as NIST Special Publication 500-263 a PDF summary is online here - »www.nist.gov/customcf/get_pdf.cf ··· d=150372
There were a half dozen methodologies used to produce rapid aging, and about 10 test methodologies to test the consequence of aging. The project detail indicates all work stopped after 2006, and the results were not only incomplete but there were breakdowns of equipment and many assumptions made about the failure and its effect on the analysis being performed.
IMO the report is incomplete, testing stopped in 2006, and was never resumed. Several testing machines broke, how that affected testing wasn't investigated. Looking at the NIST spreadsheet data, there seem more questions than answers regarding the production of the report or why testing was never finished (which is unanswered to the best of my searching).
Media longevity tests by the optical media industry (and at least one of the NIST tests) seem to assume temperature as the only stress on media over time. The assumption also is made that temperature range can be controlled over the 100 to 200 year period within a few degrees. The 200 year mark can be hit using BLock Error Rate (BLER) test. Other tests produce different results. Even the 2006 NIST tests indicated very wide / unpredictable variation for the limited testing that was in progress before the plug was pulled on the project.
The international standards organization (ISO) has defined two methods of determining longevity of CD and DVD media. They are ISO 18921:2000 for CD-ROM media and ISO 18927:2002 for CD-R media.
The latter standard involves a 2 year test. How often does the same media, stay on the market, unmodified for 2 years? Production changes occur regularly. A sku may remain constant, but the dye, manufacturing process, and substrate materials may all change a lot over 2 years. At time of purchase without testing, we don't really have a "good" guess as to longevity of any particular media. We hope it'll last a long time, but we can't be certain.
A organization called Optical Disc Archive Test Committee (»www.odat.org
) was created to assess optical data longevity. They discuss the problems introduced in this thread in considerable detail. The problem for current optical, is not very different from the problem with 2.88 MB floppy technology.
The technology moves, and changes, not only the media, but the interfaces and write / read technology. Compatibility with legacy technology is not guaranteed or in many cases even practical.
Another issue brought up by a poster, regarding estimated longevity of data on media at 100 or 200 years, is the difference between a backup and an archive.
A backup, is for intermediate retrieval of something lost. An archive is for long term storage of data. Sarbanes-Oxley requires data be archived for 20 years, GLBA (aka the Gramm-Leach-Bliley Act) requires 10 years, the Patriot Act requires an archive for 20 years, HIPAA requires a 10 year archive.
Finally, I wind up at »www.archives.gov/records-mgmt/in ··· faq.html
this is the part of our government which defines appropriate practices for data archives.
From that site I see the following -
CD/DVD experiential life expectancy is 2 to 5 years even though published life expectancies are often cited as 10 years, 25 years, or longer. However, a variety of factors discussed in the sources cited in FAQ 15, below, may result in a much shorter life span for CDs/DVDs. Life expectancies are statistically based; any specific medium may experience a critical failure before its life expectancy is reached. Additionally, the quality of your storage environment may increase or decrease the life expectancy of the media. We recommend testing your media at least every two years to assure your records are still readable.
Every 2 to 5 years (the limit determined by archive.gov), data must be migrated from old CD or DVD media to new and verified, to be within the definition the government archives suggests for best practices. These standards are likely important to any company or agency that may experience litigation and be required to produce digitally archived data.
This was an interesting thread. It brought up a number of issues to think about, starting with the 2.88 mb floppy.--
"If you put the federal government in charge of the Sahara Desert, in 5 years there'd be a shortage of sand." - Milton Friedman"