An Australian scientist hopes to restore a vintage, refrigerator-sized IBM tape drive stored in a museum to recover Apollo moon mission data the space agency misplaced nearly 40f years ago. NASA's only means of measuring moon dust during its Apollo missions has gone largely unappreciated until recently, reports Australia's ABC …
Hey, I still have
an 8" floppy disk with CP/M 2.0 on it, could they read that for me too?
Bruce O'Bruce surely
Surely a thorough magnetic scan of the tape, by say a large tape manufactures labs, Then digitally reconstruct the data using an emulator, would be a better and safer approach?
The design manual/circuit diagram for the drive would help when constructing an emulator...
Did anyone ask IBM?
Here's a strange question, did anyone think to contact IBM? Surely they would have kept the blueprints/tech specs of the old drive.
If anyone could afford to rebuild an old tape drive from scratch you would have thought NASA could.
Still perhaps that's the next option.
Can you imagine in 5 years time that will be illegal
- DRM restrictions etc!
Way to go!
I hope these guys are sucessful. Really interesting story!
Talk about the long way round
so they are trying to resurrect an old tape drive.
that would be fine if Itsy Bitsy Micro's had gone bust 20 years ago.
Surely they still know how to build a new one of these, and considering the importance of the data , a new clean drive , would be less destructive than an old 'cleaned up' drive!
I would have thought the most difficult challenge would be to get the software to read it, thus an old computer might be needed RS232'd into a nice new laptop! :)
That is hardly necessary
Having worked in the IT industry there is bound to be some one somewhere still using one daily.
Maybe constructing a new machine would be easier than trying to resurrect an old one? (I'm not sure I fully understand the problem though, surely it's just a case of copying the magnetic data to a new location)
Problems finding a tape drive?
Why can't they build another one? Surely the designs exist elsewhere?
What happens if it chews the tapes the moment they are put in?
I have nightmares like that.
Sometimes old hardware should be left to rest in peace.
Therein lies a lesson for us all
You can't beat good old paper when it comes to archival storage!
No mention of boffins, no puns on dusty tapes of dusty dust data, no alitteration.
Standards falling. Disappointed.
And there's me thinking I was clever the other week, loading up a few games from my Spectrum 48k.
Is it really so hard
to recreate the hardware from scratch? The engineering drawings must still exist, or at very least the data storage specification. we hear this kind of thing all the time - the BBC Doomsday project, this kind of tape archive, older picture formats. Surely, since the archives are on magnetic tape, as long as one has a read head wide enough (or in the case of the Doomsday, a laser travel assembly long enough), the rest is in the software?
By careful deduction...
inference, and application of reason, the contents of the tape can be determined without further processing.
Backtracking Moores Law, factoring the exponential growth in data storage density, gleaning parity information stored on other contemporary media, and deep analysis of the Goodies episode, "Frankenfido", suggests within an acceptable margin of error that the tape contains:
Being a sign-bit, its significance lies in the binary negation it applies to all data that follows. It turns out the cabin dwelling bearded loons were right. The Moon mission - at least that part of it that followed the first 126 seconds after launch - never actually happened.
Its not all beer and skittles for the luddite paranoiacs, though. Rather than an elaborate government sponsored cold war hoax, reapplying the long lost "1" to the mission data reveals Apollo 11 to be a rather successful Mars round-trip with alien rendezvous and subsequent technology transfer.
So, hairy conspiracy freaks, you were right. But not completely right. Certainly not right enough to be employable or attractive to the opposite sex. Keep on cobbling together a romantic life from those bedtime hugs from mum and the occasional alien probing. Don't get too attached to the probing, though. Aliens have standards too.
NASA--Needs Australian Science Always
I think I'll chuck "The Dish" into the DVD player.
Goes to show
There was quite a heated discussion, at a not-to-be-named forum of Govt IT directors I was at, on this subject (getting old data, not moon dust although given teh attendees you could be forgiven for making that assumption).
Anyway, Los Bureaucraticos want data kept for x years depending upon what it is classified as. Many classifications must be kept for a minimum of 25 years.
