PDA

View Full Version : Seeking suggestions for generic 'event' field name


Linda G
05 July 2008, 12:46 PM
This morning I was trying to come up with a field name for when a person attends a funeral, to distinguish from when it's a person's own funeral. Mainly because, in my smart list view, I'd like to separate out all post-death events.

Then I thought, do I really need a special event for attending a funeral? what about other occasions, etc. When I have access to old newspapers, it's amazing the number of times some families are mentioned, for a variety of reasons.

I'd like to hear your thoughts.

This doesn't seem all that important, but I'm having trouble coming up with ideas other than 'event'--meaning 'generic' event. :)

John Yates
05 July 2008, 01:18 PM
Linda,

I think you may be looking for the equivalent of "Witness" from another program. A Witness is someone who has any (not necessarily even presence) association with an event. One example I have seen is the pallbearers in a funeral. So you may not need a generic event if you could "tie" people to specific (exisiting) events in a certain manner.

P.S. Linda, I have a question about genealogy I'd rather ask you off list first. If you would, please click on my "email me" and I'll explain.

Bob White
06 July 2008, 07:47 PM
This morning I was trying to come up with a field name for when a person attends a funeral, to distinguish from when it's a person's own funeral. Mainly because, in my smart list view, I'd like to separate out all post-death events.............
My thoughts/practice: My experience is that somewhere in the 90 - 95% range that the burial date and the "person attends a funeral" date are one and the same. So, the now and then that they are different, I note it in the Note section. It's not information that I'm ever going to make a list of or need to sort, so I feel this method is more than sufficient. The very few of these I have run into were almost all the situation where a service is held in the city where a person has lived and worked -- and then, after that service, the person's body is shipped to his or her childhood home for burial in a family plot.

(Note: I have never even thought of the idea of tracking when someone attends a funeral. That's for the tabloids. With me, it's just that the funeral date and the burial date are two different dates and both should be recorded.) :)

Deena
07 May 2009, 11:01 AM
This morning I was trying to come up with a field name for when a person attends a funeral, to distinguish from when it's a person's own funeral. Mainly because, in my smart list view, I'd like to separate out all post-death events.

Then I thought, do I really need a special event for attending a funeral? what about other occasions, etc. When I have access to old newspapers, it's amazing the number of times some families are mentioned, for a variety of reasons.

I'd like to hear your thoughts.

This doesn't seem all that important, but I'm having trouble coming up with ideas other than 'event'--meaning 'generic' event. :)

I used to use CTree and it had a feature called "Witness". It was an event type basically where you could choose the event and record that a person was there. I really liked it. I used it for things like recording all the people who signed the guest register at weddings or funerals. I've been struggling where to document this information since I converted to Reunion (which I love and wouldn't change but just doesn't have this feature).

Troy
07 May 2009, 01:18 PM
My 2-cents for what it's worth.

If there are more than one type of witness events used, then perhaps you could have several of them: (e.g., witness-funeral, witness-wedding, etc.).

Just a thought.

Troy---

Linda G
07 May 2009, 04:16 PM
Wow. 10 months have passed already.

Thanks, Troy, for posting a response. I had forgotten I had asked this question.

Re-reading it now, in the context of other issues I've been 'pondering' recently, there are 2 reasons why recording this info is important to me.

First, it places an individual in a specific time and place. useful for reconstructing a person's timeline. Reunion & the usual genealogical software meet this goal, in spite of the fact that I still haven't come up with a label I like. Witness works ok...I just have to get beyond associating it with legal documents. LOL

The second reason has to do with using cluster genealogy as a tool to find out more about an individual. Knowing the groups/associates can help incredibly to resolve questions of identity. But to use this approach, it truly would help to be able to 'see' the data from the group's view. Or, in this case, from the event's view (see all individuals' associated with the event). and to be able to layer/combine/visualize different groupings. Reunion does not currently offer this type of data organization, although Ted offered an ingenious approach in another thread (http://www.reuniontalk.com/showthread.php?t=4827&page=2&pp=10). I'm not sure any of the typical genealogy software currently available addresses this goal of someone using the cluster genealogy approach. This just may be 'a horse of a different color'. More along the lines of studying communities, and the individuals that belong to them. The fact is that our ancestors were not isolated individuals, and complicated as it may seem, understanding their communities goes a long way to first identifying them, and then understanding them. If anyone knows of useful tools, I'd appreciate your suggestions. Especially any that offer mapping visualizations. Surely someone here does one-location studies!

Reiner L. Sauer
08 May 2009, 07:11 PM
Hi Linda,

This morning I was trying to come up with a field name for when a person attends a funeral, to distinguish from when it's a person's own funeral. Mainly because, in my smart list view, I'd like to separate out all post-death events.

Then I thought, do I really need a special event for attending a funeral? what about other occasions, etc. When I have access to old newspapers, it's amazing the number of times some families are mentioned, for a variety of reasons.

I'd like to hear your thoughts.

This doesn't seem all that important, but I'm having trouble coming up with ideas other than 'event'--meaning 'generic' event. :)

I'm using events field called

1. "appears/appeared" and/or
2. "mentioned"
as neutral lables and use the "memo" field to enter details.

