Slashdot Mirror
Problems in Computer Conservation
sobachatina writes "The Computer museum at The University of Amsterdam has an interesting page with examples of the problems that they run into maintaining 20+ year old hardware such as rubber rollers from card readers melting or mold growing inside CRT terminals.I hate it when I get mold growing inside my monitor!"
for going solid state all the way.
Suck the air out of the exhibits and you'd probably be able to preserve those exhibits a little bit longer.
And if all else fails, take a picture and put it up when the original machine has fallen to pieces.
I have been pwned because my
Just a few years ago. Not because It didn't do what I wanted it to. In fact I rather miss it because it was ideally suited to its task, but because various little mechanical bits of it started to get wonky and I couldn't find replacements.
I'll be able to custom build a replacement now with the new VIA stuff, and the replacement will undoubtably be "better" than the Compaq, but it's still just plain annoying to have to take a grand or so out of pocket to replace something that did it's job ( and that I only payed $50 for in the first place) and could have continued to do so ad infinitum had a few $5 parts been available.
And of course its basically working carcass is now sitting in some landfill because none of the local shops even considered it worth taking up space if I gave it to them for free.
And this could still be a continuing issue. One of the surest ways to force DRM "enabled" machines on the majority of the populace is to simply phase out the bits of the machines people already have making them impossible to keep going.
It might take 20 years, but businesses seem to find the patience they otherwise lack when it comes to ways to grind down the consumer to the level they desire.
KFG
Well, "paper" has changed over the years. old scrolls, manuscripts, and books are on thin animal skin (vellum) and can last 2,000+ years under the right conditions....the wood based stuff we usually call paper oxidizes, turning dark and crumbling in less than 100 years unless special preservation steps are taken. For paper made of cloth rag, you can get 100-200 years (paper until the 19th century was made this way). Some combinations of inkjet/laser toner and rag "paper" can last 140+ years, it's claimed.
Okay, we WILL be leaving behind mountains of trash that future cultures will probably be mining for raw materials.
"Learning is not compulsory... neither is survival."
--Dr.W.Edwards Deming
CD-Rs use an organic dye which reacts to the recording laser in a CD-R burner, causing it to melt and pit--it then becomes non-translucent and the reading laser is refracted when it attempts to read those portions of the disc.
:P
There are several known cases of bacteria and fungus attacking this organic dye, not including the obvious danger that heat and sunlight pose to it.
"Regular" CDs use a polycarbonate substrate instead which is literally stamped into the CD during an injection moulding process. THIS is a mechanical change, giving the advantage that a stamped CD could very well last 50-200 years, whereas a burned CD-R that is not hermetically sealed will be lucky to last 10-20.
It seems that what we need is an inorganic hybrid of a stamping machine and a CD-R burner, something that can (using a much more powerful laser) physically inscribe the bits into a polycarbonate-like material. The nice thing about a technology like this being adopted, is that the firmware could be modified to allow the same machine to create CDs, DVDs, and whatever else they throw at us within that form-factor. Even better would be the ability to come up with your own (Open Source?) disc data storage format.
Anyone want to play devil's advocate on that idea? Apart from cost, I could see consumables and waste being an issue.
Jason Fisher
When I was a kid I found my dad's old (Texas Instruments, I think) calculator. It had a one line red VFD display and was almost the size of a brick (it weighed sligtly less). I also found a little book of magnetic strips (about the size of a stick of gum). The calculator had a little slot in the left and right sides. You could load "programs" into the calculator by inserting the strip in one side of the calculator and a little motor with a rubber wheel attached would pull it through and spit out the other side. If you were lucky, it had read the strip right and loaded your program.
Well, after about a dozen of these loads the little rubber wheel attached to the motor fell off, no more contact with the strips. I tried tons of things, tape, pieces of plastic, shaved down washers, nothing worked. I was quite sad when I had to throw it away, it was a fun little toy. I'm sure that now, being older, I could have fixed it. Maybe if I had kept it and fixed it I could still be using it to this day.
These things wear out and break down but I think if you have enough time, money and resources you could probably keep them going forever.. But, is it worth it? For me it would have been... for the sentimental value.
Geoffeg
Consider this: In my business, people are using machines as old as 30 years and some may be quite a bit older than those. These machines have everything from punch card readers to tape readers to special floppy drives that are impossible to find nowadays. And of course, these parts go bad, and as luck would have it in this industry, replacing one of these pieces of machinery can completely break a business. Especially with the economy doing as badly as it is now, and manufacturing is at such a low that everybody in this industry is suffering. But I digress.
The point I'm trying to make is this: If these things (CRTs with mold, rubber wheels melting, etc) are critical to the operation of a really old computer, then someone needs to manufacture them, just like people still manufacture replacement parts for old cars. This is most likely a better idea than replacing these systems with new ones for the following reason: These old systems are proven. A lot was invested into making them reliable and whatever bugs exist are well known by now. Replacing these systems would introduce problems for a long time to come... problems like software not working properly, which is a problem that management has a very difficult time accepting. Try telling your boss that some buffer wasn't flushed and therefore $50,000 just went down the drain. A rubber wheel melting and being replaced is a lot easier to explain to one's manager because everybody knows what a rubber wheel is. And how much does a rubber wheel cost? Even if it has to be specially manufactured and costs the end user $100.00, that's a hell of a lot cheaper than re-engineering the whole damn computer network. And putting up with stupid management (of which I am a member) giving you shit because three months have passed and the new computer network STILL isn't operational due to some stupid SQL program or perl script that has yet to be written, and we've gone ahead and ordered that rubber wheel anyway.
Negra Modelo. Me llamo Juanito Rodriguez y soy alcoholico.