Genealogy Software

Genealogy Software

[Larry Hazel]

Does anyone know of any genealogy software written in Ada?  With source
available?  I thought I might attempt to write some.  Any suggestions for data
representation will be appreciated.  I haven't done much Ada 95, but a lot of
Ada 83.  Will need some way to represent dates going back hundreds of years
(maybe just a fixed length string since some of them will be of the form "ca.
1784").  Will need a person object (maybe abstract with separate child types for
male and female).  Will link to other person objects which represent spouses,
children, and probably backward to parents.  An event object (marriage, birth,
death, buying or selling property, ...).  Bibliographical text will also be
needed.  Storage of scanned pictures and documents.

Maybe this is too complicated for a 1 man home project.

Larry

Re: Genealogy Software

[Wes Groleau]

Larry Hazel wrote:
> Does anyone know of any genealogy software written in Ada?  With source

I have entertained thoughts of writing one with
some similarities to LifeLines.

Having a data structure in memory loosely based on GEDCOM,
and storing them as actual GEDCOM fragments.  This would
make creating a GEDCOM file as simple as

  cat header data/* trailer

Some of the other implementation details are in my head.

I haven't had the time to write anything down.
Interested in collaboration?

-- 
Wes Groleau
http://freepages.rootsweb.com/~wgroleau

Re: Genealogy Software

[Larry Kilgallen]

In article <3BD498DF.27E8B66C@otelco.net>, Larry Hazel <lhhazel@otelco.net> writes:
> Does anyone know of any genealogy software written in Ada?  With source
> available?  I thought I might attempt to write some.  Any suggestions for data
> representation will be appreciated.  I haven't done much Ada 95, but a lot of
> Ada 83.  Will need some way to represent dates going back hundreds of years
> (maybe just a fixed length string since some of them will be of the form "ca.
> 1784").  Will need a person object (maybe abstract with separate child types for
> male and female).  Will link to other person objects which represent spouses,
> children, and probably backward to parents.  An event object (marriage, birth,
> death, buying or selling property, ...).  Bibliographical text will also be
> needed.  Storage of scanned pictures and documents.
> 
> Maybe this is too complicated for a 1 man home project.

It is complicated, but in this problem domain I do not think that
means it is better suited to a larger development organization.
Discipline is the main requirement, and to some extent that may
be better achieved by an individual (who is properly motivated).

There are a lot of details to consider:

	You need to be able to express _multiple_ birth dates,
	when there are conflicting data from various sources.
	There must be a capability to tie each data element
	back to a source document.  There must be an ability
	to specify not only January 11-16, 1896 but also
	January 11 _or_ 16, 1896 for cases where the source
	document clearly is one or the other but which of
	the two is not clear.

	For marriages, some people marry each other multiple
	times (that is, the same partners changing their minds).

	I had a great uncle whose wife died, after which
	he married her sister.  There were children from both
	marriages.  That sort of arrangement formerly was quite
	common in the US.  Obviously you can get that correct
	in the computer -- the challenge is making the user
	interface clear to the humans.

I would strongly urge you not to consider writing such software
until you have tried some that already exists.  PAF was only $ 35
the last time I checked, but it really is not necessary to buy
such software for a peek.  Just find friendly genealogy fans to
"show you" their family records and you will get to see the
existing software in action.

Hang around the genealogy newsgroups/chat boards to learn areas of
satisfaction and dissatisfaction with existing software.

The last I knew, GEDCOM was still the standard interface format.
It has a real specification, available from the Church of Latter
Day Saints (Mormons) in the US.  As part of their theology that
church has done lots of genealogy work worldwide, and my wife has
found they are quite open to sharing that information with any
non-believers who are interested.  The only hitch is that film
must be viewed on their property, since that is a condition of
the agreements under which various town clerks let them film the
records (before there were computers).

