2010/7/31 Doug Blank <doug.blank@gmail.com>

For me, I wish that there was a merge that was smart enough to do the
right thing most of the time. For example, we know that there is only
one birth and death, so there is no reason to duplicate those. But if
it had to ask a question of me every once in a while, that would be
fine too.

It depends on what you use Gramps for.  If you use it to hold what you know or think is true, deduced from research you have done elsewhere, then there is obviously just one birth.  If you use it to support research, then you need to hold several birth events, because sources do not always agree.  I have 5 different dates for the marriage of a couple and I still don't know which one is right, because I haven't got the actual certificate yet. You need to keep all with the complete source references, until you find out which one is true or more likely.  Even then, you may want to keep the others to make sure you have recorded why you think this or that source got it wrong.  What I do is mark private those events I think are wrong.  This contain data I want to keep, but I don't want to publish.  At nice thing would be a trust metric for data, something like "pretty much true", through "likely", down to "known falsehood".  Quality levels on source references may be it, though.

A different thing, of course, is the synthetic birth events derived from "age at".  I am not sure.  If they exist, there is the need to get rid of them easily.  They are wrong remarkably often.  People lie or don't remember.  And sometimes they don't even know.  For example, when death certificates are written the one who knows best is not around anymore and those giving the data do not know exactly either, so the age is off by a couple of years very often.

Regards,

Julio


 
-Doug

> Michiel
>
>
> Doug Blank wrote:
>>
>> On Thu, Jul 29, 2010 at 4:38 PM, Michiel Nauta <m.d.nauta@hetnet.nl>
>> wrote:
>>>
>>> Hi,
>>>
>>> Those living on the bleeding edge will have noticed that trunk contains
>>> a new implementation of "merge". I hope it will see some testing. To
>>> facilitate that, I put up some scripts on my homepage that give access
>>> to some (a couple of million records) data in GrampsXML format. Detailed
>>> instructions and scripts are available from
>>>
>>> http://members.tele2.nl/m.d.nauta/typeless_data_entry/typeless_data_entry.html
>>
>> I've been testing the merge and it seems to be working very nicely.
>> Thanks for the work on that!
>>
>> I wonder if there isn't a couple of things we could add to Gramps to
>> make the typeless data process even smoother? But Gramps Exhibit is
>> very cool.
>>
>> Do you need Gramps Exhibit to go from the websites into Gramps? I
>> noticed that you didn't have a link to it on your page. It wasn't
>> clear what role Gramps Exhibit played in the process. I'd love to see
>> more details on the process of going from web data to gramps xml.
>>
>> Thanks!
>>
>> -Doug
>>
>>> Michiel
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> The Palm PDK Hot Apps Program offers developers who use the
>>> Plug-In Development Kit to bring their C/C++ apps to Palm for a share
>>> of $1 Million in cash or HP Products. Visit us here for more details:
>>> http://p.sf.net/sfu/dev2dev-palm
>>> _______________________________________________
>>> Gramps-devel mailing list
>>> Gramps-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/gramps-devel
>>>
>>
>

------------------------------------------------------------------------------
The Palm PDK Hot Apps Program offers developers who use the
Plug-In Development Kit to bring their C/C++ apps to Palm for a share
of $1 Million in cash or HP Products. Visit us here for more details:
http://p.sf.net/sfu/dev2dev-palm
_______________________________________________
Gramps-devel mailing list
Gramps-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/gramps-devel