Question Basic questions for beginner's genealogist

Hello all,

I'm new to genealogy, and I would like to take advantage of this post to ask you some basic questions.
I have been impressed relative to the family trees you were able to manage (relative to the link some of you have provided in your signature)
Really impressive !

Question 1
Today most of the archives are available through the Internet.
This allow us to find a lot of information.
I have found that when looking for sources, I often navigate between several dated books. (Register)
And sometimes several times to the same Register ..;



In some of the Register, I find names who will be part of my family tree, (at this time I note the date of the register)
In others register I cannot find anything relative to my family tree, so these register have to be eliminated from the consulting list .
So to my question :
How do you go about the referencing ?
In fact, and in order to be more efficient, where do you reference the registers that you have already consulted ?
For the register which are "positives", sure the sources will be referenced through webtrees, and we can consult via the "list Source" but for the "negatives"... ?

The objective being to sort out what has been done, what is being done, and what still needs to be done (by deduction)?
Are you working with an Excel sheet to help you with to manage this approach?

Any advice appreciated
Thank you

Christian

Good question...

First lay out the structure. I'd say the archive / website is a Repository (keeping in line with physical archives these are separate repositories per village, town or church) which each of the books / volumes / year registers being a Source from which you can extract Source Citations which gives evidence to a fact (there is someone with a name, related to someone else you know, that is a fact) or event (birth, death, marriage, baptism and burial being the most common you dig up from the archives).

It would not make a lot of difference whether you do research online or at a physical archive. But if you're doing it systematically and you're not doing it alone or if your memory is not reliable, then you better make notes. But it can get out of hand if all my ancestors come from one place. With three generations per century in average, the people of 200 years ago is 6 generations so 2^6 (father & mother per generation) = 64 different surnames to check.

No, when going through book indexes in the hope to find a familiar surname somewhere, you'd do that for one surname only. Otherwise you cannot concentrate, me thinks...

Important is to keep notes and work systematically. Luckily in webtrees you can stick notes to about anything.

And don't forget your personal "my" page! I've made HTML blocks with tables & links to all my grandparents with a link to their fan chart, and keep track in that table how many of their 8 great-grand-parents I had completed the research on. So I can see at a glance what still needs to be researched without flooding myself with _TODO tasks too much.

Does that help?

I am not sure I am being helpful, but in case illumination of the breadth of the problem, or rather the uniqueness of each individual genealoger's case, is useful, I go ahead. My answer is, no I don't.

I have very little use for reviewing register by register, because I cannot remember my entire family tree, and therefore would not necessarily recognise a given family member. I will have to revisit the source with a different branch in mind later. Even if I do remember all the known branches, there will be more known branches the next time I visit. (And yes, I do find that my parents have common ancestors a couple of centuries back.)

The second problem is that names are not unique. It is impossible to recognise a given individual based on name alone. Surnames were not in common use 200 years ago. Consistent spelling is only a product of legislation less than a century old. The only two viable approaches are to type up and connect every individual in the village, or look for particular individuals which you hope to find.

Third problem is that evidence is rarely definitive. I work a lot with secondary sources, solid genealogical work much of it, but the genealogists do not agree. Sometimes there are errors, inevitable when you compile hundreds of pages of data without the help of computers. More often the evidence is scant, and the genealogists have had to guess, sometimes educated guesses, other times wishful thinking. This also means that sources have to be revisited. Exhausting the interpretations take a life time.

Of course, if you start with your great grandparents and research their descendants, your problem is very different. Then you have official legal names with consistent spelling.

But when you decide to start with your great great parents, you have to start over with all the sources.

I find there are far too many sources to review them all systematically. Yet another reason why I just search for missing links in my existing tree.

:-- Georg

YES that helps
Thank you.

As a first approach, I thought that a mind map could be benefit.. to point out such information
For example : To highlight the registers which have been consulted, and the one which need to be
But I agree with you, there is no rule.
All the more as you say, it is not because a register has been consulted that we will not come back to it later..

This make sense.
However, we can see the approach by layers, and in that approach the mind map can help...

I understand that notes are mandatory, so the approach via a mind map could be a solution too..

I will see..

For now, I selected gitmind
I will see if this is beneficial, or a repetitive task to webtrees.

Christian

Mindmaps I can believe in. I use them to some extent, but only small scale on paper.

