February 19, 1998
Letters to the Editor
U.S. News and World Report
2400 N Street N.W.
Washington, DC 20037-1177
Laura Tangley
U.S. News and World Report
2400 N Street N.W.
Washington, DC 20037-1177
Deanna Marcum, President
Council on Library and Information Resources
1755 Massachusetts Avenue, NW, Suite 500
Washington, DC 20036-2188
Randy Allen
The Communications Office, Inc.
108 East Del Ray
Alexandria, Virginia 22301-1324
Ladies and Gentlemen:
I was extremely agitated when I read the article entitled "Whoops,
there
goes another CD-ROM" in the February 16, 1998 issue of U.S. News and
World Report (http://www.usnews.com/usnews/issue/980216/16digi.htm).
Statements are made in this article and attributed to the National
Media
Laboratory that are grossly inaccurate. A chart is also shown that
is a
serious misrepresentation of NML research findings. This article
adversely affects the credibility of the National Media Laboratory
(NML),
the Council on Library and Information Resources (CLIR), and U.S.
News
and World Report (USN&WR).
The article makes the statement that "tests by the National Media Lab
show that top-quality VHS tapes stored at room temperature preserve
data
dependably for just a decade." This statement is completely false.
The
statement that "average quality CD-ROMs become unreliable . . . after
5
years" is absurd. NML publications assert significantly higher life
expectancy values for magnetic tape and CD-ROM that those claimed in
the
USN&WR article.
The USN&WR chart, an interpretation of an NML chart, is erroneous and
misleading. Specifically,
** The original NML chart, part of a series of charts, was designed
to be
used in conjunction with government disposition schedules -- storage
media are judged for their suitability for storage for discrete
periods
of time. The USN&WR chart shows media lifetimes on a continuum. The
NML
chart shows that best quality CD-ROM media would be suitable for
storing
information for 50 years, but not for 100 years. Correctly read,
this
indicates that the life expectancy of this media is between 50 and
100
years. The USN&WR chart misrepresents this information and displays
a
maximum life expectancy of 50 years for CD-ROM.
** Judging whether a class of data storage media is suitable for a
specific data retention schedule is different than assigning a life
expectancy value for the same class of products. On the NML chart,
for
some storage times, all major vendors are considered acceptable for
reliably storing information, using the life expectancy of the lowest
quality media as the criteria for this group. For longer storage
times,
only better than average media is considered acceptable, using the
average life expectancy as the criteria for this group. The USN&WR
chart
unjustly misrepresents information relating to the poorest quality
media
as indicative of "All major brands."
** The NML chart addresses the life expectancy of digital magnetic
tape
formats exclusively. The USN&WR chart does not state whether the
information shown is for analog or digital formats, but the
implication
is that they are analog formats. "VHS tape" implies it is the analog
format widely used in the home. Information on the NML chart for
"Data
VHS" tape was specifically for Metrum (Honeywell) digital data
recorders
which use consumer grade VHS tape. Digitally recorded tapes are much
less forgiving of defects and have lower life expectancies than
analog
tapes. Life expectancy information on the NML chart is not relevant
for
analog formats.
** The USN&WR chart is labeled as displaying "the life expectancy of
various media used to store government documents and other
information .
. .." The ridiculously low numbers on the USN&WR chart combined with
this statement leave the reader with the impression that U.S.
government
documents are not being properly stored and important information is
at
risk of being lost. And the National Media Lab is credited as the
source
of this information. The NML chart discusses the longevity of
various
media types without specific reference to any agency, corporation, or
profession.
** The USN&WR chart displays media types, such as "half-inch
reel-to-reel
tape," that are not part of the NML chart.
The information that was misused for this article was obtained from
the
NML web site (http:\\www.nml.org). NML is principally tasked to
support
the U.S. government by providing the best commercial solutions
available
for the tasks of collecting, exploiting, disseminating and archiving
data. However, information that benefits the government can also
benefit
industry and educational institutions. For this reason, NML
publications
and other resources are freely distributed via the NML web site.
NML web site usage rights state that when NML resources are used:
- NML must be informed of each instance.
- No changes shall be made to the content or display of the
original
file.
When informed that USN&WR would be publishing an article using NML
resources, one week before going to press, NML made some simple
requests:
- That the NML life expectancy chart be used in its entirety
without
alteration.
- If modifications were to be made, that NML could not be credited
as the source of the information. NML indicated that it was
easy
for a non-technical person to misinterpret the life expectancy
information.
The above requests were not honored and USN&WR went to press with a
life
expectancy chart that is a gross misinterpretation of NML's original.
U.S. News and World Report is responsible for distributing
misleading
information regarding the longevity of information storage media.
The
information that USN&WR printed did not come from the National Medial
Lab, and we are offended that we are attributed as the source. This
has
created an embarrassing situation and a loss of credibility for the
NML.
U.S. News and World Report owes its readers and the National Media
Laboratory a sincere apology.
Sincerely,
Dr. John Van Bogart
Data Preservation Scientist
National Media Laboratory
Attachment: NML's "Life Expectancy of Various Information Storage
Media:
For Storage at 20 C and 40% RH" chart