National Bells Register?

[JohnBaldwin]

A few weeks ago I promised members of this chat list some information on
work that I had been doing towards a National Bells Register (NBR). First,
let me apologise for the fact that this didn't get to you in advance of the
Central Council meeting. I had fully intended that it should, but a number
of things conspired against achieving it.

At the Council meeting, I made a PowerPoint presentation about what I had
done and some of the issues surrounding such a project. If you would like
to have a copy of the presentation itself (355Kb), a sample report in Word
format (31K) from the little pilot data that has been loaded into the
tables, and a Word note about some of the obstacles that I foresee (23Kb),
do please send me an off-list request and I'll happily forward it to you -
ask for 'the presentation', and when you get it, look at it in Notes view to
see what I intended to say against each 'slide'. (The need to get the
presentation ready was one of the reasons for not getting back to you all!)

In what I had to say, I tried to summarise the following points:

1. The database into which the data has been loaded is an extension of the
DOS DataEase database which currently holds the Dove detail. I need to
emphasise that I DO NOT envisage that this software would be a satisfactory
basis these days for a NBR, but I was wanting to start somewhere. Moreover
I am totally familiar with it, have to use it still in my fulltime job
(which now is primarily about designing and using databases), and it is a
useful starting position in that all 3-and-upwards rings are already
identified as such in data that I currently have. (Further, trying to use
completely different approaches to a given problem makes for slow progress
with either - and at the moment I see no reason to change the basis for
Dove - it ain't yet broke [that is until Mr Gates kills off our being able
to run DOS software], and therefore I don't need, yet, to fix it. But I DO
intend, in the not too distant future to move things across to a MySQL
approach - and then the NBR pilot stuff can also come across.)

2. A VERY small amount of data has been loaded into these database tables,
taken from 'forms' prepared by the Winchester & Portsmouth DG in a tower
survey carried out in 1997-98. I am grateful to them for their agreement to
use their data this way and we must respect the fact that any data you see
displayed is theirs.

3. The basis on which fields of data were decided as being necessary was
based upon what was described as the Dawson-Hayward format, briefly shown at
a meeting of interested parties held at Leicester University in May 1999
towards the end of the CCCBR's Ringing in the Millennium project. (An
attendance list at the meeting is included within the notes of the
presentation.) At that meeting a number of papers were presented - and I can
supply, separately, if people haven't seen it, a resume of that meeting (ask
for Discussion Note 1). I have to say that most of the issues then foreseen
as potential problems have in fact turned out to be such - but others have
also since 'raised their heads.' ('Ownership of data' is another issue
which I did not really address at Llandudno but was clearly a sensitive one
at Leicester.)

4. It was, and still is, not clear to me that a NBR is primarily within the
purview of a CCCBR - where the R stands for Ringers. That turned out also
to be the view of the meeting on Monday at Llandudno. However, that equally
does not stop either the need for some standards / recommendations to be
developed (and such were requested by the Council), nor for work on this
project to cease. I still have plenty of ideas as to how it might proceed,
and also Jason Hughes' offer to take this forward is on the table, and I'd
like to work alongside him on it.

5. One of the essential tables (as is evidenced by the sample report) in
any NBR will be that of bellfounders and their working dates. To this end,
I have been working (with his agreement, and help) on getting David
Struckett's list (as in his Dictionary of Campanology) into an Excel
spreadsheet - and in due course that can readily be assimilated and used
within a database.

6. I do see the need for researchers to be able to submit their own data
into any NBR that may come about (and subsequently to 'cleanse' it). I do
NOT feel that there is (at the moment anyway) any mechanism by which people,
who have their own data in vastly varying formats, can just submit this to
someone else with the object of getting their data into a pilot NBR.
Indeed, deciding and creating a suitable and generalised mechanism for
achieving this has proved to be, for me, the major stumbling block to more
rapid and visible progress. I have looked at trying to use Excel, or Word
(tables) as possible precursors. Here is an area, in my opinion, for
someone to come up with an imaginative solution - which caters both for
'bulk input' (of already existing data which may people already have and
would be willing to share) as well as for 'one offs' (where the completion
of an 'on-line form' would be a satisfactory way of proceeding - but would
be tiresome in the extreme for more than a small number of bells: take it
from me, from what I've already had to do for getting the W&P data included!
John Scott, for example, has details of some 3000 bells which he has said he
is prepared to make available - but I believe they are all kept somehow in
Word format.) I also feel that the resulting NBR ideally should be
searchable from the internet on an ad hoc basis as well as being able to
provide 'standard' reports.

7. It may prove to be helpful, initially, at least to define a minimum
subset of data fields for each of a number of tables (eg, towers, bells,
bellframes, founders) and I hope soon to a little more work on these to
present to members of this list for their comments. What one must bear in
mind is that it is incredibly easy to decide to provide 'space for'
absolutely every small datum - and have almost no cases where all fields are
completed. Better, in my mind, is to have >90% of all entries having >90%
of the fields completed with data.

I hope this 'introduction' puts what I tried to say at Llandudno into
perspective and I offer it to you all as a contribution towards what I hope
one day could become a NBR.

John Baldwin
(029) 2055 4457