[Bell Historians] Indexing foundry marks

Fri Nov 27 11:06:38 GMT 2015

I hope an amateur local historian can submit a humble view.

 From a cataloguing/database point of view, this suggestion has a lot to 
offer. In the form of a string of text like a blazon, it can include the 
equivalent of several spreadsheet cells in one element. This could 
easily be converted into several specific cells of data. Also, the 
individual descriptions could be represented by single/dual/etc 
characters, and so a very short coded string could represent a complex 
design, especially if only the differences to 'standard' element designs 
needed to be specified (such as - in heraldry - rampant, guardant etc.).

So we could for example begin generically with 'star' and then progress 
to '6 points' then 'wavy arms' then 'diameter/height' and so on. As part 
of a code string, these four levels of description could be represented 
by for example 'S6344' (I'm imagining a height of 44mm here). Anyone who 
then uses this database on a regular basis might before too long 
immediately know what S6344 in this part of the code string means. It 
could also be followed by a further character which might be a reference 
leading to a note field recording any unusual elements of the design in 
a separate table, a bit like footnotes or endnotes.

I would like to add an observation on a different aspect of the problem, 
that of data capture in the field. I recently had one experience of 
trying to 'capture' two foundry marks on our parish bells, in response 
to a request from an author who was working on a study of kings' head 
foundry marks around the country. (He published his study in the form of 
a map.)

I'm assuming that to start a new catalogue without relying on 
pre-existing data, the project would really have to go right back to 
first-hand data acquisition of the marks on the bells in the church 
towers. Existing lists could be used to identify what to expect to find 
at each location, but from there on in, to enter specific design 
elements of individual marks would probably need to rely on direct 
personal observation, and a record of each mark would be made by 
photography.

If this is about what would happen, the photography would come with 
issues. As we are well aware, foundry marks could be described and 
classified according to a set of guidelines, using an established 
protocol as suggested with comparison to heraldry above, or by filling 
in standardised entry forms (always too restrictive in a few cases), but 
the photographic recording would at the very least need to be controlled 
in a similar way, otherwise there will be a mass of photos taken by 
photographers of different abilities, using different cameras and lenses 
set to different options and values, with or without flashes of 
different specs, in differing conditions. This could become a 
cataloguing nightmare!

When I tried to capture our king's head foundry marks with a 
bell-ringing colleague, we took lots of photos, which I admit did allow 
the author to identify and classify the marks, but it was clear that 
several photos were needed 'to be sure' (as well as some extra replies 
to specific queries) and they all suffered from inevitable variations in 
lighting, angle, focus, lens f stop value, etc. (there was no 
professional photographer present!).

In short, to accurately catalogue foundry marks using photography would 
probably require the addition of a great deal of 'superfluous' data to 
the database, in an attempt to 'understand' the images actually shown in 
the photos.

To start the ball rolling with taking this aspect of the exercise 
forward, I suggest two possible options. Others could come up with more 
I'm sure.

Suggestion 1. Whilst not an archaeologist (I'm not a lot of things), I 
have had a lot of exposure to its practice during my membership of 
Waltham Abbey Historical Society. In archaeological reports, finds such 
as pieces of pottery are very carefully described and catalogued to the 
umpteenth degree (which indeed may be another example, to add to 
heraldry, of a parallel discipline worth looking at). I was also an art 
student in my Dark Ages, and I can say that whilst it is very common to 
see line drawings of these pottery fragments in published reports 
instead of photos of them, these drawings are not mere 'artistic views' 
but are constructed according to a specific set of conventions to 
accurately represent the object's dimensions, in a way which other 
archaeologists can use as accurate references to, and comparisons with, 
finds made in their own research. This could be applied to foundry marks 
- I just wish I knew more about the process to put my case better, but 
hopefully others can pick it up and run on with it.

But how to apply this method to items which cannot be taken back to labs 
not 'in the field'? I suggest that instead of using (or at least relying 
on) photography, a casting be taken of each foundry mark, which could 
then be used to make a positive 'model' (to full scale of course) of the 
mark, for study and classification in easier working conditions than a 
belfry. We tried this with our king's head marks, but unfortunately the 
very short time period we had to work in forced us to make compromises 
which rendered the exercise a failure. In essence, all we could get our 
hands on at the time was Play-Doh (don't laugh), which of course refused 
to set hard even when left in a fridge for ages. But it did manage to 
shrink!

