TMG-L ArchivesArchiver > TMG > 2004-10 > 1097509690
From: Gordon Banks <>
Subject: Re: [TMG] RE: Age calculator & irregular dates
Date: Mon, 11 Oct 2004 08:48:10 -0700
References: <200410080150.i981oaA9028393@mail.rootsweb.com> <email@example.com> <firstname.lastname@example.org>
I kludged the problem by creating roles in the birth tag for those which
have irregular dates so that the role's sentence uses the memo for the
date it puts in reports and then I use the date field for the date I
need to calculate ages.
On Fri, 2004-10-08 at 15:04, Darrell Martin wrote:
> At 10:53 AM 10/8/04, Gordon Banks wrote:
> >The sort date does too have genealogical meaning if you put it there.
> >Since you enter the data in both fields, you are the one who determines
> >what has genealogical meaning. When TMG is unable to process a date
> >(e.g. shortly before 8 May 1275, Spring 1150) you get ages like -571
> >years or something absurd. If when there was an irregular date it used
> >the sort date for the event, then you could manually force the correct
> >age by putting a date in the sort date that TMG could recognize (e.g. 8
> >May 1275, Apr 1150). Otherwise, using accurate genealogical information
> >disables the age function in TMG. It doesn't have to be that way.
> Hi, Gordon:
> It is made explicit in descriptions of the TMG data design that the sort date is used *ONLY* for sorting Tags. It is also explicit that the sort date will "never" be output for any reason. That sort dates are not used to calculate ages is consistent with the stated intent of the design. If such a calculated age were put on a display or into print, then the sort date would -- even if only indirectly -- be "output" in violation of the design.
> I will not argue that TMG's date math is as good as it could be. I have in fact been one of the harsher critics of some aspects of that feature. But agreeing that there is need for a change is not the same as agreeing that the sort date should be part of the solution. In some ways, incomprehensible ages like "-571 years" may actually be *preferable* to ages that seem reasonable on their face, but in reality are also wrong.
> I believe that if it is possible to conclude from the evidence what a date is, with enough certainty to calculate an age from it, then one should put that conclusion in the date field of the Tag *as a regular date* and put the date expressed in other ways ("after his mother's death", "shortly before 8 May 1275", "Spring 1150", or "six weeks before his sister was born") into the Tag Memo or into the citation of the evidence, depending on the nature of the specific date in question.
> A problem remains for regular but not "full" date forms, from which TMG still does not calculate ages. I gave a rather detailed suggested solution to that problem, on TMG-L, a few weeks back as I recall.
> Darrell A. Martin
> a native Vermonter currently in exile in Illinois
> ==== TMG Mailing List ====
> Send all messages and replies to <>.