RootsWeb.com Mailing Lists
Total: 3/3
    1. Re: Data model for civil registration
    2. Richard Smith
    3. On 08/02/13 11:16, Ian Goddard wrote: > I use Gramps from time to time & this enables me to use a range in the > same way as yourself. It's possible to update this to a precise date it > I feel a certificate is necessary. As an estimate of DoB it's good > enough to evaluate alternative candidates and is sometimes better than > baptism which can be out by a few years, or more when the register > includes the note "baptism of an adult Quaker". That's interesting, because I had been going to use Gramps as an example of a piece of software that doesn't make it easy to express what I want. I can certainly add a birth event for 'calculated between 1866-11-20 and 1867-03-31' (or whatever I want). But that's not what he source says. Or I can add a custom 'birth registration' event for 'between 1867-01-01 and 1867-03-31', but custom events seem to be rather second-rate citizens in the Gramps universe. For example, if I have a baptism date, but no date of birth, in many contexts Gramps will use the baptism date as an approximation of the date of birth; but I can't see a way of making it do that with a custom 'birth registration event'. So far as I can see, this is a common problem in many programs: the built-in event types have certain special properties that can't easily be replicated with custom events, and inevitably at times the built-in events don't include everything you want. What I think I'd like is the ability to say that one event type typically follows shortly after another event type, and in the absence of information on the latter event, it would be a useful proxy. For example, in an index, I might want to distinguish a John Smith, bp 1590 in Ripon Cathedral, from a John Smith whose birth was registered in Southampton in 1905. Although I might not know where or when either was born, for the purposes of a index, those dates and places are quite sufficient. Richard

    02/08/2013 05:52:24
    1. Re: Data model for civil registration
    2. Ian Goddard
    3. Richard Smith wrote: > On 08/02/13 11:16, Ian Goddard wrote: > >> I use Gramps from time to time & this enables me to use a range in the >> same way as yourself. It's possible to update this to a precise date it >> I feel a certificate is necessary. As an estimate of DoB it's good >> enough to evaluate alternative candidates and is sometimes better than >> baptism which can be out by a few years, or more when the register >> includes the note "baptism of an adult Quaker". > > That's interesting, because I had been going to use Gramps as an example > of a piece of software that doesn't make it easy to express what I want. > I can certainly add a birth event for 'calculated between 1866-11-20 > and 1867-03-31' (or whatever I want). But that's not what he source says. What I do is use Regular - Range. I count that as good enough unless I have a more specific date. > Or I can add a custom 'birth registration' event for 'between 1867-01-01 > and 1867-03-31', but custom events seem to be rather second-rate > citizens in the Gramps universe. For example, if I have a baptism date, > but no date of birth, in many contexts Gramps will use the baptism date > as an approximation of the date of birth; but I can't see a way of > making it do that with a custom 'birth registration event'. > > So far as I can see, this is a common problem in many programs: the > built-in event types have certain special properties that can't easily > be replicated with custom events, and inevitably at times the built-in > events don't include everything you want. > > What I think I'd like is the ability to say that one event type > typically follows shortly after another event type, and in the absence > of information on the latter event, it would be a useful proxy. For > example, in an index, I might want to distinguish a John Smith, bp 1590 > in Ripon Cathedral, from a John Smith whose birth was registered in > Southampton in 1905. Although I might not know where or when either was > born, for the purposes of a index, those dates and places are quite > sufficient. What I do where all I have is a baptism is enter the baptism with the appropriate church or chapel as place and, if the father is described as "of wherever" and there's no indication that the birth wasn't at home I enter a birth date of "before" the baptism date (ideally it would be "on or before" but Gramps doesn't have that option) with the "wherever" bit as place. It's never completely right, of course, because I'm sure that some baptisms were carried out at home where a child wasn't expected to live. -- Ian The Hotmail address is my spam-bin. Real mail address is iang at austonley org uk

    02/08/2013 09:16:20
    1. Re: Data model for civil registration
    2. Martin Steer
    3. On Fri, Feb 08, 2013 at 12:52:24PM +0000, Richard Smith wrote: > >Or I can add a custom 'birth registration' event for 'between 1867-01-01 >and 1867-03-31', but custom events seem to be rather second-rate >citizens in the Gramps universe. For example, if I have a baptism date, >but no date of birth, in many contexts Gramps will use the baptism date >as an approximation of the date of birth; but I can't see a way of >making it do that with a custom 'birth registration event'. Gramps has functions to select birth_or_baptism and death_or_burial. You could hack them, probably, to add extra events. One problem with this kind of approach is that an estimated birth date, say, will obscure a well-founded date of baptism. Martin

    02/08/2013 06:28:59