Is the web permanent? Ones and zeros are a great way to archive massive amounts of data, but what’s being saved and what’s being left on the curb? And how long is permanent in an environment where one after another, advancing technologies replace obsolete ones?

Recently I heard about a remarkable undertaking on the web. Founding member of the Irish rock band The Pogues, Jem Finer, a musican with a degree in computer science has created an interesting work. “Longplayer” (http://www.longplayer.org) is a piece of music that’s playing on the web as you read this. Using four separate musical compositions and a computer algorithm, the piece began playing on January 1, 2000 and will continue playing, without any repetition until December 31, 2999.

“Longplayer” is a melding of technology and art, but more than that it’s a remarkable conceptual piece that does more to raise interesting questions about time, art and technology than it does to trigger an aesthetic experience. There is little chance that the computers that power the web, or the web itself will exist in anything like the form in which we’re familiar with in 998 years.

The vast majority of people that will have a chance to hear some of this piece have yet to be born, the computer it’s on will fail and the web will have morphed a hundred times over before it ends. A trust has been established to maintain “Longplayer” through all of these changes.

A web site I founded and manage recently re-launched with its fourth complete design revision. Of the five distinct designs the site has had over the last seven years, only the current design and the previous one survive. I have archived version 4.0 but the first three are lost to the bit bucket and I regret it.

Not that I’m in love with that earlier work, it’s just that a record has been lost. It’s history, my history and a tiny piece of what the web looked like in 1995, 1997 and 1999.

I’m not alone in this shortsightedness. There are levels of irony at play here. If you want to look up an old Journal-World story, we have an archive for that. But if you want to know what the online presentation of our content looked like in years past, you’re out of luck.

You can go to the library or other places to find out what a printed copy of the paper looked like from 1996. This sort of access to the past is relevant to journalists, historians, scholars, designers and nostalgia buffs. By erasing our past, the web builders of the present fail to fully serve the future.

I’m not the first person to worry about this. There are efforts afoot to archive the Web. Internet Archive maintains a web archive at http://web.archive.org/. They call it the Wayback Machine. Unlike Sherman and Peobody’s contraption where they’d give it a date and arrive at an unknown location, here you enter a destination URL and it offers you dates to visit. Going back to see what my site used to look like was fun.

Unfortunately many sites thwart their efforts by blocking the “robots” that collect and archive the sites. Looking up http://www.nando.com, the site for the Raleigh News and Observer, one of the nations first online newspapers, proved useless. It’s a crime that what Nando looked like in 1994 is lost to us.