Many of us pointed out that we can do this, but we had serious questions sucj as (a) who is going to be rebuilding the 'ancient' tape drives of the mid-2000's in 2030 and will there be software that 'understands' the files being restored, and (b) what guarantees are there that the media will survive in a recoverable state for that long. Oh yes, there was also (c) who pays, but the answer is always the taxpayer, somewhere along the line.
Unlike NASA moion dust data, which may well yield some useful information, a great deal of what we are told *must* be kept is absolute tosh that has bugger all use today, let alone in 25 years time. Like the contents of this post, for example.
Might shut the conspiracy theorists up
Nobody, not even NASA, is going to go to the trouble of faking a largely corrupt 40 year old tape backup.
Conspiracy theorists, get your coats
Doesn't this rather rain on your parade?
All that money - so little care
What are NASA playing at? They spend so much money getting to the moon and then they ditch the original tapes and loose the archive (ie the disk copy) of the data. This is appaling.
It would be bad if this was the only time this has happened - but its not. It seems that NASA has a history of just chucking old tapes away (because, you know, the US is full and there is no space for a tape store). They threw out a lot of the old data tapes from Pioneer 10 and 11. Just chucked the tapes, which are irreplacable, in the bin. Now, with the strange effects that are being witnessed due to the Pioneer Anomaly, there is a push to re-evaluate the data and see if our understanding of gravity is all we think it is, and hey - no tapes!
So, NASA - score EPIC FAIL.
Digital "Archives" just don't work.
Another example of digital archiving that becomes very hard to access when technology surpasses the storage medium. Remember the BBC project back in the 80s to recreate teh Doomsday Book? it was stored on some laserdisks that can only be read now by a couple of machines in museums and the effort required to update that material is deemed to expensive to bother with.
Just imagine what it will be like in another 50 years. All those Video tapes, CDs + DVDs with your family photos will be unreadable. I've still got "old-fashioned" photographs of my great great grandparents from over 100 years ago. I've got loads of photos from my childhood from 40+ years ago! How many of our children will be able to say the same in 40 years!
"Sorry? CDs? Oh we don't read those old things anymore... you'll need a specialist data recovery centre that will cost you squillions of Yuan to get your pictures back!"
I always heard
Given enough time, the Earth's magnetic field eventually rendered any sort of magnetic tape-based storage unusable. Even if they get the reader working, is it really going to be able to just pull all the data back off again as if 40 years hadn't passed?
729s were fun
With the dial on the upper left where you could change the drive number. There was a switch on the door that could take the drive right out of ready if the springs holding the door up weren't up to the job (and that switch was often bypassed because of this problem). you loaded the tape up on the left spindle with a locking hub and threaded it down and under the head area and up onto the right takeup spool and then "popped" the top left two buttons to take it to load point and make it ready...watched it suck the tape down into the columns and away you went...
What a beast...
Had strings of them on a 7080 (same basic series as the 70 and 90 machines) with a massive 80k of memory transisters that were each about the size of your pinky fingernail. If you knew how, you could actually program right into memory from the control keyboard on the console.
Did anyone ask IBM?
IBM does have a corporate archivist (I've worked with him before on a similar old-technology resurrection project.). However, I'm not sure how much this might cost, or whether documentation that old even still exists. But, it might be worth one of the project officials contacting the company.
Note that 7xx technology is generally tube type equipment (I have a couple of modules out of a 704 system, and they have tubes in them!).
Don't forget, of course, that, in addition to getting the 729 operational, they'll need a control unit to attach it to, and an appropriate host on the other side of that to drive it (S/360?).
One caution, though, is that really old magnetic tape, even when kept in good environmental conditions, tends to have the binder that holds the magnetic material (Iron Oxide) to the tape deteriorated. Thus, you sometimes only get one shot to read the data, since the process of running the tape over the read head tends to scrape the magnetic material off of the tape!
Here's some information about a 729 restoration project that may be of use to the project:
Can't they just ...
.... go back to the set where they faked the landings and re-do it?
Years in hex??
"... nearly 40f years ago ..." - wow! 1039 years ago!!