However, I'm sure that a simple 'impression making' kit could be put 
together by individual data collectors for use in the field. something 
like copying a bank vault key:

1. find a tobacco tin or jar lid wide enough and deep enough to contain 
the foundry mark.
2. obtain some hard-setting modelling clay, such as Das.
3. Take these and some PTFE releasing agent (WD40) into the belfry.
4. Lightly clean dust and cobwebs off the mark with the PTFE spray.
     This will also stop the clay from sticking to the bell.
5. Fill the lid with Das and press it onto the mark.
     If the mark is cut into the bell rather than standing proud, 
overfill the lid first.
6. Remove excess Das squeezed out of the sides of the lid.
7. Back at the lab/kitchen, let the clay set hard (some products need to 
be baked in an oven).
8. Make a positive from your mould.

The resultant set of (for want of a better term) 'death head masks' 
could be collected together in one national repository, or several 
regional ones, to form an archive for future study.

Suggestion 2: Go back to photography, but with a difference. There have 
recently been significant strides forward in 2D to 3D technology, or 
photogrammetry, and it is becoming more and more common to find this 
applied to such things as famous sculptures, museum artefacts and so on. 
In short, it is possible with some programs to accurately record an 
object in 3D using 2D sources (photographic images) - even to the point 
of being able to extract accurate dimensions of an object from photos, 
given certain bits of extra information (mainly the camera lens f stop 
setting - which is usually stored as "exif" data in jpeg image files).

The basic idea is the old stereoview card trick in reverse - take as few 
as two (ideally more) photos from different angles, and use computing 
power to turn them into a 3D model, based on the different overlapping 
of various parts of the object shown in the photos.

True, this is mostly used in very different situations to belfries, such 
as 3D representations of historic buildings, digital terrain mapping in 
3D using video footage from aircraft and drones, and so on; but the 
market is evolving fast, and I think it may (or may soon) be possible to 
use this approach to make an accurate 3D record of a foundry mark in a 
belfry using nothing more than a mobile phone camera operated by anyone 
who owns one.

Have a look at this product - I must first say that I'm not in any way 
connected to them, I only found out about it yesterday, whilst looking 
into ways to make a 3D model of our church's western façade, for an 
entirely unrelated project. This product doesn't fit that bill as it's 
aimed at recording small objects such as pet cats, and it may well be 
too 'amateur' for foundry marks as well, especially as regards 
extracting dimensions and measurements from its 3D output (though that 
could be bypassed as simply as holding a ruler next to the mark when 
taking the photos), but otherwise it might provide a very simple and 
straightforward solution to the problem in hand. It's also free.

http://www.seene.co/app/

If the general idea appeals but not this product, a comparison of the 
more usual players in this market, which is also a starting point for 
researching their capabilities and suitability, is here.

https://en.wikipedia.org/wiki/Comparison_of_photogrammetry_software

I hope this long posting is of some help.

Lawrence Greenall
Waltham Abbey, Essex

On 26/11/2015 18:25, Richard Smith richard at ex-parrot.com 
[bellhistorians] wrote:
>
> Dickon Love dickon at lovesguide.com [bellhistorians] wrote:
>
> > Has anyone done some work on developing an index system
> > for foundry stamps and other marks on bells?
>
> If I were developing a system for indexing foundry stamps,
> I'd look at how similar problems have been solved in other
> disciplines. The problem of indexing coats of arms is
> conceptually very similar. Arms can be indexed by surname
> just as foundry marks can be indexed by founder, but that's
> no help if you want to look up an unfamiliar coat of arms or
> foundry mark. A list of arms ordered by design is known as
> an "ordinary", and they are arranged in alphabetical order
> of the main thing ("principle charge"). Many geometric
> patterns have standard names by which they're indexed; other
> things are grouped, for example all animals other than birds
> get listed as "beasts". Within the categories, they're
> indexed by number, forms of decoration and other items on
> the arms. I'd take a look at one.
>
> RAS
>
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ringingworld.co.uk/pipermail/bell-historians/attachments/20151127/8b49409b/attachment.html>