The Register program was designed to output data for a single vehicle identity across all the changes that had been recorded, with the output in chronological sequence. This required strict formatting and a set of "rules" by which the program could determine a course of action for each history found. The output has a standard format that comprises of several parts:
History output was divided into two types:
The output formatting was organised to cope with extreme volumes of data, with output based upon unlimited pages being available.
A sample output is described below.
The program was developed in DOS. Part of the heading was used to set compressed text when printed. Windows programs now use font and size attributes. The dashed line includes the owner of the program. The description line contains the output report number, date and vehicle class description. The final dashed line containers the number of columns each page is formatted to. Program owner and formatted columns per page were parameters set within the program.
3 CU------[PJV'S ]----------------------------------------------------------------------------------------------------------------
[R11] 14/01/2007 Data for Code JCX Version (VA)
1974 - VR Bogie Cement Hopper - Gravity Freight
------[128]---------------------------------------------------------------------------------------------------------------------
This is a description of what the program searched for. Can give a number, a number range or a generic "all the vehicles in the class". The last description indicated a request for everything recorded that includes data logged with unknown vehicles. Examples would be
Search for 101
Search from 101 to 108
Search for all the vehicles in the class
To assist with DOS printing and ensuring all data stays on the same page, the program determines the lines per page and adds a page eject code at the end of each page output. This formatting must include a data column description header at the start of every new page.
Vehicle.............SV # Date...... Loc History............SV Notes
Data is generally accumulated from many sources with data usually being collected out of order across a long timeframe. Rarely would all the data be presented. The program method allowed for ad-hoc data collection and data entry but provided a tool that presented the items in chronological order.
The data entry below would generally have been written as: JCX 101 built new and issued to service 9/4/1974 at Newport Workshops. Recoded to JCF at North Geelong Yard 1/1978. Recoded back to JCX 101 on 4/7/1978 at North Geelong Yard. On 14/4/1979 JCX 101 was recoded to VHCX 101 at Bendigo Workshops. Converted to VHQF 204 in the 1990s.
Additional information about photographs, references to vehicles and observations were generally not included when passing on information in the manual long hand days. With formatted data and the ability to provide ad hoc entry for anything, the program provided a simple method to accumulate all types of data and with that the easy cross checking of items. Quite often we found we have photos for events that never occurred until months later.
*JCX 101.VA - 9/ 4/1974 NWS Built new -
/ 1/1978 NGY To JCF 101.VA -
JCF 101.VA - 4/ 7/1978 NGY To JCX 101.VA -
*JCX 101.VA - 14/ 4/1979 BAW To VHCX 101.VA -
VHCX 101.VA - / /1979 - Photograph #1334
/ /1980 - Photograph #1581
21/ 9/1980 TOT Photo (partial view) #1632.35: on SG train
/ 7/1984 BAW Out of Workshops (RO wagon stencil)
22/ 9/1985 CAD Photograph RO/ V/line logo
1/ 3/1989 - Reference Rosco sheet
16/ 3/1990 BOW Out of Workshops Lift and paint
circa 1990 - To VHQF 204.VA (v)
VHQF 204.VA - 27/11/1994 TOT Photograph RO: Slope sheets, Vline logo
The following history shown in full reflects the histories of the two "WW 11" histories as presented showing the use of 1st, 2nd, a car name "Victoria" and the use of the tag ">>>"
A 19.VA - 1/ 3/1887 - Built new 8 whls; WHB; Corr. 2301
circa 1887 - Note "New departmental Victoria"
circa 1888 - To AA 19.VA -
AA 19.VA - / /1894 - To Victoria .VH 94/6485a
Victoria .VH - circa 1894 - Equipment "Knowles ventilators"
5/ 8/1932 - Into Workshops -
9/ 9/1932 - Out of Workshops (36 days)
8/10/1934 - Into Workshops -
18/ 3/1935 - Out of Workshops (162 days)
10/ 8/1943 - Into Workshops -
19/11/1943 - To WW 11.VA (102 days)
*WW 11.VA 1st 9/ 9/1948 NWS Into Workshops -
14/12/1948 NWS Out of Workshops (97 days)
16/ 6/1954 - Scrapped -
circa 1955 CFD Body Behind Scout Hall, Trugunnini Rd
circa 1974 - Sighting Peter Clarke: Disappeared several years later
23/ 6/1996 CAI Body [SXC]: Behind Ventura's, Mimosa Rd MEL68H4
>>>
AA 200.VA - 24/12/1890 ELS Built new WH 90/12252
