# When will all the data start evaporating?

**URL:** https://boards.straightdope.com/t/when-will-all-the-data-start-evaporating/8338
**Category:** Factual Questions
**Created:** [March 8, 2000, 11:35am UTC](https://boards.straightdope.com/t/when-will-all-the-data-start-evaporating/8338 "2000-03-08T11:35:00Z")
**Posts on this page:** 8
**Page:** 1

<div class="post-metadata">

### Author: ![Soulman\_K](https://avatars.discourse-cdn.com/v4/letter/s/e47c2d/32.png) [@Soulman\_K](https://boards.straightdope.com/u/Soulman_K)
#### Post date: [March 8, 2000, 11:35am UTC](https://boards.straightdope.com/t/when-will-all-the-data-start-evaporating/8338/1 "2000-03-08T11:35:00Z")

</div>

How long will the digital media last? That is, when will all the data start evaporating? (Being into music, I’m particularly concerned about my CD’s and MiniDiscs.)

---

<div class="post-metadata">

### Author: ![JonF](https://avatars.discourse-cdn.com/v4/letter/j/cab0a1/32.png) [@JonF](https://boards.straightdope.com/u/JonF)
#### Post date: [March 8, 2000, 1:51pm UTC](https://boards.straightdope.com/t/when-will-all-the-data-start-evaporating/8338/2 "2000-03-08T13:51:00Z")

</div>

It depends.

There really isn’t enough data to support a definitive answer. We won’t know until a significant number of failures have happened under normal storage conditions. Environment also makes a big difference.

But there are lots of people worrying about this, as a web search will quickly show. There’s an interesting group that’s mentioned in this month’s [Dr. Dobb’s Journal](http://www.ddj.com), but I can’t dig up their URL, the magazine is at home, and the DDJ web site is responding very slowly now.

People have run accelerated life tests. According to [How Long do Digital Storage Materials Last?](http://www.safesupplies.com/article1.html), at 20 C (68 F) and 40% relative humidity, stamped CDs should definitely last 5 years, have a pretty good chance of lasting 50 years, and may last longer. This is considering _any_ data loss as a failure; audio CDs can lose lots of bits and still work fine.

Of course, this is all about as accurate as the MPG figures you see in a car dealership.

* * *

jrf

---

<div class="post-metadata">

### Author: ![scr4](https://avatars.discourse-cdn.com/v4/letter/s/59ef9b/32.png) [@scr4](https://boards.straightdope.com/u/scr4)
#### Post date: [March 8, 2000, 2:22pm UTC](https://boards.straightdope.com/t/when-will-all-the-data-start-evaporating/8338/3 "2000-03-08T14:22:00Z")

</div>

There is also a good article at [http://www.wired.com/wired/archive/6.09/saved\_pr.html](http://www.wired.com/wired/archive/6.09/saved_pr.html) - maybe not much hard data, but interesting.

---

<div class="post-metadata">

### Author: ![Satan](https://avatars.discourse-cdn.com/v4/letter/s/e9bcb4/32.png) [@Satan](https://boards.straightdope.com/u/Satan)
#### Post date: [March 8, 2000, 4:53pm UTC](https://boards.straightdope.com/t/when-will-all-the-data-start-evaporating/8338/4 "2000-03-08T16:53:00Z")

</div>

> [@](#):
>
> Originally posted by JonF:
> 
> **There really isn’t enough data to support a definitive answer.**

YOU MEAN IT’S STARTED ALREADY?!?!?

* * *

Yer pal,  
Satan

---

<div class="post-metadata">

### Author: ![pluto](https://avatars.discourse-cdn.com/v4/letter/p/f08c70/32.png) [@pluto](https://boards.straightdope.com/u/pluto)
#### Post date: [March 8, 2000, 10:26pm UTC](https://boards.straightdope.com/t/when-will-all-the-data-start-evaporating/8338/5 "2000-03-08T22:26:00Z")

</div>

> [@](#):
>
> Originally posted by Soulman\_K:  
> \*\*How long will the digital media last? That is, when will all the data start evaporating? \*\*

Those are really two different questions. The odds are the data will still be in place long after the medium has become unusable because it’s been replaced by something new. I still own several vinyl records that I presume are still more or less intact, but I no longer own a turntable. I have a book on my desk with some computer code on 5-1/4" disks that I cannot access.

