Maintaining the Church Database

Introduction

The church database is maintained on a county by county basis by the volunteers providing the individual county pages. At the county level adoption of mainatenance of that section of the database depends on the effort that the maintainer can offer along with their other commitments. The church database statistics provides information about the number of entries for each county and the state of development.

The information on this page is aimed at the county page developers and describes the files that they use and which are used to build the database. These files are not accessible on the web and the information in them is closely related to the database content but is not always in the same format that users see.

County sections

The database is split into county sections which are frequently maintained by the same person who mainatins that county section fo the Genuki pages. But we do have the ability now for another individual to separately maintain the church database entries for a county, although it is useful for them to liase with the county maintainer.

The data is maintained as a comma separated text file, for which the format is described later in this page. If you want to take on maintenance of a county section of the church database then please contact Phil Stringer to discuss it, and to obtain the initial data for that county.

We use a database table to hold information about how each county is maintained, see Phil Stringer for changes. The location where each devolved county section was kept, used to be in a specially named directory in each county, and if a churches.csv file appeared there, it was automatically used. This method is no longer used, each devolved county section has to be defined in the database before it is used. This has been done for all the existing devolved county sections of the church database. If you want to start maintaining your county section of the church database contact Phil who will set the field in the database table.

The options we have for the church database within this county table are:

Data structure

The main component of this system is a searchable database which contains basic details of the church e.g. it's name, where it was located, and when. This is sufficient to be able to dynamically generate a results page for the user. However at the next level users want to know further information such as what records exist and where to find them. This is not as easy to manage in a database and so there is the facility to permanently have a unique page for the church, which is displayed as a result of the search, and which can hold as much information as is required. The disadvantage of this approach though it is then not easy to see for which churches you have provided information about church records. So a facility has been developed to be able to supply church record information in a separate set of csv files. Collecting all the basic details first before moving to individual church pages significantly reduces the effort required to provide the information. Otherwise you end up having to make changes in two places, lots of cutting and pasting etc. which is a real pain.

churches.csv file format

The database is used to record details of all churches and places of worship used by any religious group. Regard church as a congregation rather than the building in which they worshipped. If more than one denomination use the same building then put in two entries. If they moved between buildings at different locations then again use two separate entries, with different locations and dates, but when you get to the later stage of producing a separate web page, then have both entries pointing to the same page. In addition municipal cemeteries not associated with a particular church can also be recorded here, as that is what is required when looking for burials.

The raw data is maintained as comma separated text files, with a separate file for each county. The recommended technique is to use Excel (or OpenOffice) to view and edit this file, but always saving it back in csv format. Excel lets you easily view the data in tabular format and to sort it as required. The most convenient order is probably Place, Denomination, Dedication. You can have a line of headings at the top of your columns and the program that builds the database will ignore any line in which the grid reference field contains the characters Gridref, so use that in any headings line.

The fields in churches.csv are as follows. The database upload program makes some minor adjustments to these fields and adds some more hidden ones based on your data for use in the search process. If you maintain your csv file using Excel, the following letters are those that Excel uses to identify the columns, but for Ireland and the Channel Islands you will have to mentally add one..

  1. The 3 character Chapman code for your county (upper case). This used by the search routines and provides links to the county page on search lists and to provide contact details.

    • The OS grid reference of the church [XXnnnnnn or XXnnnnnnnnnn (upper case)], for England, Scotland, Wales and the Isle of Man, The initial gridrefs are approximate ones, just giving the Km square on the map and so additional zeroes were added to convert it to the traditioanl 8 character gridref and the approximate flag set to Yes.

      The standard eight character grid references do not give a fine enough resolution when used with the sites such as streetmap.co.uk as sometimes the arrow on the map appears on the wrong side of the street. So there is also the option of providing a very exact gridref using twelve chars instead of the usual eight which are also still be supported. So SD371271 can now appear as SD3710027100.

      If you are adding a new church and don't know its location, then use the gridref for the place taken from the gazetteer. Adjust it so that it is easily recognisable as approximate [ XXnn0nn0 ] and also set the approximate field to Yes.

      A useful source of church gridrefs is "Dove's Guide" for bellringers.

    • Irish counties and the Channel Islands have an extra column, as we don't use grid references for these but use latitude and longitude. These are supplied each in their own fields as it is much easier to handle them this way. So these will be columns B (latitude) and C (longitude), and you will need to increment the letter by one for the rest of the columns.

