This is more or less the "GedToWeb" home page. This page will at some point link to both source code and freeware on the subject of automatic web page construction from genealogical data. The Church of Jesus Christ of the Latter-Day Saints have created a standard data interchange format entitled The Gedcom Standard. Currently in release 5.5, this is "a flexible, uniform format for exchanging computerized genealogical data. GEDCOM is an acronym for GEnealogical Data Communication." This from The Gedcom Standard, Release 5.5. For a reasonable article see the Wikipedia GEDCOM article. By now it should be clear that GedToWeb is most likely software that converts a gedcom file (extension .ged) to some sort of web oriented output. Had you guessed that you would be right on the money!
In general it produces a variety of files along the way towards a web version of a family tree. This is done by way of two different indexing approaches, the first through a dictionary like method based on individual name and the second by way of family name, i.e. surname. As a sample of the former consider:
Eva Cora ADAMS —
to — Randy Dwayne DEMOSS
Barbara DENNIS —
to — Ella HUMPHREYS
Lillian HUNTER —
to — Ellen Jane MYERS
Ellen Lucille MYERS
— to — Larry Saul RIPLEY
Robert E. RIPLEY —
to — Ella WALDEN
Harland G. WALKER
— to — Virginia
And if we follow the first link, we would wind up with:
Eva Cora ADAMS —
to — Betty ALLEN
Brenda ALLEN —
to — Clay Dean ANDERSON
Dona ANDERSON —
to — Georgia Ann ARNALL
Harold Dean ARNALL
— to — Ronny Bill BARBER
June BARNES —
to — John C. BELL
May BELL —
to — Richard Michael BERRY
Goldie A. BESHEARS
— to — Mary Etta BOATWRIGHT
Billy BOAZ —
to — Lillie BOWERS
David BOWER —
to — Lizzie BOYLE
Hope BOYNTON —
to — Reinee BROCKOVER
Ronnie J. BROCKOVER
— to — Billy Richard BROWN
Carl BROWN —
to — Lee BROWN
Mary Elizabeth BROWN
— to — Irvan BUNYAN
Keith Allen BURBRIDGE
— to — Roscoe D. BUTTREY
Merlee Hansen BYNUM
— to — Harriett CARTER
Vera CARTER —
to — Shirley CASON
Tandy M. CASON —
to — Leona Pearl CAYWOOD
Lorenzo Ran CAYWOOD
— to — Robert Scott CHARLTON
Shael Diane CHARLTON
— to — William Stephen CLARK
William Stephen CLARK
Jr. — to — Ausdille William COLE
Barbara Ann COLE —
to — Lloyd W. COOK
Paul COOK —
to — Walter COX
Brenda COY —
to — Nelson CROWDER
Polly CROWDER —
to — Sandra Fae DALLING
William R. DALLING
— to — Randy Dwayne DEMOSS
And with the same logic (following the first link) we would get:
Eva Cora ADAMS(____ - ____)
Freida M. ADAMS(____ - ____)
Hellen ADAMS(____ - ____)
Alma Faye ADDINGTON(____ - ____)
James ADDINGTON(____ - ____)
Prill Myrtle Wardle ADKINS(____ -
____)
Mahela ALBERTY(____ - ____)
Mahela ALBERTY(____ - ____)
Mary Kathleen ALBRIGHT(____ - ____)
Douglass ALEXANDER(____ - ____)
Douglass ALEXANDER Jr.(____ - ____)
Adam Austin ALLARD(____ - ____)
Bill ALLARD(____ - ____)
David ALLARD(____ - ____)
Donna ALLARD(____ - ____)
J. Daniel ALLARD(____ - ____)
Jim Austin ALLARD(____ - ____)
Linda ALLARD(____ - ____)
Melrose ALLARD(27 DEC 1923 - 2 AUG
1981)
Michael ALLARD(____ - ____)
Rev. Austin ALLARD(29 JUL 1909 - JAN
1988)
Sunshine ALLARD(____ - ____)
Anna Joy ALLEN(16 AUG 1933 - ____)
Barbara ALLEN(____ - 1965)
Betty ALLEN(13 DEC 1935 - ____)
And now, one more step to:
Family 1: John Merrit WOOTERS
| Unknown | ||||
| ( ____ - ____ ) m ____ | ||||
| Unknown | ||||
| ( ____ - ____ ) m ____ | ||||
| Unknown | ||||
| ( ____ - ____ ) m ____ | ||||
| Eva Cora ADAMS | ||||
| (____ - ____) | ||||
| Unknown | ||||
| ( ____ - ____ ) m ____ | ||||
| Unknown | ||||
| ( ____ - ____ ) m ____ | ||||
| Unknown | ||||
| ( ____ - ____ ) m ____ |
So with just a few steps we can narrow in on a particular person and their relationship to the overall family tree. Admittedly, Eva Cora Adams might not be the best example to use as an illustration but her name has the signal virtue of coming first in the current list! The reason that more than one step is needed for this method lies in the amount of information indexed. With the method implemented, one name in some 3,000 odd is found with no more than 4 lookups necessary. While a monolithic approach could have resulted in a single lookup, page size should always be a consideration on the web. Using a stepped algorithm allows a balance between lookup speed and individual page access time.
