Hacker News Reader

Comment by HugoDaniel | Hacker News Reader

The archivist’s final recommendation: Destroy the tapes. “There is no evidence that suggests this material is historically significant... I recommend disposal through the immediate destruction of all magnetic tapes.”

Why ? Lack of evidence does not prove that it is historically insignificant. Also being historically insignificant does not imply it has no value.

Why not convert them to a digital format and put them on the internet and let the hackers/artists/archivists/space travelers/etc decide ?

View on HN

For media of this age and in poor condition, conversion to digital format is an expensive and time consuming process - they would most likely need to hire a specialty company at a rather high rate.

In the best case, the tapes would be cleaned (very time consuming) and then read on a 'newer' 7-track tape drive (although not many of these have been manufactured for decades, the 9-track format having taken over in the '60s). A more likely case is that the tapes will also require some splicing to remove irrecoverably damaged sections that cause the drives to track poorly, and probably all transfer to new reels in the process of hand-cleaning. Then the data format on them is probably unknown. It's possible, although far from guaranteed, that there will be a COBOL-type format header, but even that can be very difficult to work from. This is particularly difficult since when you miss words due to tape damage it may not be clear how many words you missed (if the drive can't track at all), and the drive may not be helpful in exposing that this happened either.

At the end of this whole process, it's likely that the tapes contain the output of a test regime or model that isn't even parsable without the original software, which could possibly be on one of the tapes but given the era and machine (probably a CDC160 based on the CDC163 pictured) is just as likely to have been on punched paper tape which would definitely not have survived this many years. We're talking a series of unitless numbers, probably from a matrix of unknown dimensions (but possibly just a series!). It would be pure speculation what they even mean.

If the data is in BCD format (possible for this machine, particularly if any non-integer values were to be stored), then it is likely in fixed-width fields and it will not even be apparent where one number begins and another ends (depending on the data, this may be inferred by padding, but the location of the decimal point would remain hard to determine).

In historical preservation, you have to choose your battles.

I'd even argue there's no specialty company that would be like "yeah we have a process for that" - a process to restore and recover the data would have to be invented on the spot. Which would be possible, but there's not a big chance of success. Cost... that just depends on whether there's a hobbyist going "I can do it" or a company going "Give us $x monies and we'll try it".

BCD you say? Somewhere there's a COBOL programmer (who saved the world back in 1999) tanned, rested, and ready to restore this tape!

Is this what they mean when they say "never admit you know COBOL"? Is that programmer you?

"It's complicated. It screwed me now and then, but that's not a real basis for a relationship."

Possibly - but COBOL was actually very new when some of these tapes were made according to the date range in the article, so there's a high chance (I'd bet on it) that the tapes are not in COBOL record format. BCD was implemented in the machine language in the IBM machines and so it was widely used prior to COBOL, the CDC machine involved here does not have BCD in its instructions but still provided some support for BCD operations because like many smaller computers of the era these CDCs were often used to pre/postprocess tapes also used with larger IBM machines - in an analogy to more modern architecture, it was not uncommon at the time to 'map' on an IBM mainframe (collect data from other sources, often) and then take the tapes to smaller machines to 'reduce' (totalize, etc). This allowed for using time on the very expensive IBM machines more efficiently.

So you've already worked out all the tricky parts and done the calculations on the back of a napkin.

The only remaining question is, when can you start?

It's not about COBOL; it's about the operating system and hardware on which they were written. Manipulation would have been via JCL (if from a mainframe) rather than from within a program. It was typical to have a multi-file header with volume information (e.g. a unique tape ID, encoding details such as BPI and number of tracks), file details (name, creating program, record size and block size if fixed, etc.)

In the mainframe example, JCL tweaks could let you bypass any automatic label processing and get directly to the files if they are intact.

[EDIT]: I didn't see the part about "severe mould" initially; I think that would make recovery a non-starter.

And at least for some old tapes, copying them is a read-once process. You can use an old drive that has been modified to not backtrack and retry on error. It doesn't help, and makes everything else worse. You run the tape straight through, non-stop, and get three things: some bits of data (if you're lucky), a rewound plastic backing, and a small pile of iron-rich dust.

> In historical preservation, you have to choose your battles.

Though you sometimes get fun surprises. An old Lisp Machine (hmm, or ITS PDP-10?) backup tape contained an unintentional screen dump. Freezing a moment in time. Editor session in progress. And an animated bug hack crawling across the screen. An virtual insect preserved in magnetic amber.

Since the tapes were damaged by "moderate to severe mould," it likely would have taken a non insignificant amount of work to convert them to digital format. NASA doesn't want to waste it's engineers' time - this seems more like work for a historical society or organization

Think about how much budget and time and resource and nasa-career-path risk it would take for a manager at Nasa to ask for taking on a restoration project like this. Not going to happen.

Why would it have to be NASA? I feel like would be a lot of NASA / space historian people that would be interested to try to restore them without charge.

Because DOOM is real.

Reply on news.ycombinator.com