Regards
Reiner

Linda G
09 May 2009, 11:37 AM
I like your labels! Neutral, but accurate.

Thank you, Reiner!

Jillaine_Smith
11 May 2009, 09:58 AM
Linda,

I'd be VERY interested in participating in a discussion about cluster genealogy. Is such a conversation happening somewhere?

I'm familiar with and have written about social networks from other areas of my life (nonprofit networks, advocacy networks) but have not applied these ideas to genealogy (yet).

I totally concur with you about how helpful it would be to see our ancestors as part of a cluster.

I bet that out there somewhere is an expert in social network mapping who is also interested in genealogy.

-- Jillaine Smith
Bethesda, MD

Linda G
11 May 2009, 11:54 AM
Jillaine,

I've been looking (off-and-on for a year or two) and haven't found a forum where cluster methodology/tools are discussed. The term is used in NGSQ articles and the like, primarily as a 'gold standard' technique, but no hint of data management tools used.

I'm relatively new to family history research and know that I have a lot to learn about where different topics are being discussed. I've joined a couple of the 'professional' type groups, and occasionally ask online. No real answer, which sort of implies that they're using 'pencil & paper' or spreadsheets, which I find very difficult to believe. One person said they wrote their own database management tool. I haven't been able to go to national or regional conferences to try asking in person.

I would think that there are academics doing community studies that might have developed some tools useful for cluster genealogy. Now you've given me a new term to google -- 'social network mapping'. That's great! I haven't been successful with the ones I was trying. LOL

So, again, for all those reading this, all references are welcome!

John Yates
11 May 2009, 09:41 PM
I've been looking (off-and-on for a year or two) and haven't found a forum where cluster methodology/tools are discussed. The term is used in NGSQ articles and the like, primarily as a 'gold standard' technique, but no hint of data management tools used.



Linda, I just googled "cluster genealogy" and found (probably the same as you've found):

http://en.wikipedia.org/wiki/Cluster_genealogy
http://www.genealogy.com/heard100302.html
http://genealogy.about.com/od/basics/a/cluster.htm
http://www.nationmaster.com/encyclopedia/Cluster-genealogy

Here are my thoughts on "cluster genealogy". I think the methods would be the same as used in traditional genealogy. Current tools can be used if you limit your cluster to relatives (of any sort) so connections between linked list boxes can be drawn. Even the shirttail ones that you wouldn't have followed in your "regular" research. When you begin to include friends, neighbors, etc., current programs won't be of much use in recording your evidence. Or in displaying the evidence well. The additional complexity of the linked lists will need more advanced programs.

Next, I think that the limit of a human lifetime severely limits the amount of evidence that an individual can collect. So cluster genealogy by an individual would need to concentrate only on, say, a few people and their locations. So I find it no surprise no one has developed a widespread "tool" for holding such evidence. Current programs have enough problems with handling the more limited relationship evidence in them as it is.

To do a decent job of cluster genealogy, one would require sharing the information with others. We are in the Neanderthal age of that with GEDCOM. (privacy and objections to sharing and accuracy of others work... etc. etc. all aside).

What is needed is a next generation (or more) generalized (to include non-relatives) version of the World Tree. When computers do it for us from the digitized documents from all the sources that genealogists use. (Go Google book scanning!). A mark up language in SGML (XML perhaps) will need to be developed for genealogy tags. And then programs that "do" genealogy from this now on line data will be developed. A Google World Tree? And "smart" programs (bots of today evolved?) will be able to take the trees we've developed, error check them with the "original data" and add our "uniquely known" information (weighted somehow, of course), and immediately tell us we are related to George Washington, and how, etc... Since the tree will be so large we will be reduced to viewing only small pieces of it, and using programs to ask questions of the data: how am I related to X? Where did Y live? Who were Z's immediate (define that first) relatives? etc. Documents will be defined and generated that produce "The Genealogy of the (an?) ABC family of (somewhere in?) England (or its precursor) 1000-2100". With confidence levels (you can think of adding in DNA data as more evidence to go into a confidence level).

So, in my opinion, cluster genealogy will only come into its own when Google gets done with digitizing the knowledge of the world, and interested programmers codify a markup language for it, and write programs to do the research. This may nor may not happen in our lifetime. Meanwhile, I'll continue to add what I can to my tree as a hobby, doing the best I can with current techniques and programs, and some day some computer program will validate it, reject inconsistencies (say if a source I used is disproved or...), and take some of it as first hand knowledge (that can be accepted or rejected by the program or others...), and I hope to get at least a grade of 95% correct! 98%? 99.9%? Who knows. I hope to live long enough to have my little genealogy evidence "graded" and extended.

With GEDCOM being the state of the art of genealogical data sharing today, you can see how futile extended cluster genealogy is today other than the microcosm of our individual research.