Another tool. I am not sure if it still exists in webtrees or if it is only a faint memory from its predecessor phpgedview, is the research task feature, where you can record open questions which requires research.

These are not systematic approaches to exhaust your sources, but good ways to avoid forgetting good ideas when I have two ideas and can only pursue one at a time.

> Another tool. I am not sure if it still exists in webtrees or if it is only a faint memory from its predecessor phpgedview, is the research task feature, where you can record open questions which requires research.

It still exists.

1) add the "research tasks" block to your home page.
2) enable the _TODO (research task) event in the family-tree preferences
3) add research tasks to your individuals.

fisharebest wrote: It still exists.
.

YES, this is a Great help. I'm using it a lot


However, the quest here was more about to have a possibility to record general tasks, notes, not linked especially to individual.

So you can't make tasks other than under one individual? I thought one could, but then it was only a faint memory from eons past.

Here is a concept that I've been mulling around for some time. I often help people to find information about their family and need a better way to keep my searched list.

You could create a second "tree" to webtrees just for you documentation of the books you have reviewed for surnames.

Add a individual using only the surname you are currently researching.

For example McDonald (no given name) .
Add a _TODO task named Books Reviewed or "xyz Book Collection Reviewed"
Add a NOTE with a list of the Books in the collection and a Yes/No after it to indicate that you searched it for mention of that particular surname.

norwegian_sardines wrote: You could create a second "tree" to webtrees just for you documentation of the books you have reviewed for surnames.
.

Yep ! Could be a solution.
Perhaps not the most appropriated, but a solution

Finally I find that the principle to use a "mind map" is not bad..
Sure it is redundant to information that we find in webtrees,..


But it can give a good overview of what has been pointed out, what needs to be pointed out, and what has no interest.

Of course, everybody to find the best approach, and Open to everybody

Hello all,

I'm back after having tested the "mind map" orientation several days
Honestly , the current status is that for now I'm not really convinced if this brings me more pain or help..
I will continue testing.....

In parallel to this topic, I have another question about :
How are you handling the source.

This is one of the most important in genealogy.
We have to keep in mind that : "Genealogy without a source is a mythology!"

So my question is relative to
* Repository
* Source

Repository is the location where the information can be found
* Internet
* address of house
* ..

Source
The Source points to the element among all those found in the Repository

MY CURRENT APPROACH (or FIRST APPROACH)

Example:

Repository
Here about a repository which is pointing to Internet : "Linked to the civil registries of town Soultz-Les-Bains "
(Location : Internet URL to this library or repository )


Here an overview of the link.



Source
The Source points to the element highlighting the event
By analogy here "a Book" or "Volume"


The Book records are hand-written books which have been digitized
The fact here is that each Book, or volume covers ONLY 1 years,


Media Object
Of course I'm completing the source by Adding a copy of the original document that I'm attaching as Media Object

Now to my question:

The consequence here is that I will have an endless list of "Sources"


Also, meaning that I might have as many sources as individuals, and much more..

Does this approach make sense for you ?
Is there a limitation to the number of sources ?

SECOND APPROACH

OR it is better to address this issue by limiting the number of sources , and as following

Repository which is pointing to Internet : "Link to the civil registries of town Soultz-Les-Bains "
Same as "FIRST APPROACH"
..... Link

Source : --> pointing only to One location = "Civil Registry"
In our example
...... Source : Soultz-les-Bains - Etat civil

Citation details --> Pointing to the "Book + Reference in the Book where we can find the information"
....... Citation details : Soultz-les-Bains - Etat civil - Registre de Naissance 1845 : Registre N° 34


QUESTION
What is your advice:
* First approach ?
* Second approach ?

Fact is that the "First approach" should be the one which is the closest to the definition of GEDCOM
However, in my case, it will create a lot of entries for sources...
Perhaps this is definitely not an issue

Does that make sense for genealogy experts ?

Thank you for your comments and advice


Thanks

Some sense, but it would be useful to know the nature of your sources, i.e. are they essentially hand-written protocols (which have been scanned separately)? Or are the digital registers collating or structuring information from multiple protocols or parts of protocols?

In my opinion, the REPO/SOUR model makes sense only when you visit physical repositories and read unique copies of the sources. Digital records are so easily duplicated that there tend to be many REPOs, and it so much easier to reference and recover individual volumes and records.