Re: Genealogy Software

[Bruce or Tracy Jacobs]

I wrote some that would convert Gedcom into a book format in HTML.  It
is a royal
headache.  Lots of stuff to consider and interpret and keep up with. 
Cross-ref'ing
is probably the most difficult part.

Larry Kilgallen wrote:
> 
> In article <3BD498DF.27E8B66C@otelco.net>, Larry Hazel <lhhazel@otelco.net> writes:
> > Does anyone know of any genealogy software written in Ada?  With source
> > available?  I thought I might attempt to write some.  Any suggestions for data
> > representation will be appreciated.  I haven't done much Ada 95, but a lot of
> > Ada 83.  Will need some way to represent dates going back hundreds of years
> > (maybe just a fixed length string since some of them will be of the form "ca.
> > 1784").  Will need a person object (maybe abstract with separate child types for
> > male and female).  Will link to other person objects which represent spouses,
> > children, and probably backward to parents.  An event object (marriage, birth,
> > death, buying or selling property, ...).  Bibliographical text will also be
> > needed.  Storage of scanned pictures and documents.
> >
> > Maybe this is too complicated for a 1 man home project.
> 
> It is complicated, but in this problem domain I do not think that
> means it is better suited to a larger development organization.
> Discipline is the main requirement, and to some extent that may
> be better achieved by an individual (who is properly motivated).
> 
> There are a lot of details to consider:
> 
>         You need to be able to express _multiple_ birth dates,
>         when there are conflicting data from various sources.
>         There must be a capability to tie each data element
>         back to a source document.  There must be an ability
>         to specify not only January 11-16, 1896 but also
>         January 11 _or_ 16, 1896 for cases where the source
>         document clearly is one or the other but which of
>         the two is not clear.
> 
>         For marriages, some people marry each other multiple
>         times (that is, the same partners changing their minds).
> 
>         I had a great uncle whose wife died, after which
>         he married her sister.  There were children from both
>         marriages.  That sort of arrangement formerly was quite
>         common in the US.  Obviously you can get that correct
>         in the computer -- the challenge is making the user
>         interface clear to the humans.
> 
> I would strongly urge you not to consider writing such software
> until you have tried some that already exists.  PAF was only $ 35
> the last time I checked, but it really is not necessary to buy
> such software for a peek.  Just find friendly genealogy fans to
> "show you" their family records and you will get to see the
> existing software in action.
> 
> Hang around the genealogy newsgroups/chat boards to learn areas of
> satisfaction and dissatisfaction with existing software.
> 
> The last I knew, GEDCOM was still the standard interface format.
> It has a real specification, available from the Church of Latter
> Day Saints (Mormons) in the US.  As part of their theology that
> church has done lots of genealogy work worldwide, and my wife has
> found they are quite open to sharing that information with any
> non-believers who are interested.  The only hitch is that film
> must be viewed on their property, since that is a condition of
> the agreements under which various town clerks let them film the
> records (before there were computers).

Remarriages, how common is it?

[Petter Fryklund]

Not that it matters for the project, the requirement to handle the events
are already there, but I'm curious since my grandmother and grandfather on
my mothers side where married  twice with several years passing (I don't
know, but I think 7 or 8) between the two instances. My grandfather was even
married in between.


>
> For marriages, some people marry each other multiple
> times (that is, the same partners changing their minds).

It's very interesting how many different details there are to consider when
writing software. You really need to be flexible. It's also nice to have
control of your requirements and be able to change them as time goes by. I'd
say Ada would be great for such a project.

Re: Genealogy Software

[Ted Dennison]

In article <3BD498DF.27E8B66C@otelco.net>, Larry Hazel says...
>
>Ada 83.  Will need some way to represent dates going back hundreds of years
>(maybe just a fixed length string since some of them will be of the form "ca.
>1784").  Will need a person object (maybe abstract with separate child types 

I remember several threads here from days of yore about the proper way to
represent dates going back hundreds or thousands of years. It might be worth
hitting the archive on groups.google.com and looking for them.

