Titanic 2020
Aired April 15 and 16, 2000
Listen to the show. (The
recording will be available next week. Please check back.)
You must have RealAudio installed to listen to the show. Download RealAudio here.
This is Internet On The Air. I'm Joan Silvi. Are computers erasing
recent history? Details in a moment.
Funding Credit: Internet On The Air is a production of the University of Michigan
School of Information and Michigan radio, made possible by a grant from the W.K. Kellogg
Foundation.
Move over Y2K, Titanic2020 is here. Networked computing technologies, such as e-mail and
collaborative groupware, have changed the way that business gets done. But these software
applications have also created an explosion of electronic records. Few organizations
effectively manage these documents that are “born digital”.
A report released by CENSA, the Collaborative Electronic Notebook Systems Association,
compares the electronic records situation to the Titanic’s maiden voyage. With
current advances in computing, soon “the number of records produced on the planet
could be doubling every 60 minutes”. Unless organizations address this issue, CENSA
predicts that in 20 years, critical elements of the historical record may vanish. This
loss could seriously impact legal accountability and organizational memory.
According to David Wallace, Assistant Professor at the University of Michigan School of
Information, most organizations are likely to ignore this wake-up call until they
themselves lose critical data. Preserving electronic records over the long term is
complicated by the necessity of migrating digital data to updated software and hardware.
Unlike paper records which may lie dormant for decades and still be easily read, digital
data may require frequent attention or else they’ll become unreadable.
At a cost of $50 Billion, fixing Y2K was expensive. It remains to be seen what price tag
Titanic2020 will bring, but experts predict Titanic2020 will dwarf Y2K spending.
For more information about Titanic 2020 and electronic records issues, and to hear an
interview with David Wallace, visit our website at www.iota.org. For Internet On The Air,
I'm Joan Silvi.
Top of Page
Related Links
For further information, try these Web sites:
Top of Page
The Interview
IOTA interviewed David Wallace in March 2000.
This report predicts massive amounts of data
loss due to poor electronic records management. What events have led to the current ways
in which records are managed?
We started using computers in a big way in this society in the ‘60s, and that was
mostly a mainframe system. In the early ‘80s we moved to the personal computer
environment, and subsequently in the ‘90s we’ve moved to networked
communications.
What happened is that the software market and the innovations in software have been so
dramatic in the last thirty years that people have generally been more interested in the
next generation of software for immediate use. People did not have this awareness about
what the long term preservation needs were at the time. That’s pretty much the main
reason why we are where we are right now. People had a very short time horizon in thinking
about their digital data and they weren’t thinking about the long term preservation
issues.
Also, the media that we’re actually recording our communications on now is a much
more unstable media than the traditional environment of paper or microfilm. Digital media
is susceptible to hardware and software obsolescence, and the actual magnetic media itself
does not have a very long life span.
Do you agree with the report’s findings -
is this a realistic portrayal of the future?
I do think there is a lot of substance to their claim that we’re going to have a
crisis with the fragility of critical records that have been captured digitally, in terms
of their long term preservation. I think their argument as to why that is the case is
pretty valid as well.
They make the argument that people’s records and digital data are help captive within
proprietary software systems. They give a lot of interesting data (or reasons or
rationales) as to why people and companies who innovate software are not necessarily
thinking about open systems and open standards and long term preservation. A lot of
it’s about claiming marketshare and getting a reliable customer base.
If Titanic2020 does happen, what will be the
impacts on organizational and social memory?
The impacts on organizational memory are going to be drastic…Organizational memory
is based in several different areas. One is within people’s heads in their daily work
environment. The other is the information objects: documents that are captured (who said
what at a certain time) and can be referenced in the future to see what can be done in the
past.
If, in fact, we’re going to have this crisis in digital preservation (I think we
will), the implications for organizational memory are twofold. One, is that we’re
going to need an increasing reliance on people’s memories. However, we’re in an
economy right now where people do not stay in careers for 30 or 40 years, or in a specific
work location for 30 or 40 years. That’s kind of a paradox.
