Stability of Electronic Media for Publication/Data Storage in Taxonomy and Systematics.

[Stuart G. Poss]

In response to T. Schlemmermeyer  Sylvia Hope wrote:

>  I vaguely remember having read that the life time
> of a CD is about 5 YEARS. Is this right or highly over- underestimated?
> Beside this, images in the computer might get lost due to tecnical
> problems and viruses.
>
>           These may be problems, but digital images can be copied in
>           advance of their deterioration to a backup storage device
>           without any loss of information - not true of analog images.
>
>           Sylvia Hope

TAXACOM has seen this issue before but it has never been successfully resolved.  For those
interested, there is an excellent review article in Scientific American by Jeff Rothenberg dealing
with the longevity of electronic media (Ensuring the Longevity of Digitial Documents. S.A.
1995:42-47 [January issue].  An important point made in the article was that it is not only physical
integrity of media that are at issue, but also the obsolesence of hardware and software necessary to
transfer it.  Thus optical media has a physical lifetime of about 30 years, the technology
(algorithms and hardware) needed to transfer it become obsolete in about 10 years.  This varies
according to Rothenberg by media type ( magnetic tape- 5 yrs to obsolesence/1 year physical life
time; videotype 5 years to obsolesence/1-2 year physical life time; magnetic disk 5 yrs. to
obsolesence/5-10 years physical lifetime).   While paper media have a half-life in the hundreds of
years, electronic media are no where close to this yet.  Consequently, serious thought needs to be
given to how to support the required transference of information in this brave new world of digital
taxonomy and systematics.

While I believe taxonomy and systematics need to make use of digital storage and communication of
information so encoded.  We  must do so, if for no other reason to provide information critical to
the study and preservation of biodiversity in a sufficiently timely and resonsive manner to make
much of a difference in stemming its loss.  This is particularly true now that mankind has for all
practical purposes appointed itself as the primary monitor and handler of natural processes that in
the past have supported the continuance of biodiversity.  Nonetheless, we must be cognizant that we
do not yet have in place an infrastructure necessary to effect such transfers of data and
publications on anything other than an ad hoc basis, now done largely through the foresight of
individual investigators or institutions.

If we are to be able to eventually take advantage of fully electronic publication in taxonomy, where
the preservation of historical record of nomenclatural usage is of prime importance, we need to
think through a collective strategy for addressing routine and comprehensive storage and
tranfsfer.   There are many difficult questions that have no easy answers.  Since any comprehensive
effort will be expensive, who will pay?  Shall only works by those able to pay or have access to the
resources be saved for posterity?  How can we insure that obscure works, those of limited
distribution, or those that languish in the obscurity among vast electronic warehouses?  Is this
primarily the responsibility of the authors, the "publisher", librarians at the host institution,
governments, scientific societies?  If it is to be individual investigators, who will carry on after

we grow old and our own cells fail to adequately replicate?  Given its increasingly diminished
resources, is the taxonomic/systematic/scientific community even up to addressing the challenge
posed by this problem?

My own view is that we need a collective response, perhaps with societies and institutions assuming
the primary responsibility to assure continuity, with the assistance of individual scientists and
the governments at various levels.  It might not be unreasonable to argue that such archival efforts
might best be undertaken by research archives themselves so that the information being saved can be
as closely associated with the reference materials in collections used to generate it in the first
place.  However, this will require much greater resources than archive collections now receive and a
more integrated, cooperative resonse than many such institutions have in the past supported.

There appears to be a related problem with other library materials as well.  Perhaps as
institutional libraries grapple with this issue, the role and potential collaborative involvement of
libraries and research archives are clearly relevant.  Hopefully the upcomming workshop on data
standards will address this issue, at least tangentially.  Otherwise how will standards be useful,
if they have little permanence.  Perhaps considerable cost savings accrued by reducing the flow of
dollars toward traditional publication houses though use of fully electronic publication could be
used to finance some of this.  In any event, the overall set of problems doesn't appear to be going
away soon.  To the contrary it may be growing more and more acute.  While we move on to our next
project, we tend to forget that random individual bits of our last project are begining to decay.
Should I take my frustrations out on another tree?  Is continuing to "carve our names" on trees
still a viable alternative to solving this problem?

--
_____________________________________________________________________
Stuart G. Poss                       E-mail: sgposs at seahorse.ims.usm.edu
Senior Research Scientist & Curator  Tel: (228)872-4238
Gulf Coast Research Laboratory       FAX: (228)872-4204
P.O. Box 7000                        703 East Beach Blvd.
Ocean Springs, MS  39566-7000