http://www.anatechusa.com/thin_film_coating/default.html
http://www.circletrack.com/enginetech/ctrp_0408_casidiam_coating/index.html

http://www.faqs.org/patents/app/20080291570

Larry

With the volume of tape you have to transcribe, you may see some wear on your heads. That might vary the calibration (or worst case wear the heads down to where they need replacing.)

Considering how few parts you have, may I suggest considering having the heads coated by a very thin diamond layer. They might last 400 times longer.

See: Wear-resistant, diamond-like coating created by Sandia 1998
e.g. See: Diamond Tools

Diamond Coating

The hardest, strongest, and slickest is known as tetrahedral amorphous carbon, or ta-C. For example a coating of only 2 μm thickness of ta-C increases the resistance of common (ie. type 304) stainless steel against abrasive wear; changing its lifetime in such service from one week to 85 years.

Thanks

It would be possible to do what you are talking about but I would have to call the suicide prevention line shortly thereafter. There is one thing about 1960′s era equipment that most people don’t realize today. It was incredibly touchy and it required trained field engineers to maintain. My first career was as a Field Engineer (FE) in the late 1970′s. My company build a large impact line printer, card punches, and card readers, interfaced to IBM, Amdahl, and Burrough’s computers.

For a medium sized location such as a bank data processing center, you probably had 15 guys that just baby sat the hard drives, tape drives, printers, card punches, card readers, keypunch machines, the CPU, and the comms system to interface to the big beasties. ALL of the guys that did this stuff had to have months worth of training (I spent three and a half months in school 8 hours per day to learn my hardware) where you lived with the equipment, were trained by the people that built the equipment or the training department where you were intensively fed every detail in the books, and THEN you had practical troubleshooting classes where you learned the quirks of your hardware, what had a tendency to fail, and how to fix them.

Could you imagine in 50 years from now, when we all have passed from the scene, trying to get a mainframe installation like I described up and running again? Oh and by the way, there is the software to make it run in the manner that it was designed to do.

I don’t think that people in general have a proper appreciation of just how complex that technology was and is and how far we have come in terms of usability and portability. Even with this, could you imagine trying to bring a laptop from today back to operational status in 50 years. Hell most people toss their systems in 5 years. I have a Mac Pismo laptop that is nine years old and I have gone through quite a lot of work to keep it operational.

So, no while doing something like you suggest is possible, the costs would far outweigh any benefit of doing so.

Great post and I agree. A mantra that I keep saying is that it does not matter if you have a 100 year storage media if you do not have a 100 year means of reading that information from the storage media.

I would add that for sites like Anthony’s or Steve M’s, these will become a valuable historical research asset at some time in the future and just how many decades will we be able to read wordpress threads? Just think about how many articles and comments that exist already here and this is just as valuable as any other information source.

(One more thing for you to worry about Anthony)

:)

http://www.archives.gov/electronic_records_archives/index.html

The majority of the digital information we have available today will evaporate in useful terms in 15-30 years from now. Most of what you know will be unavailable to your children and grandchildren unless everyone makes an effort to preserve the essential data from the late 20th century and the early 21′st century.

Do you have essential data records in your safety deposit box?

Larry

“crosspatch (20:45:45) :

This is why I am a fan of archiving critical data in a mechanical fashion. Even a standard vinyl LP can be played back using a dixie cup and a straight pin … though not may times and not with great fidelity.

I believe our era will be looked at a thousand years from now as a dark age from which little information survives.

This is a problem across all types of data storage. The National Archives and NIST have been interested in this for some time. (by the way has anyone on this team contacted NIST to see if they have any equipment that they used in research that might assist in this project).

About 10 years ago I worked in an IT company that had a 400,000 cartridge tape library and a few thousand reel to reel tapes for computer data. I was asked to figure out a safe retention time for tapes and how often they should be read into memory and re-recorded on new media to preserve data that had records preservation limits. I contacted the folks at NIST and the National Archives on this issue, along with some of the manufactures. The short answer is no one knows how long digital magnetic tape will last. The manufactures would only specify the usability of the tape in read write cycles and would not even venture a guess on vault storage life.

In the case of the reel to reel tapes if they were wound with too much tension over time the tape would shrink and “cinch” forming buckles in the tape spool as the material slowly tried to adjust to the tension.

The gold substrate archival CD’s are probably good for decades if stored carefully and only read when necessary. One of the most reliable storage devices I could find was actually the simple flash memory chip. It can hold the charges used for data storage for up to 100 years according to the info I found, and will tolerate abuse that would destroy other storage media.