So, in a generation or two we will see how our evidence stands up, and importantly, how it is extended into one world tree. None of this diminishes our current genealogical efforts in any way. But as in all fields, the tools of the future will be amazing. Progress will be made as the computerization steps mentioned above are achieved. Meanwhile, we just do our best today with the primitive tools we have. And keep making "noises" about the future. ;-)

John

martha
12 May 2009, 03:58 AM
Jillaine,

I've been looking (off-and-on for a year or two) and haven't found a forum where cluster methodology/tools are discussed. The term is used in NGSQ articles and the like, primarily as a 'gold standard' technique, but no hint of data management tools used.

I'm relatively new to family history research and know that I have a lot to learn about where different topics are being discussed. I've joined a couple of the 'professional' type groups, and occasionally ask online. No real answer, which sort of implies that they're using 'pencil & paper' or spreadsheets, which I find very difficult to believe. One person said they wrote their own database management tool. I haven't been able to go to national or regional conferences to try asking in person.

I would think that there are academics doing community studies that might have developed some tools useful for cluster genealogy. Now you've given me a new term to google -- 'social network mapping'. That's great! I haven't been successful with the ones I was trying. LOL

So, again, for all those reading this, all references are welcome!

Linda, I saw your previous posting and question about doing a one town study. But I didn't and still don't know exactly what you are looking for, so I didn't respond. I am trying to recreate the Jewish families of a city whose entire Jewish population was murdered in the Shoah. So in that sense, yes, I am doing a one town study. But how could/would this help you?

Martha

Reiner L. Sauer
12 May 2009, 07:01 AM
Jillaine,

I've been looking (off-and-on for a year or two) and haven't found a forum where cluster methodology/tools are discussed. The term is used in NGSQ articles and the like, primarily as a 'gold standard' technique, but no hint of data management tools used.

I'm relatively new to family history research and know that I have a lot to learn about where different topics are being discussed. I've joined a couple of the 'professional' type groups, and occasionally ask online. No real answer, which sort of implies that they're using 'pencil & paper' or spreadsheets, which I find very difficult to believe. One person said they wrote their own database management tool. I haven't been able to go to national or regional conferences to try asking in person.

I would think that there are academics doing community studies that might have developed some tools useful for cluster genealogy. Now you've given me a new term to google -- 'social network mapping'. That's great! I haven't been successful with the ones I was trying. LOL

So, again, for all those reading this, all references are welcome!Hi Linda, Hi Jillaine,

If the definition of 'cluster' includes e.g. >> the population of an area or town over the last 300+ years << , then I'm doing cluster genealogy.
I have about 20 years experience with this and would like to join a discussion about this and share my experience.

Regards
Reiner

Linda G
12 May 2009, 01:27 PM
We should probably create a new thread, or at least get a more accurate subject. <g>

Martha & Reiner, I would love to hear how you each manage the data you collect. Are you using just Reunion? Have you found other tools useful?

John,

You seem to have focused on 'reconstruct the whole world' as my primary reason for looking for data management tools to support cluster genealogy research. It really isn't.

I'll try to explain my reasoning as simply as I can, hoping for clearer communication.

1. The primary goal of genealogical research is to identify kinship relationships.

2. A corresponding goal, a close second to #1, is to 'recreate' an individual's life--a timeline placed in historical context. Personally, my goals regarding level of detail vary according to individual. For some I 'like' more just because I'm interested. For some I 'need' more because I'm trying to answer some question regarding them. For some, the bare bones suffices.

3. The researcher analyzes multiple sources, and forms conclusions in order to identify kinship relationships (#1). Likewise, adding to #2.

4. Reunion (and the 'typical' genie software) is wonderful for managing information ONCE THE KINSHIP RELATIONSHIP IS DETERMINED. I'm trying to also cautiously use Reunion for SUSPECTED kinships.

5. For me, cluster genealogy offers approaches to help me evaluate questions of identity (determining a kinship relationship). Productive cluster approaches include more than kinship relationships.

*collateral relatives-->kinship relationships (for example, examining siblings/in-laws can help identify parents/spouse maiden names)
*same place/time --> geographical relationships (for example, examining abutters and considering those individuals as possible family members)
*change in same place/time (migration)-->geographical relationships
*known associates--> social relationships (for example, examining other members of a military unit to determine residence)

*and combinations of the above <vbg>

These are not my original ideas. Journal articles/books tout these approaches as effective techniques for evaluating identity problems.

The above approaches require integrating information from multiple sources (i.e. just knowing the names of the abutters is not enough!) Just like we do in Reunion, but kinship is generally not known for many of the individuals. Other relationships (geographical, social) are more critical, depending upon the technique used. This is the heart of my research problem. Just looking at information from one source isn't enough. I would like to integrate info from multiple sources and be able to 'look'' at it, ideally from multiple points of view (kinship, geographical, social).

So, John, while I admit to being sorely tempted many times to wanting to reconstruct communities as Martha and Reiner are doing, my desire to find a data management tool supporting cluster approaches is motivated by my efforts to achieve #1 (identify kinship relationships). But any tools they use (or that others use for one-location, perhaps one-surname, studies) interest me, because I might be able to apply them for my purposes.