---
T.E.D.    homepage   - http://www.telepath.com/dennison/Ted/TED.html

No trees were killed in the sending of this message. 
However a large number of electrons were terribly inconvenienced.

Re: Genealogy Software

[Larry Hazel]

Wes Groleau wrote:
> 
> Larry Hazel wrote:
> > Does anyone know of any genealogy software written in Ada?  With source
> 
> I have entertained thoughts of writing one with
> some similarities to LifeLines.
> 
> Having a data structure in memory loosely based on GEDCOM,
> and storing them as actual GEDCOM fragments.  This would
> make creating a GEDCOM file as simple as
> 
>   cat header data/* trailer
> 
> Some of the other implementation details are in my head.
> 
> I haven't had the time to write anything down.
> Interested in collaboration?

I should make it clear.  My wife is the genealogy nut, not me.  I have heard of
GEDCOM but don't know anything about it.  Guess I need to try to find a
specification somewhere.  She started with PAF about 15 years ago when it was
just about the only thing available.  She didn't like it much and moved to Roots
III, then Roots IV, then Ultimate Family Tree when it took over Roots.  Now, the
company selling UFT is bankrupt with no more upgrades or support available.  It
also appears to be based on FoxPro for Windows which is obsolete and may quit
working on newer versions of Windows.  Thus my interest in having something that
I can provide support and enhancements for.

She has tried Family Tree Maker and doesn't much like it.  I think there is one
other called "Generations" of or something similar that she hasn't tried.

As I think more about the complexity of the task, I think it would would become
a full time job with no end in site.

Larry

Re: Genealogy Software

[Larry Kilgallen]

In article <3BD57C9E.168A0DF8@otelco.net>, Larry Hazel <lhhazel@otelco.net> writes:

> I should make it clear.  My wife is the genealogy nut, not me.

Likewise.  The best reputation I have heard for a commercial product
is Reunion.  I think they are up to V6 now.  I bought V5 for my wife,
but she hasn't had time to really make use of it, so it is not even
installed yet.

Re: Genealogy Software [off-topic]

[Wes Groleau]

Larry Kilgallen wrote:
>         You need to be able to express _multiple_ birth dates,

Expressing them is not too hard.  Search and compare algorithms
for multiple and/or uncertain dates (along with all the other
genealogical oddities of dates) are a bear.

>         when there are conflicting data from various sources.

Although GEDCOM has its flaws, something similar to it is a good
data structure for data that is as variable in format as genealogy.

> I would strongly urge you not to consider writing such software
> until you have tried some that already exists.  PAF was only $ 35

Definitely.  There are many decent freeware/shareware packages.
Give some of them a thorough workout and see what's wrong with
them before writing the requirements for your own.

GEDitCOM, Lifelines, GIM, and Brother's Keeper are four that
I find worthwhile, in that order of preference.  PAF has a
so-so user interface, but it's GEDCOM implementation is the
worst ever.  (Ironic that the people who invented GEDCOM also
produce the worst implementation of it!)

For what it's worth, no matter what genealogy package I try,
I eventually get fed up with its limitations.  I finally
ended up editing my GEDCOM file directly, and using the
software only for error checking and searching.  Which is
why I planned to write my own software--only the stuff above
it on my priority list adds up to more than 168 hrs/wk !

There is a GEDCOM-L mailing list devoted to enhancement ideas
for GEDCOM.  Last I checked, it seemed to be expannding into
general technical discussions about things "similar to" or derived
from GEDCOM and implementation ideas.

> ... (Mormons) ... are quite open to sharing that information with any
> non-believers who are interested.  The only hitch is that film
> must be viewed on their property, since that is a condition of

Not necessarily.  Libraries can get stuff from the LDS archives
by inter-library loan.  Same rental fee as if done at a Mormon facility
and still, must use it at the receiving library (but how many people
have microfilm readers at home anyway?)  And of course, the GEDCOM
specs are not nearly so private.