  2. Yes or No indicating whether the location is an approximate one. For all the Yes/No fields the upload program is quite flexible as it just looks for an upper case Y anywhere within the field. If it finds one then it treats it as Yes, anything else including an empty field is treated as No.

    The code has used the length of the grid reference to determine whether we have a really precise location. i.e. 6 digits means nearby or within 100 yards, and 10 digits for a really exact one. But as Irish churches use lat/lon this means we don't have a method of defining a precise location for them. So to provide this facility a further option is now available in this field of P for a precise location. This will also work for those with grid references and will take precedence over the length of the grid reference.

  3. The place name.

  4. The URL for your town/parish page covering the above place name. This is used to provide the 'Up Arrow' link on final web page giving details of the church.

    This of course depends on the town/parish structure used for the corresponding county web pages. The staring point for that has been the civil parishes that existed in the early 19th century. This has worked reasonably well for alot of counties but less so for those where ther has been a lot of develoment in later years. For those additonal places have their own pages where this makes it look more sensible. For example in Lancashire it is broken down into townships rather than civil parishes due to the huge increase insize as a result of the industrial revolution. And large new towns e.g. Blackpool that have appeared since then.

    This field indicates which town/parish web page cover the area where this church is located, so the starting point is the 'parish list' in the wbe pages. Then to work ot which of the smaller places are covered by this town/parish page the information should hopefully be in the gazetteer. If that hasn't been done yet then another way to find this is via map links. From the web page for the church databse entry click the link for 'Vision of Britain' under Maps and then click 'Historical Units & statistics' and the links there give you the answer in most cases.

  5. The dedication (where there is one).

  6. The URL of a GENUKI page specifically for the church which is displayed instead of the generic page giving details of the church. For many counties this field is not being used and where it is blank a dynamically generated web page is displayed when a user asks for details of this church. The csv file and associated software though does not contain the detail that you may want to provide in the web pages. In thse cases you will need to abandon the dynamic pages and produce static ones which are displayed instead as a result of searches.

    Leave this blank until you need to create a unique page for the church. You will need to do this if you want to add in further information such as a description of it's history or where to find the records, in more detail. However I am contemplating a further database table in the future to describe where to find the records, transcripts, indexes etc.

  7. The name of the street where the church is located. Obviously this is there to help people find the church if they want to visit it, but if the church has no dedication then this is frequently used to refer to the church. E.g. compare "St.John's" or "St.Peter's" with "Park Street Methodist". If this is a rural area without a street name then you can enter a single space character. This prevents questions appearing in the Help area asking for a street name.

  8. A code used to group the denominations by type. This is only used internally for searching and in choosing some of the text in the web pages. The next field with the detailed denomination is always used in the displays provided to the user. For some paces of worship the word 'church' is not appropriate e.g. for Quakers 'Meeting House' is the appropriate word. This field is used when a choice has to be made by the code that generates the web pages.

    This generic code is used in searches to look for broad types of church e.g. Methodist instead of all the possible variations e.g. Wesleyan Methodist, Primitive Methodist etc. etc. The ONLY codes that can be used are those in the search menu on the churches web page i.e.

    ANG Anglican church (CofE, CinW etc.) For backward compatibility only. The upload program now converts all ANG values to EST, apart from Church of Scotland where it converts them all to PBY.
    EST Established church (CofE, CinW etc.) However be wary of calling anything 'Established church' as it isn't politically correct for some of them :-)
    BAPT Baptist.
    CEM A cemetery without an associated place of worship (ignoring chapels just used for funerals). For separate burial grounds use the CEM value rather than that for any associated denomination. Otherwsie the code that generates the pages would call it a church etc. rather than a cemetery. Use CEM for a Crematorium as well.
    CONG Congregational or Independent
    IND Independent For backward compatibility only. The upload program now converts all IND values to CONG.
    JEW Jewish
    METH Methodist
    PBY Presbyterian
    QUAKE Society of Friends (Quakers)
    RC Roman Catholic and any specials.
    UNIT Unitarian
    SEC Secession churches which is the usual term for those that seceded from the Church of Scotland. This includes Free Church of Scotland which my experts tell me that although it seceded later than the rest and was a large group, it is usually grouped as a secession church.
    EPIS Episcopal Church of Scotland
    OTHER Other
    We may find the need to add further major categories if the need arises.

