Message-Id: <mailto:199505101511.KAA05309@library.wustl.edu> Date: Wed, 10 May 1995 11:08:19 -0400 From: Jim Lindner <mailto:vidipax@PANIX.COM> Subject: Re: digital image tape storage To: Multiple recipients of list IMAGELIB
I have really enjoyed this thread over the last couple of days, and would like to add a few SHORT comments. Our company is a videotape restoration company, and as many of you are aware, many high density digital formats are variations of video formats, and as such, we have had a great deal of experience with these tapes after they have been sitting around for awhile. I encourage you to check out our www home page, we have several articles on issues in magnetic media restoration, and links to other resources as well. Our address is http://www.panix.com/~vidipax/A few quickies... Many of the problems that you should expect to see are media failures, and some of these relate to binder failure for magnetic media. These type of failures have been called "sticky shed" and the cause is the binder (or glue) system that holds the magnetic particles to the tape. The chemistry in the binders has essentially gone unchanged since magnetic tape was invented, although there have been some improvements. Current research seems to indicate that the culprit is a combination of oxidation and hydrolysis, so keep the tape cool and dry CONSISTENTLY to slow down the reaction.
Evaporated metal tape has been very bad news in the video area, particularly with Hi8 where the tape recording is extremely dense. Problems for this media primarily relate to abrasion with use over the heads and big dropout problems that result when the abraision causes the metal to flake off of the base (there is no real binder in this product). Of course, large dropouts of this type can be catastrophic with data.
CD-R (CD ROM'S) have other opportunities for media failure as well, and I strongly recommend talking to the media manufacturers for their recommendations. I think that you will find that this media was NOT developed with this application in mind, and there are several media failure issues being researched. One obvious area of concern is the potential catastrophic loss of data due to problems with the index track. Another has to do with error correction, and the exceeding of error correction parameters as the media ages. (if you make a recording that has problems which are error corrected automatically when new, additional error over time could cause an uncorrectable error).
Our recommendation in this area is very conservative. Basically we think that ANY decision that is made now is going to be wrong with the hindsight of 50 years from now. For that reason, no ONE strategy has a high probability of success, and so the best strategy is to use multiple strategies, using several media types as well as proprietary hardware / software products, and several copies that are physically separated from other copies due to the potential of natural disaster (fire, theft, flood). It is extremely difficult to recover from a single copy that has problems, it is easier when there are several to work with. Spread the risk, you have a higher probability of long term success.
Jim Lindner
Jim Lindner VidiPax The Magnetic Media Restoration Company mailto:vidipax@panix.com