Just goes to show the absolute necessity of testing that you can recover data from an archive every now and again, and of keeping data on media that is both within its specified life-time, and on a system that is still relevant.
"It's going to have to be a custom job to get it working again. It's certainly not simple, there's a lot of circuitry in there, it's old, it's not as clean as it should be, and there's a lot of work to do," said Guy Holmes of SpectrumData."
*Sucks through teeth* ... Yeah, I think we can do it, but it's gonna cost ya!
Enter contract to keep 10 people fully employed on a high wage for 5 years to play with old kit.
No, it’s not hard to build new hardware.
The engineering drawings do still exist; they are safely backed up on an IBM 729 Mark V tape.
I went for a joke alert; I was going to Paris, 'cos she knows all about hard ware ;-)
An excellent demonstration of why the world cannot afford proprietary formats or standards.
Seems like an extreme solution
There were a number of drives that read 7-track 200, 556 and 800 bpi IBM tapes. The 729 had fairly agressive tape handling with vacuum column tape loops and servo motors that would start and stop tape movement very rapidly.
For reading old tapes it might be better to read at constant speed using a gentler drive mechanism, and simply digitize the read amp output as the tape runs past the head. After that, its all digital signal processing.
but I imagine there's a Linux driver already written for the drive ...
To all the Doomsday comments - it's actually Domesday.
The main reason you haven't seen an online version of it is licensing issues. Because no one thought of asking the metadata creators - schools and pupils - if they minded if their work was published on the Internet. Not surprising really as we weren't even at Web 0.1 never mind 1.0 in those dark days.
However, the BBC Domesday project was preserved, see this: http://www.ariadne.ac.uk/issue36/tna/
Mine's the one with the Philips LV-ROM player remote in the pocket...
@ Stuart Gray
Asked he: "Is it really so hard to recreate the hardware from scratch?"
Quite possibly there are proprietary details to the construction that were only documented on a need-to-know basis and have been lost to retirement and death.
In situations like this, it helps to be a Buddhist and to thereby understand that change is always just around the corner.
I used to smile at the dimwit managers I worked with who'd read "Paperless Offices for Dummies" and thought they understood the issues. 'Twas amusing to ask them how could we read the big floppies that stored ancient files from a Xerox word processor back around 1980. Or the lovingly preserved tape backups of IMS databases dating from 1985, where the program source code had long since been junked, and even if still extant, could no longer be compiled.
There really is data that demands long-term retention. Military pension data is one, because sometimes ancient soldiers marry sweet young things who become entitled to a widow's pension. The US Civil War ended in 1865, but some participants (drummer boys, e.g.) were quite young, say 15 y.o. at the time. Thus a geezer born in 1850 could in 1930 (aet. 80) marry a 20 y.o. born in 1910 who today is a mere 98 y.o. and still receiving a pension check every month.
It's really quite surprising how quickly information becomes unusable. I have what purports to be a list of the addresses my family lived at from 1936 on, but between garbling of the information, the building of freeways and interstate highways, changes in street names, renumberings of street addresses, and alteration in city boundaries, I have been unable to pin down several of these addresses using Google maps.
Such is life.
@Years in hex??
And here I was thinking I was the only one who noticed that.
/Mines the one with the backup tapes in the pocket
What's the chance...
... that the data they get off it will probably fit on a 3.5" floppy?
Old data... real problem.
I wonder how many people know about the risks of long-term data storage. During my lifetime, I've seen lots of media turn antique:
- My dad's programs are stuck in a backup tape reel that can't be read
- I lost most of my Mac files on both my Jasmine Removable 45 and MDS88 drives
- My dad's Iomega Jaz cartridges are dead
- And well, not only am I deprived of a 5.25" floppy disk reader, but even if I had one, I'm pretty sure I won't be able to read my old C64 programs.
Even in non-computer media, VHS recorder/players are hard to find already; I haven't seen Compact Cassette Walkmans for some time now, and my sister didn't even know how to operate a record player until recently.
I've already lost most of my old data, and it is very difficult to find any data from before 1997 already. And that's without counting the "dead tech" stuff, like my C64 proggies. :(
The IBM 729 tape drive recorded data on half-inch magnetic tapes in the fashion which was the industry standard. The specs for 7-track and 9-track tape drives are preserved in many places, and many companies produced tape drives that handled these tapes.