This report is important because it underscores the whole notion of migration (which is
transferring digital information from one hardware/software platform to another). What
we’ve tended to find, and what this report is good at demonstrating, is that
proprietary software formats are not very good about forward and backward migration.
Forward migration is being able to read data in previous iterations of software, say, Word
95 into Word 97. Backward migration or compatibility is the ability to read a Word 97
document in a Word 95 platform.
What they tend to find is that proprietary software companies are not always very diligent
about accommodating backwards and forwards migration. In instances in which they do try to
account for it, the documents don’t get translated exactly. You lose some
functionality in both directions. The authenticity and reliability of the documents can
get injured because you’re not having the same exact document that was originally
created. A lot of our legal [requirements] for admissibility of electronic documents, or
documents period, is based on these notions of authenticity and reliability.
What are the solutions - can Titanic2020 be
prevented?
I don’t see one stream solving this problem. The picture that [CENSA] tries to
paint is they liken this data loss to the Titanic’s maiden voyage. I think
there’s a lot to be said for the argument that you need large scale crises to get
people’s attention. Maybe we will need some critical data losses to get people’s
attention. They’re trying to raise people’s awareness to stave off what they
call the “icebergs of the information superhighway”.
What people can start thinking about is this notion of open standards, looking at
information in software that can be easily captured and migrated to other software
platforms as well as other (proprietary) software environments. The problem, of course, is
that the people creating the software don’t have that same interest necessarily all
the time.
The report is pretty good at citing what the proprietary software [industry’s]
inclinations are towards open standards. They talk about the cost involved, the poor
current acceptance, the short life time of most standards, and what they call the
“not invented here” syndrome. I think, basically, people recognize the value of
open standards, but a lot of proprietary systems’ marketshare is tied to the
specifics of their software. To create an open standard, is almost like opening up the
door and letting your customers go to other stores. And that’s viewed as a negative
consequence. However, we do need open standards.
One standard that has become a de facto standard (not an open standard) on the Web for
documents Adobe Acrobat’s .PDF file. [Adobe] actually gives pretty good information
about how .PDF is unique in some regards and that is has a long-term preservation
orientation. I think they’ve committed to supporting it for up to 25 years into the
future. Maybe, as we hit some critical data losses in the next four to five years,
it’s the kind of wake up call that people need.
If you tell people that they need to plan for it now, that this is going to be a problem
in the future, they have many more competing concerns for their current resources and
expenditures. They’re not likely to take this very seriously until they hear about
another company or a government or an organization that loses critical data in the same
kind of software environment that they’re living in. That tends to get people’s
attention. My sense is that people are not going to be very pro-active in solving this
problem. They’re going to need to get some sort of wake up call through critical data
loss.
How do these predictions compare with your
research on e-mail policy?
We’re also living in a time of great inconsistency, or troubling times in a
regard, because we’ve gone ahead and committed ourselves dramatically to using
computers and computerized networks to create documents. Be we don’t have an
infrastructure for supporting the digital preservation of electronic records. The de facto
standard in society for things like e-mail is to print it out onto paper and file it.
What’s interesting is that we are using e-mail in ways that we never could have
imagined ten years ago. The sheer volume of e-mail that is created is pretty overwhelming
to most people’s ability to manage their own information. To me, this notion that
people are somehow going to be pro-active and print out their e-mail and file it - some
people may be doing that - but for the most part, in my mind, this is going to be the
biggest misperception about how people are preservation information right now. I
don’t think they are going to be printing things out. The future archival record is
going to suffer dramatically from this strategy. If you look at the sheer magnitude of
e-mail - To print it out and file it is well beyond most people’s patience or
resources.
To reiterate, society is now creating, capturing, and managing its information resources
through computers. What we have not done, is to come up with a way to preserve this
information over the long term. We’re very diligent about using computers to create
new kinds of information, but we haven’t been as diligent about protecting the
survivability of this information for future generations.
Please direct questions or comments to iota.webmaster@umich.edu.
Last Updated April 11, 2000
|
