That link is another example of the enormous amount of truly awful information that floats around the Internet, and allows urban legends to grow into myths.

ALL types of media can be produced by manufacturers who don't know what they are doing. Examples: videotape from the 1970s used bad glue, and the oxide ends up shedding; CDs and DVDs from Memorex and other manufacturers used horrible dye, and those discs lose their data; laserdiscs manufactured in the first few years of that technology used bad glue to bind the top and bottom half of the discs together, and they were prone to "laser rot" after just a few years. And of course movie film from certain manufacturers had two dye layers that were prone to fade, leaving only red and white.

I can go on.

The point is that each of those same media types is capable of lasting a long, long time. I have transferred movie film from the 1920s that still runs through the projector like it was developed yesterday. Same with the other media I mentioned.

As for CDs and DVDs, I have done tests, and my media is as good as the day I first burned it. That's almost twenty years ago for my oldest CDs.

Now, let's talk about solid state media. I understand the physics of how they work, and it has always disturbed me. It is inherently unstable. I have tried, on several occasions, to find accelerated aging tests for this media and so far have found nothing. By contrast, I have found almost a dozen such tests for CDs and DVDs, and the results are extremely positive and very reassuring. If you used good discs and you store them well, they will outlive you (easier said for me than for others ...).

So, don't get me wrong, I'm not trying to start my own urban legend about solid state media, and I certainly understand its convenience and capacity advantages compared to DVDs. I use them every day and am buying more all the time.

However, here's the point of the post:

I do not yet trust them for archiving, and until I see some credible testing, I am not going to rely on them for anything other than 1-3 year storage, for working backup. One of my main jobs is media restoration and archiving, so this is something I take seriously.

The ONLY time I had failed CDs was when I first began to burn them in 2000 using a brand-new "6-2-2" CD burner (6x read, 2x burn!) and no-name CD-Rs from CompUSA. Within months the aluminum began flaking off in huge chunks.

I also had failed DVD-R Ritek disks after a few years of storage about a decade ago. No flaking (couldn't happen anyway, since a DVD-R is a sandwich construction) but many disks just couldn't be read any more. But since switching to Taiyo Yuden and Verbatim years ago, the disks are very stable.

My impression is similar to John Meyers: properly stored, videotape and film are very long-lasting. I think a hard drive will be, too, assuming there's a way to play it back in the future. Flash memory? I don't know; I'd love to see some aging analysis.

Hey, I've got it! Let's archive to punch cards! Paper lasts forever! I'll leave it to someone else tonight to calculate how many boxes of punch cards it would take to archive an hour of HD video.

"videotape from the 1970s used bad glue"

It wasn't just videotape, Ampex audio tape such as 406/407, 456/457, 499 and Scotch/3M Pro tapes like 206/207, 226/227, 808, and 986 were used by pretty much everyone in the recording business and virtually NONE of those tapes are playable these days unless you "bake" them under strictly controlled conditions. The binder (glue) used in these tapes contained polyurethane that soaks up water (hydrolysis) and causes the urethane to rise to the tape's surface. The typical symptom is squealing when the tape passes the playback head or other fixed parts of a tape player. The squealing is audible directly from the tape and also transmitted electronically through the output of the tape recorder. Continuous use of a squealing tape risks permanently damaging the tape, as oxide is sometimes torn off the tape. This flaking residue can be seen and can feel gummy while still on the tape's surface. Baking the tape temporarily restores the tape by driving the water molecules from the binder so that it can be safely copied to another tape or a different format. After baking, the tape usually remains in good condition for approximately a week to maybe a month tops. I've had to bake lot of tapes. What a hassle!

(Tapes from Maxell, TDK, Quantegy and BASF did not exhibit the problem.)