Web search for GEDCOM 5.5 specification will find several mirrors
of the GEDCOM spec in various formats.

-- 
Wes Groleau
http://freepages.rootsweb.com/~wgroleau

Re: Remarriages, how common is it? [off-topic]

[Wes Groleau]

Petter Fryklund wrote:
> are already there, but I'm curious since my grandmother and grandfather on
> my mothers side where married  twice with several years passing (I don't
> 
> > For marriages, some people marry each other multiple

In my mind the biggest flaw in GEDCOM is the assumption
that every person is a child of exactly one family and
husband or wife of zero to many.

I'd love to see something simlar to GEDCOM except that
people are related to other people directly instead of
indirectly by family group.

Family could still be there as one case of a group
record which has members with roles and time spans.

-- 
Wes Groleau
http://freepages.rootsweb.com/~wgroleau

Re: Remarriages, how common is it? [off-topic]

[Ted Dennison]

In article <3BD590E9.DF0A8D31@sparc01.ftw.rsc.raytheon.com>, Wes Groleau says...
>In my mind the biggest flaw in GEDCOM is the assumption
>that every person is a child of exactly one family and

That would be a bit of a problem. How would it handle adoptions? Presumably
you'd like to track back both the genetic and adoptive lines. I'm not a
geneology person myself, so perhaps I'm wrong about what they care about.

---
T.E.D.    homepage   - http://www.telepath.com/dennison/Ted/TED.html

No trees were killed in the sending of this message. 
However a large number of electrons were terribly inconvenienced.

Re: Remarriages, how common is it? [off-topic]

[Wes Groleau]

Ted Dennison wrote:
> In article <3BD590E9.DF0A8D31@sparc01.ftw.rsc.raytheon.com>, Wes Groleau says...
> >In my mind the biggest flaw in GEDCOM is the assumption
> >that every person is a child of exactly one family and
> 
> That would be a bit of a problem. How would it handle adoptions? Presumably
> you'd like to track back both the genetic and adoptive lines. I'm not a
> geneology person myself, so perhaps I'm wrong about what they care about.

That's one of the many reasons I consider it a flaw.
And I'm not alone.  Technically, the GEDCOM syntax
would allow being in multiple families, but the semantics
don't and most software packages obey the semantics better
than they obey the syntax.

-- 
Wes Groleau
http://freepages.rootsweb.com/~wgroleau

Re: Genealogy Software then slightly [OFF TOPIC]

[Martin Dowie]

"Larry Hazel" <lhhazel@otelco.net> wrote in message
news:3BD498DF.27E8B66C@otelco.net...
> Does anyone know of any genealogy software written in Ada?  With source
> available?  I thought I might attempt to write some.  Any suggestions for
data
> representation will be appreciated.  I haven't done much Ada 95, but a lot
of
> Ada 83.  Will need some way to represent dates going back hundreds of
years
> (maybe just a fixed length string since some of them will be of the form
"ca.
> 1784").  Will need a person object (maybe abstract with separate child
types for
> male and female).  Will link to other person objects which represent
spouses,
> children, and probably backward to parents.  An event object (marriage,
birth,
> death, buying or selling property, ...).  Bibliographical text will also
be
> needed.  Storage of scanned pictures and documents.
>
> Maybe this is too complicated for a 1 man home project.

It seems from the responses that the requirements analysis isn't finished
yet - perhaps you should try comp.lang.UML (doesn't exist! why????) first
and
then code it up in Ada! ;-)

Has anyone tried the Ada Code Generator that comes with the freeware UML
tool from Pragsoft? www.pragsoft.com

I'm not advocating it or the UML part is good, but it can be run in a
freeware
mode and does come with (allegedly) a host of code generators...

