Re: TODO for 2.3 branch

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

Re: TODO for 2.3 branch

Jerome send me the following mail:

> Hi,
> Most of my notes are shared ... on an existing source ;)
> But what happen if you load a GEDCOM with shared note ?
>>   n  @<XREF:NOTE>@ NOTE <SUBMITTER_TEXT>  {1:1}
>>     +1 [ CONC | CONT] <SUBMITTER_TEXT>  {0:M}
>>     +1 <<SOURCE_CITATION>>  {0:M}
>>       +2 TYPE <USER_REFERENCE_TYPE>  {0:1}
>>     +1 RIN <AUTOMATED_RECORD_ID>  {0:1}
>>     +1 <<CHANGE_DATE>>  {0:1}
> Maybe GRAMPS should import this note as new source with note ! Is it
> the actual way ? If not, why not adding this on TODO for 2.3 branch ?

Looking at the GEDCOM, it allows for a note structure to refer to an
which is then the above record.
However, GEDCOM apparently allows to add to NOTE_STRUCTURE an XREF:SOUR
or to a
note record a source citation. Again, GEDCOM is very dual: they say:
"SOURce tags subordinate to the note structure can only contain a
pointer to an
associated source record. Any non-structured reference to a source from a note
structure can be included as part of the note text itself."
But then in the note record, they put an entire source citation structure.

So, this means GEDCOM can have independent note records with their own source
citation! Also, shared notes exist in GEDCOM as Jerome correctly notices.
So I suppose they mean, you have a note, and the note is made based on
a source,
which it links in the note structure.

Having notes link to sources would complicate things a lot. For
starters, notes
would need a source tab. It looks overly complicated to me and bad design.

I would suggest we try to support the following workflow, and only the

1/notes are never shared. They are part of an object. Independent notes are
visible in the noteview and are for familytree notes or for use in the reports
(eg an introduction, footnote, ...). Notes that need to be shared, should be
notes linked to a source. As a source can be linked from almost everywhere.
The reason: make a clear distinction in the user interface between a
genealogical source, and a note. If both can do the same thing (which they can
in GEDCOM), users will mix them up.

2/A person creating a note based on sources, is actually making what is
called a
tertiary source. So, the person should not make a note which link to
the sources
he uses, he should make a new source and add the note to the source or to the
sourceref object.
We could in the future add to the source the possibility to link it to other
sources (indicating which source this source is constructed from), but
personally I think this is a bit overkill. Not many users would have sources
which refer to other sources, and even if they do, they could add a note of
type bibliography to the source object.

3/On import of an above GEDCOM structure we should not complicate things in my
opinion. We can have in GRAMPS:
an object --> notes
          --> sourceref --- source
          --> sourceref --- source --> notes
So we support notes with an object, or notes derived from a single source
while in GEDCOM it is apparently possible to have
an object --> note --> sources
          --> noteref --> sources
                |-> note --> sourcerefs -- source
          --> sourcerefs -- source --> notes
                |-> note
That is complicated!!!

So we could import a GEDCOM object with a note that has ONE source as a note
with the sourceref object between this source and the object. If there are
MULTIPLE sources, the note should be a source, with a note with the text and a
note with the bibliography (all used sources). Importing shared notes is not
possible however, or they must be made a source with a note ...
Ok, the GEDCOM approach is defenitely too complicated to explain to users, and
would make the user interface very confusing. I would stick with don't share
notes, and only import from GEDCOM what we think is feasible, and as best as

I have been working on giving all notes different types (place note, person
note, ...) in 2.3, so on upgrade the noteview gives info in the type field
which can be used to filter the notes.
The original idea was not to show unconnected notes in the noteview. With the
notetype having more meaning, that might no longer be needed, as you can
normally quickly filter them based on notetype, however, there is no
possibility at the moment to know where a note is used, so the usability of
showing connected notes in the noteview is low.


This message was sent using IMP, the Internet Messaging Program.

This email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
Gramps-devel mailing list
[hidden email]