The information base for the research efforts and the resultant CD/DVD/website has been many and varied. In the main, my personal sightings and photographs have been used as the basis of information for the 1970s to the 1980s. In that 'saga' I was able to speak with many people from all levels of the railway structure. As well I was able to read and copy internal correspondence that provided 'snapshots' to the average goings on.
In the mid 1980s I was lucky enough to obtain several boxes of correspondence that dated from the 1920s to the 1970s. This material provided invaluable as an insight to the typical workings and decision making processes. In fact, the answer to an unknown question often provided the clue to the question and the thinking at the time.
I was lucky enough to have open access to Newport Workshops records and this led ultimately to the Rolling Stock 'Register' books. About 70% of the data in these books were hand transcribed over nine years. Ultimately some 300,000 records were organised into a database. This database ( 'Reg5', a program written by the author) was able to store the data and provide tables and cross referenced information.
What is important is that the 'railway thinking' and methodology of rolling stock activity and record keeping remained unchanged from the 1900s to the mid 1970s. By then, external pressures and several administration moves fragmented the 'paranoia' created in the 1890s by the 'Speight vs Syme' case of 1894. The inroads of rail fans through the 1970s certainly assisted in information dissemination.
It must be pointed out that at this time when I started, information was very hard to get. Rolling stock classes were known back to the past but little information was available apart from VR Annual Reports and operation publications for some era's. Particularly perplexing were the electric car codes and their heritage. There were were few plans, not much data, overall a 'spotty' picture. It was not until many years later this was a direct result of the 1894 Speight vs Syme court case where VR was proven to have misappropriated funds because their book keeping was not accurate. It was shown that VR received maintenance funds for rolling stock that was non existant; either "lost" or scrapped. From these times there seemed to have been a "lockdown" of information, a case of "need to know". It was not until the 1970s and the gradual influx of railfans into the job that barriers were broken down.
The rolling stock data collected satisfied two main research thrusts:
Perhaps the most significant fact of this work is that the information has come from non published sources. In many cases, the data contained is from the 'Registers', with this information being confirmed or improved on from other ledgers maintained at the time. It is important to note that the classes mentioned in the CD / DVD / web site ran and have an official source and can be considered a 'definitive' list. Despite many 'home grown' lists and researchers forming personal interpretations, significant information is required to refute the core data and code lists.
As many vehicles as possible were photographed between 1973 and 1985. Photography included sides, ends, roof, doors, interiors, lettering and many other topics to assist a modeling interest. Photos of construction, repair, derailments, storage, scrapping and wagon bodies throughout the state were 'hunted down' to provide examples of all facets of rolling stock history. In all some 17,000 vehicles were photographed with most of the vehicles being photographed two to three times in various stages of service and modification.
As the information began to 'pour in' the data logistics became overwhelming. To this date (1984) any historical recording merely duplicated the existing written systems of ledgers. Indeed early computer methods were simply text based systems based on the ledger method. The enormous manual effort required to maintain the data resulted in the paranoia of passing on information and extreme effort to extract meaningful information. To assist in the data collation, a computer program was written that enabled the information to be cross referenced despite entries being added in an ad-hoc fashion. It was crucial that data entry did not require prior knowledge of a vehicle as is usual with ledger type entries. Despite being developed in 1988, the program has not been duplicated. The program gave the author the ability to request any vehicle history and the data would be returned as a chronological data paragraph across all code and number changes. In the eight years whilst the the program evolved, data entry of 230,000 histories was added for some 81,000 vehicles.Previously I have not acknowledged sources because the information has either been destroyed, archived or is generally inaccessible to the casual researcher. Indeed many of the conversations and questions with people cannot be duplicated because they're dead.
The pool of information comes from
Compiled by Peter J. Vincent, 26/11/2018