Doug McDonald wrote: > Bob Velke wrote: > > You have people, and you have that relationship. That's > what genealogy is. Then you can accrete tons of data on > top of it. No! You have *evidence* - parish registers, civil registers, census returns, wills, diaries, newspapers, all manner of things. And then you can accrete hypotheses on top of it. If you want a simple data model let me offer you the humble 6x4 index card. Extract the data from evidence, e.g. Walter Dearnley father of William in baptism 1760 William Dearnley son of Walter in baptism 1760 Walter Dearnally father of John in baptism 1762 John Dearnally son of Walter Deanally in baptism in 1762 Walter Darnley father of Walter in baptism in 1763 Walter Darnley son of Walter in baptism 1763 Walter Dearnely father of Walter in 1740 Walter Dearnely son of Walter in 1740 and so on. This is *analysis* The Dearnley surname, as I've found out has many spelling variations - but this way we've preserved them all. Now let's construct a hypothesis - all the cards for Walter as father in the 1760s together with Walter the son in 1740 all refer to the same person. So, in addition to the humble 6x4 card we take the humble paper clip and clip then together. If we like we could take another card and write our preferred spelling on the front, something like Walter Dearnley father of family in the 1760s. This is *reconstruction*. Note that our analysis records are distinct from our reconstruction record. Is this important? It surely is. If we then find another item of evidence, the baptism of Walter son of William Darnley in 1741 we realise that part of our reconstruction is probably wrong. So we unclip the Walter Darnley as father in 1763 from the Dearnley bunch and clip it onto the Walter Darnley as son in 1741 card. Then we discover the burial of Walter son of William Darnley in 1742. So we can unclip our Walter as father card again and wonder what to do with it - were we right first time? - did William have a later son Walter? Maybe we just leave it on its own for the time being until we have more evidence. The 6x4 card is humble but it's powerful because we have flexibility how we use it. That would be my starting point for a data model. Does anyone know of a software package which offers that flexibility? Or, in any other way, separates analytical from reconstruction data? Or starts with the premise that any hypothesis could be wrong and that, therefore we must be able to unpick our identifications as easily as we make them? -- Ian Hotmail is for spammers. Real mail address is igoddard at nildram co uk

Ian Goddard wrote: > Does anyone know of a software package which offers that flexibility? ? For all its flaws, GEDCOM is still more flexible than most of the software I've tried. -- Wes Groleau Answer not a fool according to his folly, lest thou also be like unto him. Answer a fool according to his folly, lest he be wise according to his own conceit. -- Solomon Are you saying there's no good way to answer a fool? -- Groleau