It seems to me that Reunion could be relatively easily extended to include social relationships. At least for storing the information. Reporting is another issue, perhaps.

Geographical relationships seem like more of a challenge, because, I think, one would really like an easy way for the user to create mapping visualizations--and that would be a brand new world for Reunion.

Jillaine_Smith
12 May 2009, 05:39 PM
Linda and others,

I have two examples of how cluster genealogy and genealogy software can support each other to identify relationships:

Example #1 - "My Elisabetha"
I had two goals:
1. tie my gr-gr-grandmother Elisabetha Betz to an emigrating family thus identifying her parents or at least her siblings.
2. confirm the location in Germany of this Betz family so I could work it back further. (I had a region, but not a town.)

I noticed on Buffalo NY baptism and marriage records that Betz individuals showed up as godparents to Elisabetha's children and witnesses to her marriage. My software at the time already had a "Witness" field for marriages. I also created two custom fields for individuals:

a) Godparents of
b) Godparents are

This enabled me to generate a report of a list of godparents and compare it to those who had godparents (for a particular time period).

I then started looking at the individuals who showed up in these roles, including who THEY married. Some of the same names kept showing up again and again. I figured there must be some relationship.

Then I searched the IGI and created a separate family file for the BETZ families I found that a) had sets of children with the same names as showed up in Buffalo, and b) were within a radius of a certain number of miles of the region I knew they were from.

By examining both, I found a family in a particular town that while not in the right generation, had a set of siblings with the same given names. My cousin ordered the microfilm for that town, and sure enough, not only found "My Elisabetha" but confirmed her siblings. By chance, I was in Germany when all this happened and was able to visit the little town from where my ancestors hailed from.

Example #2 (i'll keep it short):
My husband's colonial ancestors included a lot of confusing overlaps, and confirming which Hannah Rice married which Richard Taylor was a nightmare complicated by a lot of unsubstantiated "published" data. Mapping the families and relationships of Sudbury, MA founding families, including birth, marriage, death information, wills and probate information, who served under which captain in King Philip's war, etc. etc. all led to -- at a minimum -- proving who Richard Taylor, for example, was NOT. This all required researching people who ended up not being directly related to my husband, but who were intimately connected with his ancestral families. Social network mapping software (or Google tags or whatever) incorporated into genealogy software could have helped tremendously, and no doubt could have provided some visual "maps" of relationships that not only help me, but also help others.

AND... in both examples, but especially #2, these kinds of networks/clusters also help flesh out stories and histories-- the real "flesh" of the data of b/m/d. And that's what I'm REALLY interested in...

-- Jillaine

Reiner L. Sauer
12 May 2009, 06:38 PM
Linda,

We should probably create a new thread, or at least get a more accurate subject. <g>



Let me start a new thread under "Cluster Genealogy".
My response will be under this new reference

Regards
Reiner

Reiner L. Sauer
12 May 2009, 07:44 PM
Hi Linda,



Martha & Reiner, I would love to hear how you each manage the data you collect. Are you using just Reunion? Have you found other tools useful?

John,

You seem to have focused on 'reconstruct the whole world' as my primary reason for looking for data management tools to support cluster genealogy research. It really isn't.

I'll try to explain my reasoning as simply as I can, hoping for clearer communication.

1. The primary goal of genealogical research is to identify kinship relationships.

2. A corresponding goal, a close second to #1, is to 'recreate' an individual's life--a timeline placed in historical context. Personally, my goals regarding level of detail vary according to individual. For some I 'like' more just because I'm interested. For some I 'need' more because I'm trying to answer some question regarding them. For some, the bare bones suffices.

3. The researcher analyzes multiple sources, and forms conclusions in order to identify kinship relationships (#1). Likewise, adding to #2.

4. Reunion (and the 'typical' genie software) is wonderful for managing information ONCE THE KINSHIP RELATIONSHIP IS DETERMINED. I'm trying to also cautiously use Reunion for SUSPECTED kinships.

5. For me, cluster genealogy offers approaches to help me evaluate questions of identity (determining a kinship relationship). Productive cluster approaches include more than kinship relationships.

*collateral relatives-->kinship relationships (for example, examining siblings/in-laws can help identify parents/spouse maiden names)
*same place/time --> geographical relationships (for example, examining abutters and considering those individuals as possible family members)
*change in same place/time (migration)-->geographical relationships
*known associates--> social relationships (for example, examining other members of a military unit to determine residence)

*and combinations of the above <vbg>

These are not my original ideas. Journal articles/books tout these approaches as effective techniques for evaluating identity problems.

The above approaches require integrating information from multiple sources (i.e. just knowing the names of the abutters is not enough!) Just like we do in Reunion, but kinship is generally not known for many of the individuals. Other relationships (geographical, social) are more critical, depending upon the technique used. This is the heart of my research problem. Just looking at information from one source isn't enough. I would like to integrate info from multiple sources and be able to 'look'' at it, ideally from multiple points of view (kinship, geographical, social).