Restoring a 729 tape drive in the Australian Computer Museum might well be the cheapest option, but I would expect that there are newer half-inch tape drives in working condition in the United States that could also be used.
It's true that 7-track drives became obsolete, and so most newer half-inch tape drives are 9-track units, however.
Given that the tapes have likely deteriorated, it might well be better to simply take the magnetic recording head from the old 729 drive - rather than fabricating a new one - and put it in a new tape drive designed to
a) handle the tape gently, and
b) record the signal from the heads as an analogue signal (but then digitizing that), so that it can be subjected to careful analysis to determine the correct values for the bits.
On 7-track tape - and old 9-track tapes up to 1600 bpi - there's only a parity bit for each character, plus a CRC checksum at the end. So they don't have the benefit of an error-correcting code.
they turn up on Ebay every now and again.
Domesday content provider
F*ck me - just found my entry in the Domesday 1986 book. Apparently, when I was aged 11 my favourite TV shows were Dungeons and Dragons and Street Hawk. I hated Sunday TV because of all that Religious and Political TV programming.
Also, there's a peice on the playground game of 'Foxey' written by the girl I had my very first crush on!
Twenty-odd years of typing user generated content later and I'm not sure my stuff is any better written... but that's Web 2.0 for you!
Dealing with this now...
...at work, where I've recently provided a 5.25" floppy drive, a QIC150 tape drive and an iOmega multi-disc drive (to read a 20MB Bernoulli cartridge) to various people at work so they can recover data from old media. In the past, I've loaned a 2/4GB DAT drive. If need be, I can provide 8" floppy drives; no idea if the system I used them with (a Lobo MAX-80 Z-80 system with 128K of memory) will boot, though. I have CP/M 2.2 and 3.0 for it.
I'll have to agree with Dave, though, that the risk of oxide separation is high. Perhaps they could run the media through a solution of iron particles (which I've seen done, many years ago) and scan it as it comes out (the particles stick to the magnetized spots) with a video camera. Converting the images to digital data should be straightforward, and there'll be no friction against the oxide, lessening the possibility of damage. Plus, they don't have to get that monster of a drive running, then figure out how to interface to it...
Oh, a tip--IBM uses reversed logic notation--an IBM "true" is 0V, an IBM "false" is +5V
Mine is the worn jacket purchased in 1979--really.
Several odds & ends ...
Recreating the hardware from scratch is a LOT harder than it sounds. Sure, the plans exist ... but what about the tooling used to create the drive mechanism? The chassis? Etc. Remember, these things may have been produced in small numbers by today's standards, but they were still mass-produced ... and the production line is long-since gone. Yes, they could produce a one-off re-creation ... but it would essentially be a prototype and not trustworthy with irreplaceable data.
The "old, dirty" existing unit isn't trustworthy, either. Not even when cleaned up (at least I wouldn't trust it).
Paper isn't an option. Scanning data stored on reams of paper is bloody silly when it was already digitized once. Ones and zeros are infinitely copyable, losslessly. Paper and other analog recording systems are not (this includes punched tape and Hollerith cards). The trick is to anticipate the need, and copy from the old form of digital storage to the new one before it is too late. We've known this for decades ... I have all my personal correspondence archived, dating back to right around the time Ethernet was invented ... Yes, my data is encrypted. It's also stored on air-gapped media, in several diverse locations. Not paranoid, just an old habit leftover from when hardware wasn't all that reliable.
I can still read 8" floppies ... My Heath H11-A is roaring away in the corner as I type. GAWD/ESS, but that thing is loud. Weighs a ton, too ... Looks like both floppy drives and all 64K of RAM are still alive, though. Not bad for a 30 year old home-built hobby box! :-)
Digital archives DO work (see above). You'd think someone at NASA would have (mostly) automated the procedure years ago. It's not hard; I'm pretty certain the scripts I use for my own stuff would scale to NASA-sized chunks of data.