SV: Remarriages, how common is it? [off-topic]

[Bror Johansson]

Wes Groleau <wwgrol@sparc01.ftw.rsc.raytheon.com> skrev i
diskussionsgruppsmeddelandet:3BD590E9.DF0A8D31@sparc01.ftw.rsc.raytheon.com.
..
>
>
> Petter Fryklund wrote:
> > are already there, but I'm curious since my grandmother and grandfather
on
> > my mothers side where married  twice with several years passing (I don't
> >
> > > For marriages, some people marry each other multiple
>
> In my mind the biggest flaw in GEDCOM is the assumption
> that every person is a child of exactly one family and
> husband or wife of zero to many.
>
> I'd love to see something simlar to GEDCOM except that
> people are related to other people directly instead of
> indirectly by family group.
>
> Family could still be there as one case of a group
> record which has members with roles and time spans.
>

As far as I have been informed about the forthcoming GEDCOM/XML (from The
Family and Church History Department of the Church of Jesus Christ of
Latter-day Saints) it will be capable of handling most - if not all -
(well-)known defiencies of present-day GEDCOM.

/BJ

Re: Genealogy Software

[Arthur Evans Jr]

In article <PleB7.39487$ev2.46198@www.newsranger.com>, Ted
Dennison<dennison@telepath.com> wrote:

> In article <3BD498DF.27E8B66C@otelco.net>, Larry Hazel says...
> >
> >Ada 83.  Will need some way to represent dates going back hundreds of years
> >(maybe just a fixed length string since some of them will be of the form "ca.
> >1784").  Will need a person object (maybe abstract with separate child types 
> 
> I remember several threads here from days of yore about the proper way to
> represent dates going back hundreds or thousands of years. It might be worth
> hitting the archive on groups.google.com and looking for them.

There's an ISO standard for today as 2001.10.24 (or maybe 2001-10-24).
Dates stored like this go back further than any genealogist is likely
to need (well, unless you can trace back to Adam).  Moreover, they
collate, in that an alphabetic comparison between two dates reveals
which is later.

This solution of course does not address date ranges, or approximate
dates.

Art Evans

Re: Genealogy Software [off-topic]

[Arthur Evans Jr]

In article <3BD58F53.7A3F89B@sparc01.ftw.rsc.raytheon.com>,
wwgrol@sparc01.ftw.rsc.raytheon.com wrote:

> Although GEDCOM has its flaws, something similar to it is a good
> data structure for data that is as variable in format as genealogy.

GEDCOM is designed for communication, not for internal representation.
If I were doing this, I would study GEDCOM and select an internal
representation that can readily be converted to or from it, but I doubt
that I would use GEDCOM itself.

I suggest you subscribe to soc.genealogy.computing, which is probably a
better place to continue this discussion.  I used to subscribe to it but
do so no longer.  I just checked, and it's quite active.

Art Evans

Re: Genealogy Software

[Larry Kilgallen]

In article <ev_remove_this_ans-2410010907240001@192.168.1.254>, ev_remove_this_ans@evans.pgh.pa.us (Arthur Evans Jr) writes:

> There's an ISO standard for today as 2001.10.24 (or maybe 2001-10-24).
> Dates stored like this go back further than any genealogist is likely
> to need (well, unless you can trace back to Adam).  Moreover, they
> collate, in that an alphabetic comparison between two dates reveals
> which is later.
> 
> This solution of course does not address date ranges, or approximate
> dates.

Or issues regarding which calendar system was used to record that date.
Today dates around the world are synchronized, but that was not so
for a while after the pope ordered a change.

Re: Genealogy Software [off-topic]

[Wes Groleau]

Arthur Evans Jr wrote:
> If I were doing this, I would study GEDCOM and select an internal
> representation that can readily be converted to or from it, but I doubt
> that I would use GEDCOM itself.

That's what I mean.  Many of the irritating limitations
of some software are caused by trying to squeeze the data
into a less-than-optimal data structure.

A dynamic tree/map/network/list type of structure corresponds
to the GEDCOM model of independent records with nesting and
pointers between records.

-- 
Wes Groleau
http://freepages.rootsweb.com/~wgroleau