So, John, while I admit to being sorely tempted many times to wanting to reconstruct communities as Martha and Reiner are doing, my desire to find a data management tool supporting cluster approaches is motivated by my efforts to achieve #1 (identify kinship relationships). But any tools they use (or that others use for one-location, perhaps one-surname, studies) interest me, because I might be able to apply them for my purposes.

It seems to me that Reunion could be relatively easily extended to include social relationships. At least for storing the information. Reporting is another issue, perhaps.

Geographical relationships seem like more of a challenge, because, I think, one would really like an easy way for the user to create mapping visualizations--and that would be a brand new world for Reunion.

Reunion is my program of choice for 18/19 years now. First, I used Reunion for own family research, only. Soon thereafter in about 1992/93, I started to use Reunion for the purpose of cluster genealogy. What I did was to record every detail about every soul, which was ever recorded the official and inoffical records of the place (village), I used to lived in. These were either people who were born there, lived there, went to school, married, worked there, fought in battle, was abutter, died, or just appeared at a baptism, wedding, court case etc. etc. Just document the relationships among those people, but never ask the question as to whether they are related to you, or whether the informtion is at all relevant. Just record the info. You'll find out about the relevance later. If they are not related to you, they are related to someone else.

In 1996, I published my research on a CD-ROM (available at the LDS and with Leister). This research covered about 10000 people, covered a period of a couple of hundred years, and a thousand pages of condensed and printed information.

Biggest problem at the time was the Reunion-limit of 99 different sources, and limited number of fact, event and memo fields.

At the time, I decided against a "world-project", but just focused on the cluster genealogy of a rather small but geographically defined place. Instead, I focused on a couple of tens of square miles, constantly hoping for future powerful match-and-merge routines, so that the results of my research could eventually be combined with the research results of neighboring townships.

This limitation may have been partially due to the fact, that -- when I started this project -- I worked and lived about 7000 miles away from home, and the internet was still in its infancy.

After the initial publication, I expanded my research to a couple of other neighboring places, which are part of the same township. Now, my research covers more than 70000 people and more than 14000 different sources.

There's no longer a limit with the number sources Reunion can support. Now, Reunion also supports a virtually unlimited number of repetitive facts, events and memo fields. I'm using more than 300 of such fields in order to cover the most mondane data of the people in the township; and about 60-70 different source-types to record the details about documents, church record, civil records, books, obituaries, newspaper articles etc etc.

There are still items on my wish list for the program (see my list), but cluster genealogy CAN be done using Reunion. Actually, I don't know any other program which offers the same convenience, stability and feature to tackle such big projects.

That's it for today. In future, I'm glad to share with you my Reunion record structure and addtional tools, which I'm using in this context.

Regards
Reiner

Jillaine_Smith
12 May 2009, 11:10 PM
Linda and others, another thing to know about are the new-ish genealogy wikis that are sprouting up.

I particularly like WeRelate.org-- technically in beta. Supports the ability to upload your GEDCOM to a wikipedia-style site for genealogy/family history. Unlike sites like WorldConnect-- which has multiple (!) gedcoms of the same people, the goal of WeRelate is a single "tree" or collection of people. You are contributing your GEDCOM to a community and like wikipedia, others can edit it.

Now, if they could just map social network mapping software over the top of that...

-- Jillaine

martha
13 May 2009, 03:16 AM
Martha & Reiner, I would love to hear how you each manage the data you collect. Are you using just Reunion? Have you found other tools useful?


Linda, Reunion is so flexible and easy to use that I have not used anything else, except a spreadsheet sometimes to see everything at a glance. But Reunion is my basic tool. I put into the database every single little tidbit I find out about anyone. Every person ever recorded or mentioned is included, even those about whom people only recall that "s/he had two twin boys and a daughter, but I can't recall their names." Slowly I have connected the dots and have created accurate family ties. Jewish naming patterns have helped when I was uncertain as to family connections.

Some people left the area and either emigrated or just moved on to the big city, so my research branched out to other countries or cities, to include offspring of the families in the database from the village I began with. My greatest problem is trying to find people still alive in our day and age. None are left in the village as they were all murdered. So far, 4 mass graves have been located, but there is no way to identify the people who were murdered. It is estimated that some 800 - 1,000 bodies have been found so far, but we know there are more than that around the area somewhere. Of course, my greatest fear is that I will leave someone out, that someone who once had a life will not be recorded due to my omission. That thought forces me to be very creative in finding information, believe me.

If there is anything specific that I can help you with, please don't hesitate to ask me.

Martha

Linda G
16 May 2009, 08:18 PM
Thank you, Reiner & Martha!

Sorry I haven't replied earlier. I was so astounded by both your answers that I needed a few days to think. <g>

You each make it sound so easy, that of course Reunion handles your community projects. So what is my problem, I thought to myself.

I think it's this--I've been using Reunion primarily to record individuals with KNOWN kinship relationships. My data management problem occurs before I've identified kinships (or convinced myself that I have).

Are you using Reunion before you know who's who, and what records are talking about the same person? That's the piece I'd very much like to hear about.