The second method is based on family names themselves. Again, a familiar approach is used, this one similar to a textbook index, i.e.:
ADAMS, ADDINGTON, ADKINS, ALBERTY, ALBRIGHT, ALEXANDER, ALLARD, ALLEN, ALTMAN, ANDERSON, ANTONINA, ARBUCKLE, ARCULIN, ARDOIN, ARMSTRONG, ARNALL, ARNBRECHT, ARNOLD, AUBER, AUGUSTINE, AVERY, AYRES,
Here clicking on ADAMS would land us on the second from the last page in the first example.
Needless to say this is not particularly easy to maintain by hand, hence the need for a program to generate the information on demand from a more user-friendly source.
Here are links to:
As with all such items of free-lunch-ware, what you see is what you get, no warrantee implied or otherwise for any purpose. If they work, great. If not tell me about it and I'll fix it (with any luck and if I feel like it<g>!)
Note: [Thu Aug 23 08:12:52 2012] The above 'non' links are a place holder until I organize the source code and otherwise get my act together. I've located what I think is the current (circa 2010 or so) code and am now puzzling over how to present it. I'm toying with the idea of setting up a github repository and providing a link to that. That way I can offer both the source code and an executable (or two) as seems reasonable. To do that however, I've first have to find the .exe(s) in question or re-create a build process that will be more or less cross platform. Since I'm only now getting back to this, it obvously will take some time. It will happen, I'm just not willing to say when!
I'd also point out that I've actually implemented (and tested) a GEDCOM server written in Perl. There will at some point be a link and article on my Perl page Software > Perl It will likely also wind up on github as that seems to be the best way to share code in this the 21st century!
Update: [Fri Sep 21 09:07:27 2012] That was then, this is now! The current approach now depends on a heavily modified version of gedcomToHTML.pl from GedcomToHTML. As an example see S1.html. This explains why all of the links above are disabled. At some point the plan is to modify GedToWeb to generate similar output to match the current display. That will take a substantial amount of work but given that the current numbers for my family tree is now standing at ‘3610 individuals, 1366 families, 3 sources, 0 repositories’, it will be well worth the effort! Considering that this amounts to ‘3649 File(s) 9,406,699 bytes’ with the current system, clearly converting this to a compiled approach is pretty much necessary. Another reason to match the gedcomToHTL.pl method is that it uses both a config file as well as templates which allows a great deal of flexibility in designing the output. You might say that I'm using the perl script to prototype the features for the eventual C++ executable!
Looking ahead, I'm thinking that the best approach is to first convert the core of the existing code into a library. This would allow a variety of possibilities. Two come immediately to mind; first, a web page generator as described above. Second, as the heart of a gedcom editor—although there are some pretty impressive pieces of software already out there, this would be open source and available for abuse use by all and sundry!
So this is pretty much a ‘stay tuned’ kind of situation— so do so!