Yes, it would be better to read at constant speed, using a gentler drive mechanism, possibly after baking in an oven for a prescibed time, at a prescribed temp. That's how the professionals do it. Sometimes using electron microscopes and specialized stepper motors to do the reading/positioning. If these clowns are really trying to use old gear to read brittle old tape, they need their heads examining.
No, the Earth's magnetic field does not "permanently erase" this kind of thing. At least not over your lifetime. I just read a couple of 30 year old 8 inch backup floppies on the PDP-11 ... That data is already in my permanent archive, but it's kinda nice to see it in situ for the first time in 15 years (or so). The eldest nephew is fascinated ... "no wonder you like text-only games, it's all you had!" he sez ... Knowing how to store the stuff helps.
I also have access to original air-check tape from KFJC (South SF Bay college radio station), recorded on 1/2 inch tape in the years 1966-1969. I digitized them last year, they sound better than most other original recordings of the same era (Yes, the Grateful Dead's stuff sounded the same as it always does. Might be boring, but at least you know what you're going to get ...).
As a side note, probably the earth's largest "tape drive" is the mid-Atlantic rift. As the North and South poles swap geography, the magnetic domains in the solidifying lava swap. Over geologic time, you get a magnetic banding effect. Useful for all kinds of time-stamping :-)
Yes, there is probably a Linux driver written for that tape drive. If there wasn't before the cache of data was found, it was probably available overnight. If THAT didn't happen, it'll probably be available by this time tomorrow ...
Australian Computer Museum Society:
Yes, cost of recovering the data is the major issue - & proving it is valuable. The US Computer History Museum has restored one of these drives & has recovered 1960s data & has offered to help the Aussie project. The ACMS (www.acms.org.au) is happy to be involved and to find that one of its c.30,000 saved items has generated so much interest. Now if we just had some real support from industry or government we might actually have a museum, and a service to recover old data, rather than the continual scramble to just save the collection, manuals, stories...
Daniel B - C64 + Drive - I have one
I have a C64 WITH disk drive.
Although what i'd do with the data once read is a bigger problem. C64 to PC link?
Archiving Hiccup or c**kup
My organic long term storage module is a bit vague on this but... , was not NASA ordered to destroy all Apollo related documention by congress as part of the terms & conditions for getting funding for the present shuttle? I always assumed the Lunar data would of been exempt from this but has anyone actually read the small print?
Paris because she knows the difference
A number of years ago I transferred many C64 progs from my 1541 5 1/2" drive to my Performa 6360 via 2 phone lines and a good 'ol ATDT command (and others, among which I've forgotten now) on my C64's 28.8k modem.. I don't see why that still wouldn't still work, providing you have 2 phone lines (or tie up a friend's phone line), a modem on the C-64, a fax/modem on the PC and a lot of time to spare for the transfers to complete. ;) Obviously, emulators for both the PC & Mac are numerous..
A couple of years ago I felt nostalgic and pulled the whole setup out of storage and my floppies were still readable.. I don't remember having any errors, surprisingly.
I wonder if the drivers are on Windows Update?
Vintage Data Processor
Refurbishing a 729 is a ambitious project. First there is the availability of parts and a controller. Then there is the cost , a system to attach it to and the available system time to transfer the data to another media. And, while I have great respect for IBM technology. Who knows how reliable a 40+ year old 729 tape drive will be when it comes to processing that many tapes. There may be better alternatives. Here are a few:
. A GOOGLE search on "tape drive" followed by a within search, "7 track". Turned up at least one firm that specializes in tape data recovery. For a project of this size and sensitivity you should deal with a firm that has experience and understands how to handle fragile materials.
. There are newer 9 track drives capable of reading 7 track reel tapes available. Finding one that is available along with system time may be a bit of a challenge. But if you put out the word hopefully someone will respond with an offer.
- Opportunity selfie: Martian winds have given the spunky ol' rover a spring cleaning
- Spanish village called 'Kill the Jews' mulls rebranding exercise
- Reddit users discover iOS malware threat
- Pics R.I.P. LADEE: Probe smashes into lunar surface at 3,600mph
- Ex–Apple CEO John Sculley: Ousting Steve Jobs 'was a mistake'