/ /1894 - To ABAB 39.VA 94/7981
ABAB 39.VA - circa 1894 - Modified 2nd Class Perf. Seating
circa 1910 - To AB 37.VB -
AB 37.VB - 22/12/1911 - Modified 45' - 57'3"
25/11/1930 - Into Workshops -
19/12/1930 - Out of Workshops (25 days)
13/ 9/1940 - Into Workshops -
19/11/1940 - Out of Workshops 68 days
21/ 5/1957 NMW To WW 11.VA Npt: 30/5/57
*WW 11.VA 2nd 6/ 4/1977 - Scrapped -
A peculiarity of a "step" history output is that a vehicle being searched for may be the last entry of a paragraph and will not be marked with an asterisk ( '*' ). In such a case a special line is added at the bottom with an asterisk an dashed line beneath the vehicle being searched. Example below:
BB 15.VA - 11/ 4/1892 NWS Built new WH 92/2701
29/ 6/1910 - Modified 45' - 57'3"
circa 1910 - To B 15.VE -
B 15.VE - 16/ 2/1921 - To BT 9.VB -
BT 9.VB - 31/ 1/1924 - To T 35.VC -
T 35.VC - 28/10/1970 - Off Register -
3/ 2/1971 BAW To WW 175.VA -
* ---------------
Notes for an item that are unable to be displayed on the same line are referenced and placed below the data output.
*WW 96.VA - 10/ 6/1981 - Scrapped -
28/11/1982 BAE Body See Note 1
Below the output the notes are formatted to fit the page as shown:
Notes as referenced......
1: [DMG]: Property J Blackall & Sons. Gone by 9/1/1991 - to MRPS at 1994;
[BXY]: at Malmsbury for auction 12/97
2: [GXO]: Located on own track with canopy. Occupant to go after
operation on hip. Car to be offered to ARHS/groups. PJV: Occupant
operated Crane 45.
Here is where all the vehicle classes are described. This avoids detailed explanations amongst the item data. This information is implied when reading Registers and book entries. List is displayed in alpha numeric order by class
Code SV Date Sys Description Type JCF VA 1977 - VR Bogie Cement Hopper - Gravity Freight JCX VA 1974 - VR Bogie Cement Hopper - Gravity Freight VHCX VA 1979 - VR Bogie Cement Hopper - Gravity Freight VHFX VA 1990 - VR Bogie Fertiliser Transport Freight VHQF VA 1987 - VR Bogie Quarry Products Hopper Freight VHRX VA 1987 - VR Bogie Quarry Rock Hopper Freight
Location "codes" were required to format a short single line history. The concept was copied from the VELAS, later ROSCO systems. To avoid laborious cross referencing of codes across multiple programs running around the state, the program had the ability to generate a code for any location using the first two letters of a location plus the last letter. This option was then presented to the user if unused by the program. If already in use the program started at "AAA" and worked through until it found an unused letter group. If the user typed in a specific letter group, the program would accept it unless it was already in use. So whilst most locations have a corresponding VELAS code, many locations prior to the computerisation are made up. Comparison between operating REG programs will spot differences for the same location. The idea was to identify the location and provide an easy to remember system.
Code Location / Builder - - ARL Arrival Yard, Melbourne BAW Ballarat North Workshops BOW Bendigo Workshops BRN Brooklyn, Victoria CAD Canal Zone, Melb Yd NGY North Geelong Yard NMW North Melbourne Workshops NWS Newport Workshops SDY South Dynon TOT Tottenham Yard
To provide some referencing ability, major references are standardised then added to the notes as required. References are delimited by square brackets ( [....] ). These references are then itemised in alpha order if they have been described. If not described they will be ignored.
In general, where the data is either unique or the data is quite clear, no reference is provided. Because the intent was NOT to have proof of every entry. Instead, unusual or conflicting information would be provided with a source back to the data for referral.
References... [NB] : Norm Bray [ROSCO]: V/line or NRC ROSCO information [RO] : Rob O'Regan
The program finishes all outputs with a simple "End of output" message to indicate the report finished without the program crashing. To assist with searches, the program times each report in seconds. This was track the anticipated delays on older machines as the data. This program started started on an XT type PC and has gradually been moved to faster machines as they have become available.
<1s> End of output.
The conversion of written records to a database system involved
many decisions about data storage and the meaning of the
data as presented in books. Much of the material is without dates
and only part of the story is actually told; ie, there is a 'hidden meaning'
behind some data and you're only getting the literal 'official' sanitised version.
