Sign In

Communications of the ACM

Communications of the ACM

'We're Going Backward!'

ACM Past President Vinton G. Cerf

In caves in Lascaux, France, magnificent artworks were discovered from 17,300 years ago. Cuneiform clay tablets written over 5,000 years ago are still readable today (if you happen to know Akkadian, Eblaite, Elamite, Hattic, Hittite, Hurrian, Luwian, Sumerian, Urartian, or Old Persian). Egyptian hieroglyphic writing was more or less contemporary with cuneiform and papyrus manuscripts dating from about 4,600 years ago have survived. The Greeks and the Romans carved letters in stone and these are still eminently readable over 2,000 years later.

Vellum and parchment manuscripts dating to 4,400 years ago still exist, albeit in fragmentary form. On the other hand, illuminated manuscripts on parchment or vellum dating from 1000 A.D. are still magnificent in appearance and eminently readable if one is familiar with the Latin or Greek of the period and the stylized penmanship of the age.

In art galleries and museums, we enjoy paintings dating from the 15th century and frescoes from even earlier times. We find Chinese block printing on paper from the 8th century, 1,200 years ago. The rag paper used before the 19th century leaves us with books that are still well preserved. We even have photographs on glass plates or on tin that date to the 1800s.

Perhaps by now you are noticing a trend in the narrative. As we move toward the present, the media of our expression seems to have decreasing longevity. Of course, newer media have not been around as long as the older ones so their longevity has not been demonstrated but I think it is arguable that the more recent media do not have the resilience of stone or baked clay. Modern photographs may not last more than 150200 years before they fade or disintegrate. Modern books, unless archival paper is used, may not last more than 100 years.

I have written more than once in this column about my concerns for the longevity of digital media and our ability to correctly interpret digital content, absent the software that produced it. I won't repeat these arguments here, but a recent experience produced a kind of cognitive dissonance for me on this topic. I had gone to my library of science fiction paperbacks and pulled out a copy of Robert Heinlein's Double Star that I had purchased about 50 years ago for 35 cents. I tried to read it, but out of fear for breaking the binding, and noting the font was pretty small, I turned to the Kindle library and downloaded a copy for $6.99, or something like that, and read the book on my laptop with a font size that didn't require glasses! So, despite having carefully kept the original paperback, I found myself resorting to an online copy for convenience and feeling lucky it was obtainable.

It seems inescapable that our society will need to find its own formula for underwriting the cost of preserving knowledge in media that will have some permanence.

This experience set me to thinking again about the ephemeral nature of our artifacts and the possibility that the centuries well before ours will be better known than ours will be unless we are persistent about preserving digital content. The earlier media seem to have a kind of timeless longevity while modern media from the 1800s forward seem to have shrinking lifetimes. Just as the monks and Muslims of the Middle Ages preserved content by copying into new media, won't we need to do the same for our modern content?

These thoughts immediately raise the question of financial support for such work. In the past, there were patrons and the religious orders of the Catholic Church as well as the centers of Islamic science and learning that underwrote the cost of such preservation. It seems inescapable that our society will need to find its own formula for underwriting the cost of preserving knowledge in media that will have some permanence. That many of the digital objects to be preserved will require executable software for their rendering is also inescapable. Unless we face this challenge in a direct way, the truly impressive knowledge we have collectively produced in the past 100 years or so may simply evaporate with time.

Back to Top


Vinton G. Cerf is vice president and Chief Internet Evangelist at Google. He served as ACM president from 2012-2014.

Copyright held by author.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2016 ACM, Inc.


P J Narayanan

This is indeed an important issue. However, the volumes make everything difficult.

In the old times, production in contemporary media was expensive, difficult, and hence the volume was low. This also translated to only truly valuable information being stored in the media. Conversely, everything put into the media became valuable. The digital media today is cheap, indiscriminate, and voluminous. This automatically reduces its value!

We can only create alternate (and perhaps expensive) channels to preserve information and artifacts that are truly valuable. The medium of choice may well be stones, baked clay, or copper plates!

P J Narayanan
IIIT, Hyderabad

Bill Curtis

A very important issue. The motion picture and television industries have lost untold numbers of movies and shows due to the decay or loss of old film. There are efforts to preserve the content of old films, but they struggle to keep up with rate at which this visual trove is decaying. It's the essence of preserving our art, culture, and history. While there are gaziggabytes of electronic information to preserve, advancing text and pictographic analysis technologies will make them accessible to future genmerations.

Richard Gabriel

Long Tien Nguyen and Alan Kay had a paper in Onward! 2015 about an approach to this problem - The Cuneiform Tablets of 2015. It is in the Digital Library ( and a preprint is on the Viewpoints Research Institute site (

Richard P. Gabriel
IBM Research

Warren Scheinin

Like the White Album, as noted in Men in Black, if its important, it will be transcribed to the next generation of popular media. Do we really need Napiers Bones or a working slide rule? Have you seen a copy of the CRC Standard Mathematical Tables and Formulae recently, much less used one?

Yes, I have containers full of 8 floppies and towers of 3 diskettes I can no longer access (as is the case with most 8 track tape owners), but then there is nothing on them I dont have on my current terabyte drive. And yet, we are currently reading the contents of charred manuscripts without opening them.

So not to worry just yet. My 60 year old copy of Robert Heinlein's Double Star, albeit a bit brown around the edges, using my newly installed cataract lens, is still readable. Youre welcome to borrow it.

Joshua Muskovitz

While my instinct is to share in the lament that much of our modern media has a much shorter expected lifespan than the examples from antiquity, I must point out the obvious observer bias present. How many other types of media have been created during those ancient times that have since turned to dust? By definition, we can never know. We have no basis to assume that those forms which have survived constitute any significant portion of all media from the past.

One must wonder just how much loss has occurred since the dawn of time. We generate content at an alarmingly increasing rate. Are we losing content at an increasing rate? And is this measured in raw loss, or as a percentage?

That having been said, there are projects that exist today which at a minimum attempt to raise the level of discussion. One such example is The Long Now Foundation ( They write: "The Long Now Foundation hopes to provide a counterpoint to today's accelerating culture and help make long-term thinking more common. We hope to foster responsibility in the framework of the next 10,000 years." Whether they can succeed, only time will tell. Or not.

Displaying all 5 comments

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account
Article Contents: