I try to understand how the religion of a person is represented in
GEDCOM 5.5. I came across the RELI token, but could not find any
standardized values for this token. In some GEDCOM files I find
localized descriptions of the religion, but it seems to me that such a
localized version is not suitable if GEDCOM files are exchanged across
language boundaries.
e.g. how would I represent the fact that a person is "roman-catholic"
RELI roman-catholic
RELI römisch-katholisch
RELI roemisch-katholisch
RELI RK
RELI RC
I think that for the purpose of exchanging GEDCOM files some sort of
language independent taxonomy for these tokens would be required.
Does anybody know of such a taxonomy?
Best regards
Carsten Leue
I have never thought of this as interesting
the religion is pretty obvious from the location of baptism or
christening, confirmation, marriage and burial
otherwise I am using notes more and more becasue they display better
when uploaded to world connect for example
Hugh W
--
a wonderful artist in Denmark
http://www.ingerlisekristoffersen.dk/
Beta blogger
http://snaps4.blogspot.com/ photographs and walks
old blogger GENEALOGE
http://hughw36.blogspot.com/ MAIN BLOG
It's not just the RELI field, but just about every field in GEDCOM that
lacks standardised representations (even in one language), let alone
several. Places, date formats etc all vary between software program and
personal usage.
Kerry
I guess the main issue is that a missing taxonomy complicates the
exchange of GEDCOM files significantly, due to different ways to
express the same information and most importantly across language
boundaries. A genealogical software could translate standardized
tokens that identify the religion and display them in the desired
locale to the end user. With "free text" this is not possble.
Yes, that's unfortunately a significant weakness of the format.
Do you know of any taxonomy that has evolved as "common usage pattern"
although not standardized?
>> [6 quoted lines suppressed]
>
> It's not just the RELI field, but just about every field in GEDCOM that
> lacks standardised representations (even in one language), let alone
> several. Places, date formats etc all vary between software program and
> personal usage.
>
> Kerry
After all GEDCOM is a standard for exchanging information between
computers, it wasn't intended to enforce any form of standardisation on the
data being exchanged.
DATE however is interesting, in one part of the standard it is defined as :
DATE {DATE}: =
The time of an event in a calendar format.
It then goes on to add various "Subtypes"
DATE_PHRASE: = {Size=1:35}
(<TEXT>)
Any statement offered as a date when the year is not recognizable to a date
parser, but which gives information about when an event occurred. The date
phrase is enclosed in matching parentheses.
and
DATE_RANGE: = {Size=8:35}
[
BEF <DATE> |
AFT <DATE> |
BET <DATE> AND <DATE>
I think a lot of software ignores these sub-types and just assigns what the
user enters as a date(with any local checking) to DATE. Examining one
Gedcom at random I find
1 BIRT
2 DATE ABT 1838
2 PLAC IRELAND
which I think should have strictly been
1 BIRT
2 DATE
3 DATE_RANGE ABT 1838
2 PLAC IRELAND
or possibly
1 BIRT
2 DATE
3 DATE_PHRASE (ABT 1838)
2 PLAC IRELAND
and in the same Gedcom
1 DEAT
2 DATE 10 May 1984
2 PLAC LONDON ENGLAND
2 CAUS HEART ATTACK
1 BURI
2 DATE BURNLEY LANCS
Which is pure nonsense but shows that the software didn't do any elementary
checking.
Incidentally unless dates are escaped to another calendar GREGORIAN is
assumed and the format IS defined as :
DATE_GREG: = {Size=4:35}
[ <YEAR_GREG> | <MONTH> <YEAR_GREG> | <DAY> <MONTH> <YEAR_GREG> ]
Although /presentation/ may differ with different software / locales.
The point is that even when the gedcom standard does allow for choices of
values it is up to the software being used to validate and make sense of
the data. Even if your software prevented you from entering "Jedi Knight"
into the RELI field, you can't assume you would never receive a record with
that text in it. If you did would you want your software to reject it ?
BobC
--
Posted via a free Usenet account from http://www.teranews.com
I rather suspect that the lack of a standardized format or
of a definition in the GED standards suggests that the issue
is not one of importance to many people in the US. And GED
was a child of the US and of the LDS Church.
As with anything else, if it's important to you, put it in
the NOTES/more about sections, because those sections tend
to GED in and out fairly cleanly.
IMO.
Cheryl
Bob,
The form as given in the Gedcom you found is correct. This is an
approximate date, not a date range or a date period. There's no such tags
as DATE_RANGE, DATE_PERIOD or DATE_PHRASE anyway. A date range is used when
you know an event happened at a particular time, but all you know is that it
occurred some time between two dates. For example, a person is listed as
age 30-40 in the 1840 Census. His birth would be:
2 DATE BET 1800 AND 1810
A date period is used to describe a condition that lasted for some time, for
example military service:
2 DATE FROM MAR 1863 TO JUL 1864
> and in the same Gedcom
>
> 1 DEAT
> 2 DATE 10 May 1984
(I cut off the rest) Dates in a Gedcom are usually seen capitalized.
Nevertheless this one is still technically correct since capitalization is
ignored in the line value and May has three letters. However be aware that
the month in a date must be written as a three-letter code, so
2 DATE 10 January 1984
would have been illegal.
Any date description that does not fit the patterns listed in the Gedcom
spec may be used in a date phrase:
2 DATE (Third Sunday in Lent)
Having spent a decade or so doing IT standards in ISO and other
organisations, it might be the exact opposite -- too many people thought it
was important but they all wanted something different and no consensus could
be reached.
Can you imagine getting any group of people to agree on the "complete list
of religions/occupations/whatever)" in order to assign codes to them? Before
you know it, you'd have the butcher, the baker and the candlestick maker,
closely followed by the sausage maker, the meat smoker, the organic butcher,
the kosher butcher, the halil butcher, the vegetarian butcher, the vegan
butcher, the organic vegetarian butcher, the free range organic vegetarian
kosher butcher, ...
Kerry
(G) Point well-taken. That pretty much sums up what
happened to the "standard" part of GEDCOM Standard 5.5,
doesn't it? ;) Ahhh, yes, those were the arguments!
Cheryl
It is even worse if you want to standardize on the language that the
data is in, or have some sort of translatable keyword. The system in
place permits the exchange of data that is not clobbered by some machine
interpretation. It is up to the user to read the data. I think that is
the way it should work.
Perhaps a program could be written to translate or standardize the data
from a GEDCOM database to a form that the user might want but otherwise
it should not be in the standards committee approach.
Dale
--
_ _ Dale DePriest
/`) _ // http://users.cwnet.com/dalede
o/_/ (_(_X_(` For GPS and GPS/PDAs
Bill,
Thank you for your correction - I have now re-read the spec and can better
understand that part of it.
I did get rather away from the OPs question which was really to do with a
taxonomy rather than Gedcom. The point I was trying to make (obviously
rather badly) was that you could get rubbish into the Gedcom (such as the
"DATE BURNLEY LANCS" example I quoted) and it was up to the user to check
the data for consistency. Also that some software can't accept as input
any of the valid date formats other than DD MMM YYYY / MMM YYYY / YYYY.
Thanks BobC
(G) What was that Church Flip Wilson used to do a skit on?
Was it in there?
Cheryl
And there were over 12000 Australians who listed their religion as Jedi in
the 2001 census.
Kerry
Mea culpa. Apparently there were over 70 000 Jedis in Australia in 2001.
There is NO particular representation ...
After the tag RELI, you put the value which you want: Chrétien, Moslem,
Atheist, Boudiw...
The receiving program then has to read: Chrétien, Moslem, Atheist,
Boudiw ...
Only a set of letters (characters).
It belongs to the user to give them a contents !!!
Regards,
--
Téhenne Saint-Denis de la Réunion
Logiciel de généalogie ohmiGene (Mac & PC): http://www.nauze.com/
Comparatif Import-Export Gedcom : http://www.nauze.com/gedcom/
Digest du format Gedcom : http://www.nauze.com/gmc/indexGED.html
> or possibly
>
> 1 BIRT
> 2 DATE
> 3 DATE_PHRASE (ABT 1838)
> 2 PLAC IRELAND
NO, the right writing is :
1 BIRT
2 DATE ABT 1838
2 PLAC IRELAND
or, less good
1 BIRT
2 DATE (ABOUT 1838)
2 PLAC IRELAND