In these instances it is still possible to retrieve the data – someone, somewhere has a turntable and a 5-14" disk drive. But there are instances where the technology simply no longer exists. I’ve worked on programs where that was very nearly the case and we were nursing the last remaining devices that could read the medium we were using. I have heard (UL warning!) that a lot of space data that NASA collected years ago is in this condition: either the reading device no longer exists or the software to interpret the formatting used is gone.

* * *

I have the body of a god – Buddha!

---

<div class="post-metadata">

### Author: ![Oblio](https://avatars.discourse-cdn.com/v4/letter/o/c4cdca/32.png) [@Oblio](https://boards.straightdope.com/u/Oblio)
#### Post date: [March 9, 2000, 1:17am UTC](https://boards.straightdope.com/t/when-will-all-the-data-start-evaporating/8338/6 "2000-03-09T01:17:00Z")

</div>

Several times a week we gingerly load an 8" floppy into our radar target simulator, hoping that it boots up. One day it will not boot, hopefully we will have the replacement system completed by then.

I recall hearing/seeing (sorry, no cite) about a company that addressed the issue of data permanence and provided the technology to etch data onto a metal medium. The medium was chosen for resistance to corrosion and general stability over millenia. I think the data was recorded somewhat like microfiche at a fairly high density with a lifetime many magnitudes greater than more traditional means (CD,tape,papyrus,post-it notes etc). The issue of who could interpret the ‘hieroglyphics’ after untold eons buried in post apocalyptic ruins is another issue.

* * *

A point in every direction is like no point at all

---

<div class="post-metadata">

### Author: ![Ringo](https://avatars.discourse-cdn.com/v4/letter/r/779978/32.png) [@Ringo](https://boards.straightdope.com/u/Ringo)
#### Post date: [March 9, 2000, 3:10am UTC](https://boards.straightdope.com/t/when-will-all-the-data-start-evaporating/8338/7 "2000-03-09T03:10:00Z")

</div>

A couple of thoughts:

1. The lifetimes of media are estimates for most of the market life of any medium; afterall, when you see a hard drive advertisement with a “spec” of 150,000 hours MTBF (mean time between failure) you know they didn’t test the things for 17 years - they used some accelerated wear tests and some estimates. And considering all the hard drives that _have_ failed, at least a few are going to have to last a century or more to validate the methods used - I think some just might.

2. The life of a medium - not just the cartridges, discs, tapes or whatever, but the standard itself - faces an increasingly tough environment with the quick pace of change. A few dinosaurs of relatively recent make have been mentioned. I’m starting a new job and gave my workstation specs to the new employer, tomorrow I’ll talk to their IT people about procuring it and I just _know_ they’re going to have a better idea than using an 8 mm tape drive. There _are_ better options for tape backup, but my industry is married to 8 mm - that’s what all the data is on.

Data transcription has birthed quite a few small businesses - hmmmm, giving me ideas…

---

<div class="post-metadata">

### Author: ![JonF](https://avatars.discourse-cdn.com/v4/letter/j/cab0a1/32.png) [@JonF](https://boards.straightdope.com/u/JonF)
#### Post date: [March 9, 2000, 11:57am UTC](https://boards.straightdope.com/t/when-will-all-the-data-start-evaporating/8338/8 "2000-03-09T11:57:00Z")

</div>

> [@](#):
>
> The odds are the data will still be in place long after the medium has become unusable because it’s been replaced by something new…- someone, somewhere has a turntable and a 5-14" disk drive. But there are instances where the technology simply no longer exists

[The Dead media Project](http://griffin.multimedia.edu/~deadmedia/frame.html)

* * *

jrf