For example, last fall I started trying to identify the parents and siblings of my 4g-grandfather. I had some clues and first looked up the original sources to verify them. The question immediately became were there 2 families with the same father's name in an area. Was I looking at 2 sets of parents and children, or one? The surname is COOK. The location Connecticut pre-American Revolution. Well, there are a lot of COOKs in that town in that time period (church, land records, some probate, military), not to mention surrounding towns and in CT in general. None of the records come out and answer the question. I've collected a lot of records and would like to identify families--sorting what goes together from different sources, etc, etc. Hoping that the relationships between other COOKs and associated families will make the answer to my question clearer.

Recently, working on another family line (CARRERE, Baltimore, MD, 1800s, California 1900s), the same thing is happening. Lots of records (voter registration, passenger lists, directories, land, etc) and the need to sort out what belongs to who, with the goal of identifying the families and their connection (or lack thereof) with each other.

Do you use Reunion for this process?

Thank you so much for sharing your experiences!

Reiner L. Sauer
16 May 2009, 09:52 PM
Hello Linda,

Thank you, Reiner & Martha!

Sorry I haven't replied earlier. I was so astounded by both your answers that I needed a few days to think. <g>

You each make it sound so easy, that of course Reunion handles your community projects. So what is my problem, I thought to myself.

I think it's this--I've been using Reunion primarily to record individuals with KNOWN kinship relationships. My data management problem occurs before I've identified kinships (or convinced myself that I have).

Are you using Reunion before you know who's who, and what records are talking about the same person? That's the piece I'd very much like to hear about.

For example, last fall I started trying to identify the parents and siblings of my 4g-grandfather. I had some clues and first looked up the original sources to verify them. The question immediately became were there 2 families with the same father's name in an area. Was I looking at 2 sets of parents and children, or one? The surname is COOK. The location Connecticut pre-American Revolution. Well, there are a lot of COOKs in that town in that time period (church, land records, some probate, military), not to mention surrounding towns and in CT in general. None of the records come out and answer the question. I've collected a lot of records and would like to identify families--sorting what goes together from different sources, etc, etc. Hoping that the relationships between other COOKs and associated families will make the answer to my question clearer.

Recently, working on another family line (CARRERE, Baltimore, MD, 1800s, California 1900s), the same thing is happening. Lots of records (voter registration, passenger lists, directories, land, etc) and the need to sort out what belongs to who, with the goal of identifying the families and their connection (or lack thereof) with each other.

Do you use Reunion for this process?

Thank you so much for sharing your experiences!

Actually, cluster genealogy is not difficult, it may just be VERY time consuming. Nevertheless, it is very rewarding, because once you've processed a source, you rarely have to tough it again. With family genealogy you tend to consult the same source over and over and over, again.

First of all, I think key is that you keep your information in ONE (Reunion) file. Many years ago, I thought that it would be very smart to separae data across different files (or systems). Nowadays, there are only two types of files for me, first a (ONE) Reunion file, and second, source files i.e. thousands of them.

With cluster genealogy, you define the cluster first. What should be the cluster? A township like Greenwich, Connecticut? A defined geographic location is best suited, because the available records tend to follow similar borders. Which period do you wish to cover? 1640 until 1900? Should it include todays locations Cos Cob, Riverside, or even Byram? Define the geographical borders, exactly. Make sure that the geographical definition is in sync with your time line, which you plan to cover.

After you have defined your township, you should go out and obtain e.g. EVERY census record. One for every seven years, or whatever is available.

Now comes the big difference when doing cluster genealogy. You need to take a source-centric view. Previously, when you were doing family research and reviewed a census report, you only extracted the census data relating to your family. Now, when doing cluster genealogy you record every person, every relation from the census of that year, every age, every tidbit, regardsless of whether they appear to be related (to you or each other) or not. Most of them are not related, probably not yet. It doesn't matter. You need to enter every person, every relationship, every residence.

After you are finished with the first census, move on to the next census and repeat the process. Sounds terrifying isn't it?

Whenever you are tempted to enter a new person, check the index first, to make sure he/she is really new. You only want to have one person card per individual. Don't even think about doing a short cut and hope for a match-and-merge process to match all people later. Do a proper manual duplicate check before creating new person cards.

That's it for today. Let me know whether you've questions so far.

Regards
Reiner

theKiwi
17 May 2009, 01:00 AM
Are you using Reunion before you know who's who, and what records are talking about the same person? That's the piece I'd very much like to hear about.

Yes - I put everything I find into a single Reunion file, and if that results in unknown duplicates - eg a record of a set of parents from a birth record and a record of a couple from a marriage record - then so be it. As I find these and satisfy myself they're the same person/people it's easy enough to merge them and their sources together.

Roger

Linda G
17 May 2009, 02:14 PM
Nowadays, there are only two types of files for me, first a (ONE) Reunion file, and second, source files i.e. thousands of them.

Reiner, could you tell me more about the source files? Do you still use them? Are they Reunion files? I guess my question is, when you process a source do you enter directly into the Reunion file, or into some intermediate framework first?

If you had more than one cluster project (defined by place/time), would you recommend a Reunion file for each? Or one file for everything one does (cluster projects, 'normal' family research, etc)?