The main records were the 'Rolling Stock Registers' of which there have
been three versions, not including ROSCO and the new RAMS system.
Version 1: All records prior to 1886 ( destroyed )
Version 2: The 1886 - 1960 Register books. These books only contain
data for rolling stock that was in service AT 1886 or later.
Rolling stock built and scrapped prior to 1886 is not
indicated.
Version 3: Loose leaf typed sheets presented as seven volumes
Freight (3), Tank wagons, Carriages, Suburban, Sundry
Three quarters of version 2 and version 3 have been hand transcribed
by Rob O'Regan and myself in a form that preserved the basic details
and gave an off site record once access was lost. This took some
10 years.
These records are 'Item based', ie the histories are recorded by vehicle.
Therefore notes and conversions for many stock have no dates. Extracting
dates relies on events of the time, hand writing, ink and pencil scribble:
each with their own meaning.
In fact, much of the data from the Registers has been misinterpreted
because the data is overwritten, there are no dates and some work was
pre-written and the page was later used for other stock. Records
were placed into the pre-written number spots even though that vehicle
never used the number shown.
However there are other records that were
used to make the 'Rolling Stock Register' entries. These records
have not yet been unravelled. These include:
1..Circle Books 1911 - 1985
2..Newport Registers 1958 - 1985
3..Workshops In/Out 1919 -1952 (ARHS archives)
4..Capacity books 1914-1951
All these records are chronological based and allow insight into the
real happenings of the rolling stock and conversions.
But to return to a few basic assumptions made in the data entry of this
work. There are several periods in rail history where code changes and
events were not recorded in the main registers. The Registers are a
capital expenditure record, not a record of workshops activity. The following
assumptions have been made due to the lack of data. The records can be edited
once more data is found.
The main assumptions are:
Memorandum dates: Many records are referenced by a memo reference such as
'97/1245a'. This is an trail to show that the work done
can be referenced to an authority. From 1886 to the 1900s
the year shown was the year the work was done. Howver, in
some cases the work was done much later. Extreme example
was Joint Stock that returned to South Australia in 1904
yet the correspondence given was '97/xxx'. This in fact
was the correspondence that initiated toilets on trains
for the MEL-ADE corridor in 1897 and was still a reference.
So until specific dates can be found, this is the best
we've got.
1910: In 1910, the VR changed from a long letter system to a simplified
one. No dates, apart from rare examples are given. It is erroneous
to tie in two unrelated activities to a single date given to one,
just because it occurs within the same time frame. It could be
inferred but not proven. My assumption was to place all code conversons
at 6/1910. For work within the same time frame I have placed the
conversion as an approximate date prior to the work done.
After reconsidering that the work took some 18 months, I now believe
the work was done within about three weekends of July 1910.
A quick thumbnail calculation of car numbers, repaint rate, and
workers clearly shows it is possible. More effort was required
to perform a triennial stocktake.
I to IA: There are no dates for many of these conversions. So I have adopted
the date '6/1936' as a default. To tie in with
autocoupling conversions (1928 - 1938), I have assumed the
I to IA occurred prior to autocoupling, even though this is
incorrect. Conversions of I to IA was a result of IB scrappings:
approx 80-100 8/10/11t I/IB's were scrapped from 1928 at a
constant rate. I to IA conversions occurred to replace this
smaller stock. But apart from a few memo scraps, there are no dates.
Any assumption made would be incorrect.
The 'Accuracy of data' piece was in response to those who forward and print
hunches and theories in the main body of their submitted data.
Whilst I agree there is room for such questioning, this material is better
served as a footnote in the absence of data to back it up. This
promotes and perpetuates like an 'urban myth'.
The program was designed to store data and retrieve the data upon
request. This varies from the standard methodology of text entry and
always being able to see the work. Outputs are clearly dated to show
the data on hand at that date.
The outputs generated are either 'Local' histories ( specific to the
vehicle asked for ) or 'Step' histories ( track backwards and forwards
to assemble a 'Built New' to 'Scrap' continuous history )
The program can also:
1...Provide a timeline of entry to service to show discrepancies
(It proved a myth about poor construction in the 1870s)
2...Create a vehicle list for any date between 1886-1960,
with reduced accuracy to 1985 (scrap data not added yet). Some
dates create a 'data drift' due to assumptions.
3...Create a table of activities: by activity and year for a class
4...Create a list of activities by location, by year
The input of photo has been halted due to a new
method of indexing - FTS.
File mstepm written by Peter J. Vincent, updated 3/3/2008