    I used the denominations of Lancs churches that I already have in the database to produce this initial list. It obviously needs expanding especially for Scotland and anywhere else with different denominations. So please get back to me with suggested additions. Note that we are not trying to produce a list of every denomination, this is to sub-divide them in to similar groups to make it easier for users to find a church that may be useful from our database. So you need to see which types can be grouped together, and if there is just one or two of a denomination which doesn't warrant there own entry in this list, then just code them under 'OTHER'.

  9. The exact denomination e.g. Methodist New Connexion. This is what is shown on the web pages generated rather than the code in the field above. Many denominations have split and regrouped over the years so put in the one that you think will be most useful. This is likely to be the original denomination rather than the broader group when they recombine later. However don't take this too literally and put all the old Anglican churches down as Roman Catholic :-)

  10. The founding date of the church. If you don't know put 1. If a user knows when an event took place, this field can be used to exclude churches from the list presented to them.

    More information about dates.

    This field is used to construct another field internal to the final database which is used when searching by date.

  11. The closing date of the church. Use 9998 as the default value for new entries.

    More information about dates.

    This field is used to construct another field internal to the final database which is used when searching by date.

  12. The URL for a picture of the church. This is just for the image itself and is used on the final results page. E.g. click here and select 'Sacred Heart' for an example. This will be used in the src= parameter in a <img> tag.

    You can either supply a full url e.g. http://........./parish/xxxx.jpg or a shortened form */parish/xxxx.jpg The * is replaced by the county part of the url when the database is built.

    Please be aware of copyright and don't just grab any image. As there is no way of in this system of relating a web image to the originator, don't link to any image you find on somebody else's site.

    If you want to have multiple pictures, then this can be done if your county is at www.genuki.org.uk. This restriction is because the script looks for the existence of files. If the main picture is called xxxxx.jpg then it will look for others named xxxxx1.jpg, xxxxx2.jpg etc. and stop when the next in the sequence is not found. Style sheets are used to display these on the right of the screen. To facilitate this it also looks for small versions of the images to use as links, xxxxxnsm.jpg e.g. StJohn2sm.jpg. If the sm images do not exist it will use the full size ones, but force the width to be 200 pixels. This achieves the desired effect but makes the download time longer than it need be. The sm images should be 200 pixels wide for landscape images, and 150 pixels wide for portrait images.

    There are subsequent fields that can be used to display copyright information, and usage conditions etc. There is also a separate facility to include a separate section of text under the picture which is specific to the county where you can put any particular usage conditions. This text is common to all pictures for churches in that county. Held in /big/churchdb/conditions/xxx where xxx=Chapman code.

  13. The URL for the church's own web page. If you have done a search and can't find one, then enter a single N.

  14. Y or N indicating whether the church has ever had a graveyard. If present it generates some appropriate text under the Cemeteries heading and will also be used to enable searches for burial places.

  15. Y or N indicating whether the distinct page for the church has a Church History section. This is something only used within the database for generating Help messages. This field is only here so that there is central list of what has been added to individual web pages that have been written and avoids having to look at them individually.

  16. Y or N indicating whether the distinct page for the church has a Church Records section. Usage as in previous field.

  17. The name of the person holding the copyright to the picture whose URL is provided in the field above. This is used to put appropriate text under the picture that is displayed. If you don't want anything to appear, just leave this field blank.

    Another alternative where you have a group of church pictures with similar access conditions use a special code which is used to put specific text that you want including under the picture. E.g. If it is a picture of a church long ago demolished in an image library, you may have negotiated rights to link to it with additional text acknowledging the fact. The code used here must first be provided by Phil to relate to a specific file on the server which can hold whatever text is required, such as acknowledgement, usage rights, links to the library etc.

  18. Conditions. Anything else to do with copyright. The previous field has the name of the owner, this one has been used to hold an associated date where permission was granted to use a picture from an old newspaper.

    If you have used a picture form Geograph, put !GCC in this field and the Geograph Creative Commons text will appear beneath the picture. We also need the Geograph profile id of the person who took the picture. You will have already put their name in the previous field ans you can place the profile id number also in the previous field. Separate the two values with a slash character e.g. Doug Elliot/434.