I'll give you an example. Norwegian church records are hand-written books, each volume belonging to a single office. Usually the minister keeps records, but sometimes chaplains and vergers kept records as well, the verger typically duplicating the minister's records.

Thus there is a physical book which can be consulted in a reading room at the appropriate state archive. Each of the archives would be one REPO and each volume a SOUR, I think. Each volume covers many years, so I usually I find more than one ancestor per volume.

Now, each volume has also been scanned in an online repository. That, to me, is the same SOUR, but a second REPO. Interestingly, this is not a problem, GEDCOM 5.5.1 allows multiple REPOs on one SOUR.

Many of the records have also been transcribed in searchable format, available in the online REPO. This is a new SOUR with, strictly speaking, different information. It is of course supposed to be a copy, but human error happens and sometimes the unstructured hand-written form hides information which does not fit into the database.

One more feature is important for my decisions. The online REPO has changed at least once in the ten years that I have had an interest, with the result that URLs noted 8-10 years ago are no longer valid. *sigh*

In all this, the physical volume is the most stable reference. They can be moved to a different repository, but they are unlikely to be destroyed or even split up, because of their historical value. As long as I have recorded the name of the volume, I should be able to recover it, regardless any reorganisation of the national archives, in the virtual as well as the physical world.

As part of the source citation (pointing to the descriptor of the physical volume), I create a media object with a URL to the event in the searchable database. Easier lookup but less persistent. If I am my most consciencious mood (or the case is particularly tricky), I look up the scanned version and make a copy of the page, as a locally stored media file.

Now, I still have a problem, because I struggle to fit the descriptor of the physical volume into the GEDCOM fields, and I struggle to keep the scanned and transcribed forms separate. Quite simply, keeping them separate takes a very dedicated and consciencious mind, and mine usually falls short.

Now, I do not know how this correlates with your French, sorry I should only infer francophone, records!?

Thanks Georg.
Based on your reply I have updated my topic above.
Hope it is more clear now.

My fundamental question is to know if my FIRST APPROACH makes sense, or if I'm completely wrong

Yes, that approach is not nonsense. But equally so for the other.
I'm terribly sorry if this does not help you make a decision at all.

It doesn't matter what approach you take, it will never be perfect and it's always a compromise.
There will always be people that will criticise your approah, and that person will eventually be you.

I think I went through all the stages, here neatly ordered from one extreme to the other but in reality going back & forth:
- create a source record for a single page of a book where the book is a repo
- document scans stored on my own server even though you can find them very easy online elsewhere
- create a source record for the book (page is now just citation in an event) and the archive the book can be found in is a repo
- source citations contain full text transcripts
- create a source for the town archives without any repository
- oh sod that all with image scans and text transcripts, its too much work and no one will ever see my hidden sources. Whoever wants to check it will know what archive to visit to simply check that fact.
- besides, I'm getting false search hits when searching names in my own data! Better keep it a bit more clean: names should remain in Source records and not bleed through to people's data
- citation detail contains just the url of a webpage, linked to a source record named "The Internet". Two years later: page not found. Well, too bad...
- not citing any sources at all

In the end the only thing that matters is that your work can be verified. Even without citing any sources, it can be verified! But only with greater difficulty

bertkoor wrote: It doesn't matter what approach you take, it will never be perfect and it's always a compromise.

Thanks !
Sure I cannot say that it helps me a lot to fix my issue..
BUT, all has been said here:-)

Honestly this topic gives me headache ..

However, your answer is really helpful because it informed me now..., that there is no rule

I agree with Bert.

And since what you essentially want to avoid is your future self shouting at yourself for taking a shitty approach, it is worth imagining what is going to happen, with each approach, when your problem changes and the tree grows.

It is worth considering the kinds of reports you (may) want to make. Personally, I run my own scripts to set reports in prose, and generally there are two ways to set the references. Either as footnotes/endnotes or as a reference list. If you use footnotes, it does not matter the slightest if all the sources are unique. Each citation is a footnote. If, however, you use a reference list, it becomes bulky if you have a large number of sources cited only once. This is a case for your second approach if and only if you (might) use a reference list. If we are considering this kind of system, whose environment has a multitude of local maxima, we cannot understand the system unless we know something of the method and history of its evolution.

The other scenario worth imagining is that the current database which you use closes down. A different may open, but using a different system. How can you maximise your chances of recovering a given SOUR entity from a new provider?

