|
In thinking over the suggestions above, and also from a couple of phone calls and emails, it is becoming clearer as to what might be manageable or not. For the physical microfilms, public access probably won’t be realistic.
Cards are grouped into two main lots, Australian and UK. The way in which they are ordered differs. For the Australian drawings, the storage order is shown in the example below.
AYH3685
EXP3685
HYA3685
HYB3685
AYH3686
EXP3686
HYA3686
HYB3686
AYH3687
EXP3687
HYA3687
HYA3687
HYA3687
HYA3687
HYA3687
HYA3687
HYA3687
As you can see, the cards are stored numerically, and then alphabetically. It’s fine once you get used to it. But this means that AYA4656 is not stored next to AYA4657 for example.
For the UK drawings, the order is slightly different as shown for some drawings ending in 7488.
AHA7488
AHH7488
ATA7488
CZK7488
CZK7488
EAM7488
14A7488
1G7488
2K7488
4K7488
11K7488
21K7488
24K7488
Here, the cards are sorted numerically (suffix), then by the “BMC” system (alphabetical prefix), then by the “Austin” system in which the letter part of the prefix is sorted first, and then the number part of the prefix. There is a logic to it.
A particular part may exist on an assembly drawing, or included on another drawing which has a different number, or be on a drawing all of its own. Often, to locate a drawing, it is necessary to dig through a few layers of cross references to find the actual drawing needed, and then you might find because the drawing was so large originally, it was microfilmed over three or four frames which then have to be stitched together if you want to see the whole thing at once.
Since multiple cards are usually involved in any one search and sprinkled all over the place, for public access, one must rely on the person putting the cards back exactly where they came from. More than 20 or so cards out of sequence then a card might as well be considered lost and can then only be found by accident. Can the general public to be fastidious in this respect?
As Doug mentioned, once you’ve located the card(s) you wanted, then it is necessary to view the 35mm microfilm on the card. You can do this with a microfiche reader, or a dedicated aperture card scanner – the latter being far preferable but quite expensive and usually only found in libraries. Then, you need to print the image – and many of the drawings require adjustments to contrast and brightness – sometimes for different portions of the same image, to get something readable.
The whole process might take up to an hour by the time you have a drawing in your hand.
I am sure you are all fascinated by this lengthy description but if you would like to understand the steps required, such as would have an engineer at the factory, this is what had to happen.
The microfilms, being high resolution 35mm film, requires storage in an appropriate atmosphere (dry and cool) and now some have started to fade quite a bit after their 40 years of existence.
While I think it is important that the physical cards be preserved for historical reasons, and they are the closest version to the original paper drawings we have, access by interested persons unsupervised, without the required equipment, is probably not workable. The only way the original cards can be used is if a small number of dedicated persons have complete control over the whole thing to ensure they are properly stored and looked after – and Peter Davis at the factory had four blokes full time on this activity.
This brings us to the digital versions of these drawings.
There are two aspects to this – an actual digital scan of the microfilms, and then a computerised listing, or catalogue, of what’s actually there.
For digital scanning, I used an ancient Wicks and Wilson aperture card scanner which runs only on Windows XP using an old style of DMA interface card. Somehow I’ve kept this thing going for over 10 years (already some 30 years old when I got it) and its scanned about 200,000 cards and still going, but you really need to know what you are doing with it and be willing to repair gears, drive belts, circuits, and make adjustments to the optics and so on to get it to produce something good. The files are TIFF format and it can scan up to 400 dpi, but due to the nature of the software, once a file gets larger than about 12 MB it crashes and then you have to reduce to 300 dpi – still more than adequate to allow significant zooming in.
The cards themselves are usually plain cardboard with handwritten drawing numbers on them, but a very small number of cards are punched with what is called Hollerith data – in which the scanner can interpret as text and name the file with the part number. But, for about 99% of cases, the files just come out SCAN0001.TIFF and so on.
So, in order to get all this into a usable state, one has to handle each and every of the cards and write down the part number and match it against the digital file name. I think it took me about five years to get to this point.
What is even more useful is to have a Description of the part, the date of the drawing, and the model to which the part applies. Unfortunately, this is only on the actual drawing and so one has to page through each image file, zoom in, rotate, flip, invert, squint, etc and write this down against the part number.
The picture below shows a very small part of the overall catalogue file, everything is hand typed in. The whole database has 200,000 records.
On one occasion, I asked John Smidt if I could borrow his parts list and match the part number with the description had had already assigned to it. I thought this would save a lot of time at least for Mini parts, but alas, John’s descriptions did not often match what was written in the parts book or on the drawing – he would write something like “Door Handle early model with sliding windows”, or “Badge, second hand”. In the end, I took the Descriptions direct from the drawings and these often do not match with the Service Parts books – what we would refer to as a “gasket” would be called a “Joint Washer” on the drawing.
To manage all this, I wrote an application in Visual Basic which allows any drawing to be located quickly based upon a simple search of part number, description and model, and then displayed on the screen. The catalogue can also be searched using SQL language statements for complicated searches. A list of part numbers could be dispensed within seconds. The App could cross check for missing drawings, inconsistent filenames, import scanned part numbers from a parts book, and many other functions. The images can be zoomed, rotated, panned, clipped, saved to JPG, copied to clipboard, etc. There’s no way I could have produced those books without a facility like this, and it took years to develop.
Now, should all this be made public access? Yes! you say – but there are some problems. The whole thing took about 100 DVD’s to back up, and this gives you an idea of the storage space needed on someone’s server. An App is required (I’ve written one) with a web interface which requires ongoing maintenance for bug fixes and improvements, let alone its initial development. The IP of the content is undetermined and there could be liability problems arising from the manufacture of parts from the drawings, and once a drawing is downloaded, it might as well be given away to everyone (look what happened to Craig Watsons’ magazine).
So, while my thread was initially about long term preservation so the cards, I think it worthwhile that you realise the magnitude of the effort that has gone into this project as a whole so that the material is given the respect it deserves. A terrific resource and testament to the efforts of those in the factory - if only they had an IBM PC at the time.
Regards, Tony Cripps
You do not have the required permissions to view the files attached to this post.
|
