🚀 go-pugleaf

Steve Hayes wrote:
> On Thu, 29 Jul 2010 21:08:02 +0100, Ian Goddard <goddai01@hotmail.co.uk>
> wrote:
> 
>> I use GRAMPS.  If I add a document as a source against an event or a 
>> person (and the same source can be linked to as many events and people 
>> as needed) it will automatically be allocated an ID (by default the 
>> letter S for source) and a serial.  On the source form there is a Data 
>> tab.  If I click on that I can add a data item.  That offers me the 
>> option of entering a key/value pair.  I could enter "Location" as the 
>> key and the filing reference as the value.  If I were subsequently to 
>> reorganise the filing I could go to each source as it gets refiled and 
>> update the location.
> 
> But I generally try to keep unrelated people out of my main genealogy file,
> and I don't want to create person records for them there. But if I eventually
> find proof of relationship, I want to be able to find the documents where I've
> seen their names before. 

I know the feeling!  Because many of my lines belong to families 
resident in the area for many generations I have a lot of problems just 
working out who's who.

One part of the solution is to get as much as possible of the raw data - 
PRs & census data - into an RDBMS.  As a good deal of it is available as 
PDF & pushing data around like this is part of what I used to do as a 
living this is straightforward and enables me to grab bunches of records 
with SQL.

Another part is to use a spreadsheet to get an overview.  The steps are:

1.  Get all the baptismal records for the name and area under 
consideration into a spreadsheet.  If they've been exported from the 
RDBMS this is easy, otherwise it involves a lot of typing :(

2.  There may be scope for a bit of reconnaissance by ordering by, say, 
father's name, mother's name and date, father's name, home and date or 
father's name, occupation and date to get an idea of what families might 
be there.

3.  Order by date and then add a few blank rows at the top.

4.  Add an extra column at the start.  The original data is now in 
columns B to H or thereabouts.  Enter the earliest year in A1 and  =A1+1 
into A2.  Copy A2 into the following rows of the A column until you 
reach the latest year.

5.  Select the subrow B to H (or whatever) of the first baptism & drag 
it up to its appropriate year.  If the next baptism is of the same 
family (father's name and mother's name, home, father's occupation or 
whatever's available) drag it up to its year.  If it's a different 
family drag it across to the next block of columns from I or whatever at 
the appropriate year.

6.  Continue this process using as many blocks of columns as needed. 
When it becomes clear that there are no more children in the first 
family you'd allocated to the first block you can put another family 
underneath.  As families become complete delimit them with a border.

7.  As the workspace gets wider you will probably need to replicate the 
year column.  It may also help to shade every 10th row with a background 
colour.  You can also save width by eliminating the surname column as 
this is redundant.

8.  Attach notes to the start of a family block showing the candidate 
marriages.

9.  If a child's name is repeated check for burials and attach a note if 
you find one.  If you don't you may need to review whether this is the 
same family.

10. As the worksheet builds you can start working out which children of 
one generation are the fathers of the next.

11. When the relevant families have been identified by this means they 
can be entered into the genealogy program.

A spreadsheet isn't ideal.  I'd like something which would treat the 
subrows as single objects which would make the selection and dragging 
easier.  Ideally a genealogy program would incorporate this.

As I usually do this on a single parish or even an area within a parish 
there's no need to record the source on a record-by-record basis as 
they're all the same.

The last part of the solution is to set up a separate Gramps database 
for any batch of data I want to keep separate such as gedcoms downloaded 
from IGI for some of the non-conformist chapels who aren't in the RDBMS 
derived from the PRs.  I can export selections of these as XML & 
incorporate them into the relevant main database as needed.  Recently 
I've also set up a Gramps database containing nothing but locations to 
act as a gazetteer to be incorporated into the other databases so I have 
consistency in that respect.

> 
> I've been wanting to try it for some time, but after unsuccessfully trying to
> install Linux for 11 years, I doubt that I ever will. 
> 

Hmm.  I'm not sure I can advise much on dual boot Windows/Linux.  If I 
need to run Windows it's run as a guest in a VirtualBox environment 
hosted on Linux.  You could try the reverse.  Also I believe the main 
Linux distributions are pretty good at installing themselves alongside 
an existing Windows installation providing the disk resources can be 
made available.  However you may want to look at Wubi ( 
http://wubi-installer.org/ ) which says it installs Ubuntu on a Windows 
machine as a dual boot without repartitioning disks; it uses Windows 
files as its disk space.

You can, however, install Gramps on Windows directly.  I have tested 
gramps4win: 
http://sourceforge.net/apps/mediawiki/gramps4win/index.php?title=Main_Page

Although it claims to be a beta with appropriate warnings in fact it 
installs a normal Gramps release although that release is no longer 
current.  It will, however, let you check out Gramps in a Windows 
environment.

-- 
Ian

The Hotmail address is my spam-bin.  Real mail address is iang
at austonley org uk