In my case, since I hope to eventually extend my family lines via these projects, I want to define a hopefully error-free process to migrate newly identified family members into my tree. My instinct has always been to keep to one Reunion file. But my instinct has also been to enter only individuals that I'm 'pretty sure' are mine. You all are using Reunion differently, with obvious success, so I'm trying to see beyond my 'instincts' right now.

Roger, thank you for responding. From your reply, I understand that you enter all info directly into a Reunion file and use Reunion's capabilities to identify and resolve duplicates. Have you found it useful to distinguish the research status of individuals (proved vs unproved, or however you might describe it)?

I can't argue with success. <vbg> I clearly have to get over my reluctance to muck up my family file and seriously consider using it for recording more than well-identified kin.

theKiwi
17 May 2009, 06:42 PM
Roger, thank you for responding. From your reply, I understand that you enter all info directly into a Reunion file and use Reunion's capabilities to identify and resolve duplicates. Have you found it useful to distinguish the research status of individuals (proved vs unproved, or however you might describe it)?

Yes I enter it all directly into Reunion, and give it sources to indentify where it came from.

Then in an ideal world when I'm doing more entering, I will realise/know at the time that some of the new stuff is duplicate, and so not create duplicates - for example I spent quite a bit of time at Scotlands People and extracted search results pages of all of the marriages of anyone called Moffat (by Soundex) Once I had that I entered them all into Reunion as couples, all separate from each other.

Then I got all the births of anyone called Moffat from Scotlands People, which resulted in Excel sheets that had columns of

childName, FatherName, MotherName plus other details like date and place

I sorted them by parent names to hopefully get all the same parents adjacent to each other and then started entering them- hopefully with the correct parents from the marriages.

So I ended up with a bunch of 2 generation families at first.

Then from other sources like headstone inscriptions, people writing to tell me etc I discovered that the say William Moffat who was a child in this family, was the same William Moffat who was a spouse in another family, so I merge them together, and suddenly have a 3 generation family - and so on and so on.

I still haven't entered all of my search results yet, but the end goal was to try and figure out as many of the Moffats in Berwickshire and Roxburghshire as I could, with the hope that at least then I'd know who my great great great grandfather wasn't, so that perhaps it might point me to who he was.

Roger

martha
19 May 2009, 01:36 PM
I think it's this--I've been using Reunion primarily to record individuals with KNOWN kinship relationships. My data management problem occurs before I've identified kinships (or convinced myself that I have).

Are you using Reunion before you know who's who, and what records are talking about the same person? That's the piece I'd very much like to hear about.

Linda, for me, this is very easy to do. My purpose, of course, is to restore on paper those hundreds of families who were murdered. So this is my step by step procedure:

1] I find some source with the name, and I immediately enter the person into my database;
2] I try to find all resources that might give me a hint as to family connections, any personal details, jobs, schools, community organisations, housing lists, clubs, sugar lists - interviews with people who left the community before the war - whatever I can find. All that information also goes into the person card in Reunion;
3] During all this research, I am also discovering the names and relationships or possible relationships of more people, so each gets entered into Reunion.

There are 3 designations I use before the name: [Probable connection], [possible connection], [unconnected]. Names which have no pre-designation are factual connections without a doubt. That way, in the index, all of the people in each designated group are located together and I can search for them rapidly that way. Whenever a positive connection is made, I remove the designation in brackets, or if [unconnected] becomes a [possible connection] based on some new evidence, then I change the designation. In the notes are all the reasons that led me to whatever I have noted within the brackets.

As connections are made, the people involved convert to regular parts of the database, with no bracketed prefix. Some people, alas, remain forever unconnected because I just cannot find out any information on them. But they stay in the database.

I hope this answered your questions. Everything is documented with footnotes within Reunion.

If you have any other question, please don't hesitate to ask.

Martha

Reiner L. Sauer
19 May 2009, 07:55 PM
Hi Linda,

Somehow, although I've tried to move the discussion to the title "Cluster Genealogy" it's always seems to appear under "events". (:-)

Anyway, here's my response to your questions:

Reiner, could you tell me more about the source files? Do you still use them? Are they Reunion files? I guess my question is, when you process a source do you enter directly into the Reunion file, or into some intermediate framework first?

If you had more than one cluster project (defined by place/time), would you recommend a Reunion file for each? Or one file for everything one does (cluster projects, 'normal' family research, etc)?

In my case, since I hope to eventually extend my family lines via these projects, I want to define a hopefully error-free process to migrate newly identified family members into my tree. My instinct has always been to keep to one Reunion file. But my instinct has also been to enter only individuals that I'm 'pretty sure' are mine. You all are using Reunion differently, with obvious success, so I'm trying to see beyond my 'instincts' right now.

Roger, thank you for responding. From your reply, I understand that you enter all info directly into a Reunion file and use Reunion's capabilities to identify and resolve duplicates. Have you found it useful to distinguish the research status of individuals (proved vs unproved, or however you might describe it)?