  19. A flag field. The contents aren't used by anything within the database, but can be used as a search constraint. For Lancs this contains a flag for those churches published by the Lancashire Parish Register Society, and it is used to show a Google map of them.

  20. A unique numeric id for this entry within the county. If it is not supplied an id is dynamically generated by a database rebuild, but the dynamic id will change if churches are added or deleted. This overcomes the problem, but you need to ensure that this value is unique within the county. The dynamically chosen ids start are in range starting at 10000. The id is used in urls to show individual churches if one has not been specified in field G.
  21. Text that will be placed under the Church History heading. This can contain html for formatting, but cannot contain any line breaks or newline characters.
  22. Text that will be placed under the Church Records heading. This can contain html for formatting, but cannot contain any line breaks or newline characters.
  23. Text that will be placed under the Cemeteries heading. This can contain html for formatting, but cannot contain any line breaks or newline characters.
  24. Link to connect entries for churches that have moved location over time. Insert the Id of the most modern location. This has not been implemented yet.

Sources of information

There are many sources of information which can be used to identify churches that exist now or have existed in the past. The major difficulty though is that each source uses different ways of describing the location of the individual churches and there are a significant number that provide insufficient information. This can lead to duplicate entries that need an expert with local knowledge to identify and resolve. Many of the older sources were targetted at an audience that knew the information already and so an implied reference was enough. Nowadays that isn't common knowledge and research is required to resolve the conflicts. Some of the sources of conficts are:

The best starting point is usually the lists published by record offices detailing their holdings of parish registers. That will give you the basic infrastructure of the older churches. The areas that will be most deficient will be those churches that haven't deposited registers and many of the non-conformists will be in this group. N.B. Don't fall into the trap of using the earliest date in the deposited registers as the founding date of the church. You need to use other sources to check that so use the form <nnnn.

Then start looking at directories, and to start with one dated around 1900 should give most of the churches that are normally used for research. Later than that you will get s few more but lose more as especially non-conformists started to combine. You will probably find that the Anglican and larger/older churches of other faiths are covered reasonably well and more detail is given in the large towns. But for smaller places you may get something like. "There are also Wesleyan, and Primitive Methodist chapels".

Then of course there are likely to be websites that will help and are particularly useful for finding churches founded fairly recently. Try a search engine and look for "xxxx diocese" and that should locate details of Anglican and Catholic churches. The amount of information present will vary. Addresses may be those of the minister, dates may be of the church building, and watch out for joint benefices where some churches have had to close. For other churches using the old term of non-conformist the web information is not as useful as with thse there has tended to be a lot of joining back together and you see the present structure and names and it isn't easy matching that with the older individual churches.

Order of construction

Experience has shown that approaching the task of building up the database in a layered way minimises the amount of effort required.
  1. Add entries for churches in lists of deposited records, using approximate gridrefs from the gazetter.
  2. Use a gazetteer from around 1900 to add in the many without deposited records. When this has been done you should have a good framework containing a high proportion of the churches in your county.
  3. Then ask for people with local knowledge to give you the exact location of the church, whether it is still open, and the street on which it is located. Switch on the facility for the search pages to ask for help as this makes it easy see what is required in the lists of churches that the search pages produce.
  4. Only start building individual pages for each church once you have the basic information including a precise location as this will miminimise the number of changes that need to be made to them. The best way to create a page is simply to search for it on the web, and to save the page that is automatically generated. Only ask for further information such as where records are deposited when you are ready to create individual pages.

New/moved churches

Most churches have stayed on the same site over the years even though they may have occupied more than one building. But some have changed locations usuually because the congregation has grown to a size that is too large for the current church building. In these cases create a separte entry in the database for each location with appropriate closing and opening dates for each site. Then when you get to the stage of producing an individual web page for the church, make each entry point to the same page and describe the change in location in the Church History section. N.B. Don't create multiple entries for those cases where they knock down a church and erect a new one on the same site.

Viewing auto-generated pages when individual church pages are in use

If you do put individual web pages in place for individual churches before all the basic information is available, its is sometimes useful to be able to compare the page with that which would have been automatically. E.g. to cut and paste. To do this search and get a list of links up including the church in which you are interested. Click 'Nearby Churches' at the top of the page to give you a URL for the list of churches. Add another parameter ,AGP=Yto the end of the URL and that will give a list with all links to auto-generated pages.