In one experiment they baked a flash drive into a custard pie and it worked just fine when removed and cleaned up!

That said you have two storage issue both the media and the methods and equipment to recover the data, both software and hardware.

How many of you can recover data from a wps document, even if you can read the floppy disk it is stored on?

You also need to consider the data configuration, for example on images jpg, and gif are two of the best supported formats and will probably persist for many decades. Even if they are not the format is simple enough and well documented enough (right now) that one off code could be written to recover the image if necessary. More troublesome is things like interface connections between hardware devices. How many of you still have a floppy drive available to you? How many of you can plug in a serial interface to your computer?

Right now you can still find adapters USB to serial but even the USB interface has morphed into multiple versions. If you work with critical data, please consider storing both the data on archival media and the means to recover it, both hardware and software.

This is going to be a big problem for libraries in the near future and there are studies on the topic if you look for them.

http://www.itl.nist.gov/iad/894.05/gipwog/Oct-14-04.html

In the case of critical data, store it in several different media formats if possible to ensure it will be readable by some future system. For example store an important text document as .doc .rtf and .odt or pdf formats would greatly increase the odds of some future system being able to read at least one file.

In the case of spread sheets the comma delimited file could be saved as a backup to the normal spreadsheet format you normally work in.

In the case of images, jpg, gif, and tiff files are well supported file formats and not likely to disappear any time soon as they do their job very well and are widely supported in multiple software systems.

In the interest of history about the AGW debate, I would encourage Anthony to look into an archival backup of this forum, and donation of a copy to a large library, as it will 50 years from now, be a priceless archive of the mood of the time and the minutia of the debate.

Larry

http://www.realclimate.org/index.php/archives/2009/04/farewell-to-our-readers/

That looks like parody of Monckton’s style. IMO they’ve put that article in on the Precautionary Principle – might need it anyway so let’s just take precautions! :)

3×2 (15:52:57) :

Just thinking that my N95 could have formed the core of a 60’s satellite. I was reading this thread while eyeing up 250 MSPS A/D converters over at TI.

As one finds out in short order (doing new prod dev) the trick is getting the processing organized downstream of those fast TI A/Ds (think FPGAs b/c you’ve got no time for a standard ‘processor’, or even a DSP at that kind of raw rate!)

Reply: I’ll take that as a go to give him your email address. ~ charles the moderator

A walk along memory lane on the techniques used in these old beasts can be found here:
http://blog.modernmechanix.com/2008/12/13/a-portable-color-recorder/?Qwd=./RadioElectronics/3-1967/vtr&Qif=vtr_1.jpg&Qiv=thumbs&Qis=XL#qdig

For those wishing to overdose I recommend: http://www.labguysworld.com/Cat_Ampex.htm

PS. LabVIEW and a fast A/D card (maybe used in a full PXI chassis w/the appropriately fast controller) would make a more than suitable ‘demodulator’ if the support/creation of the real hardware proves too time consuming; LabVIEW Signal Processing Toolkit might also prove useful (full disclosure: not a salesman, just a user of said product). Google: LabVIEW demodulation http://www.google.com/search?sourceid=navclient&ie=UTF-8&rlz=1T4GPEA_enUS309US309&q=LabVIEW+demodulation+

The largest problem is the software. The image is 37,134×16,200 at 16 bit greyscale, a tad larger than most home monitors. The earth part is quite small, ~250M using the same .tiff

The National Gallery / Southampton University / Imperial College have created something called VIPS (or NIP2) which is designed to handle large images (such as digitised old masters), larger than available memory, can even work over networks. (yes I have actually loaded the image, best bet is crop out the earth part and save to file, then work on that)

It can do a lot but is not like Photoshop. Been described as a spreadsheet for images. Can do some very unusual things, if you can work out how to drive it.
“The GUI aims to be about half-way between Photoshop and Excel. It is very bad at retouching photographs, but very handy for the many other imaging tasks that programs like Photoshop get used for.”

Cross platform, Unix, Mac, XP. Don’t know if Vista is ok and the Windows version is less stable. Very useful piece of software to know about.

http://www.vips.ecs.soton.ac.uk/index.php?title=VIPS

Be warned, you might be disappointed, is not a casual image.

Reply: Gimp opens the image just fine although it does have to convert the image from 16 bits per channel to 8 bits per channel which, except for the most serious scientist should not be a problem. Open source and free, and it has a much more Photoshop like interface. It does take a minute or so to load the image on my desktop, a dual core XP machine with 3Gb RAM. ~ charles the moderator.