I can't argue with success. <vbg> I clearly have to get over my reluctance to muck up my family file and seriously consider using it for recording more than well-identified kin.

About the source files.
For example: In case you have the copy of a census record (e.g PDF), record the information as a source (1) in Reunion.

If this source is easily readable (and doesn't require transcription), then enter source 1 as the reference. Attach this PDF as a multimedia file to source 1.

If it is difficult to read, and should someone have transcribed the details from the census file into e.g. a spreadsheet, then record the information (.XLS or .DOC) as obtained from another source (i.e. 2) in Reunion. Attach this .DOC as a multimedia file to source 2. When quoting the transcription (which is someone else's work) use source 2 as a reference. Needless to say that the transcriber should be shown as the author of the transcription.

I recommend that you only work with ONE Reunion file. Don't follow your instinct in this case. Many years ago, I also split the data across different files. One of them included my family, the other one referred to the cluster. Then I noticed that my family tree overlaid the cluster and this resulted in hundreds of duplicate persons and even worse, hundred of duplicate sources. It took years to correct it, I'm still suffering from this early error. Keep your data in ONE file, there's no reason not to do it. The file can be easily split, but it's difficult to merge multiple files.

Remember, by using flags it's always very easy to separate/export persons based on whether they are part of your original family project, or your cluster genealogy. One thing is for sure, if you do it that way you always will have on source list, only.

As to the research status: Yes, you should set up a variety of child-statusses in order to mark the children.

After all, in the end you will be surprised who's all related to you. Don't discard them upfront. In cluster genealogy you record everybody and everything.

Regards
Reiner

Linda G
20 May 2009, 02:23 PM
Somehow, although I've tried to move the discussion to the title "Cluster Genealogy" it's always seems to appear under "events". (:-)
I know. I think the Leister folk have to do something. Or one of us have to start a new thread. As long as we can find our replies, it's fine with me.
Anyway, here's my response to your questions:

About the source files.
Thanks, Reiner. You're describing source files as I have them. I just wanted to make sure you weren't using Reunion (or some other software) in some intermediate phase before entering new individuals/information into your primary Reunion file.


I recommend that you only work with ONE Reunion file. ...One thing is for sure, if you do it that way you always will have on source list, only.
One Reunion file is my preference, too. One source list is very important to me, since I use the Reunion source numbers as identifiers outside of Reunion (file names, for example). It's very, very important to me that once a source has a Reunion number that I don't do something unintentionally to change it. Multiple family files certainly increase that risk.

Well, now my next step is to enter into my Reunion file all the info I've been accumulating for my cluster projects. This will be exciting! I've processed the info about as much as I want to into spreadsheets.

You and Martha have given ideas for how to identify these individuals (special child statuses, flags, special name conventions). It's not clear to me whether Roger uses ways to label individuals more likely to be duplicates, etc. But I have a variety of approaches to think about and try.

Thank you, Reiner, Martha & Roger, for helping me think this through! I really appreciate all that you've shared.

Reiner, I read your 'Wish List' posts from 2005. http://www.reuniontalk.com/showthread.php?t=252&page=1

I would also like to see in a future version of Reunion:
#1 Central place-name register
#2 Cluster Genealogy Report / Ortsfamilienbuch (OFB)
#12 Links to other persons in the Reunion file

Later, I'll add my support for these items in the wish list forum.

Thanks again!

Linda G
20 May 2009, 02:45 PM
I have two examples of how cluster genealogy and genealogy software can support each other to identify relationships:

Example #1 - "My Elisabetha"
I had two goals:
1. tie my gr-gr-grandmother Elisabetha Betz to an emigrating family thus identifying her parents or at least her siblings.
2. confirm the location in Germany of this Betz family so I could work it back further. (I had a region, but not a town.)

I noticed on Buffalo NY baptism and marriage records that Betz individuals showed up as godparents to Elisabetha's children and witnesses to her marriage. My software at the time already had a "Witness" field for marriages. I also created two custom fields for individuals:

a) Godparents of
b) Godparents are

This enabled me to generate a report of a list of godparents and compare it to those who had godparents (for a particular time period).


Jillaine, I've been thinking about this type of solution (creating fields to reflect relationship information to 'other' individuals).

For your example, how do you handle if someone is godparent to more than one child at different times/places?

It sounds like just knowing that they were godparents/had godparents and being able to see a list of godchild/godparent names was enough for you to analyze your data. Did I understand you correctly? Or did you end up having to add the fields multiple times for certain individuals?


Example #2 (i'll keep it short):
My husband's colonial ancestors included a lot of confusing overlaps, and confirming which Hannah Rice married which Richard Taylor was a nightmare complicated by a lot of unsubstantiated "published" data.

This sounds very much like my colonial/revolution era COOK family in Connecticut. Fun isn't it!

Troy
26 May 2009, 08:38 AM
Linda G. & Reiner & Others---

Just a positive comment on your use of "cluster genealogy"; I admire your efforts and your confidence in that regard.

I feel overwhelmed in just trying to research and find those folks related to the wife and myself.

You all should be commended.

Good luck with your efforts.

Troy---