If you can deal with all such imagined scenarios, I think you are safe. Thus the rest of my post is based on how my mind works, and not what will make you a happy genealogist.

IMHO each source should be the smallest possible entity which could meaningfully be distributed separately. This is easy to see with books. An antology, or an encyclopedia, where each paper does not depend on the rest of the book, the paper is a source. For monographs, where the chapters constitute a whole, the book is a source. If we consider original church/civil records, it is also simple. Nobody in their right mind would rip the book into individual pages, but once the book has been scanned, we have a different kettle of fish entirely.

Admittedly, I struggle with multi-volume books, but since the volumes can be borrowed individually (and individually do get lost) at the local library, I have arrived at the conclusion taht each volume should be one source.

What it boils down to, really, I think is that the SOUR/REPO/OBJE model is broken. A hierarchical system of sources would have been simpler, where the chapter and the book are both sources, one linked to be a component of the other. But we are not rewriting the GEDCOM standard, are we?

I agree with Bert and Georg. Nevertheless here my thoughts on it.

I would prefer your second approach. I would use And not one source for every year. Why?
If there is a physical book as a source, then you can use a page number and line number as an index to find any entry in this book. If your Internet-Source has a consistent structure you can use the year and a page and a line number as an index. So whenever there is a uniform structure I combine sources and put the details to the index.

I would not combine different Registre into one source, as for example because these different Registre may have different access rules or years.

About the repository: for me, a repository is a house (archive, my home, ...) or a web-page. This is the same as you have done.

First let me say that I appreciate that you are thinking about how to record Source Material, most people don't use much sourcing let alone think about how best to record the information.

I'm not sure that the REPO->SOUR->Source_Citation is broken per say, but it has a few issues. Remember... The Source is the high order information and the citation is the specifics within that source(chapter, page, paragraph) if you document that information.

The whole reason for recording the REPO/SOUR/CITATION information is so that you or more likely someone in the future, can find your information again to review you findings. With that in mind I like to set my SOURce records up in a way that reflect the most obvious way to refer to the source so that when (not if) the repository that currently has the source either goes out of business, reorders/rearranges the material or the material is totally lost, you then have a way to describe what you used to find the data and someone somewhere can say, "Oh I have a copy of that in my basement" -or- they have a different format of the same data but not an original.

While I do believe that the REPOsitory is very important for faster retrieval, describing the actual source in a why that helps the researcher locate the data either the original or a transcription is more important.

So a complete description of the census, not just the year, but any and all location codes.

georg talked a little about Norwegian church records being hand-written books, in most cases they started out as ledgers divided by the event of thing being recorded, some events had separate ledgers some ledgers had two or three parts for one for each recorded event. As time went on this locally controlled information moved to regional storage and the separate ledgers were combined into books. Even later the data moved again to even larger repositories and was digitized and indexed to help researchers find things by year and event (but the books and ledges remained the same) In these cases I tried to identify the original source material so that when the information was again reorganized I could ask my friends at the state repository what I was looking for and they (via their finding aids) could look for the item or tell me where it was stored.

That is the best advice I can give. "Record your source so that you (or someone else) can locate the information in the same format -or- some alternative format."

So first of all, many thanks for all these different answers.
Your different opinions are always very interesting and helpful.
It helps me a lot to progress in my approach.
It's been a while that I have adopted webtrees, and that I'm trying to settle down the rule foundations for myself, and of course especially for the people who might help me to contribute for my family tree.

In the approach, I'm always trying to respect the general Rule (in our case to match GEDCOM philosophy) but also be open minded, and I'm must be convinced that what I'm doing makes sense.
I must confess that I'm also a partisan of KIS : Keep It Simple.

All these rules must agree on equal terms
So, here we are about the issue.
In my mind there is something which is not matching this equality..

I know that, even I have approximately 50 members in my family tree, it is still not too late to adjust the rules that I'm trying to respect and which have been highlighted here (in French)


Before to overcome to the new issue:
Fact is that for now I always found a solution to the several issues I faced, thanks to the different possibilities of webtrees .
Example of issues fixed
* different names of people
* Media Object
* Photos

Now coming back to my issue.
Me too I'm convinced that the REPO->SOUR->Source_Citation is the right approach.
It make sense to indicate the location where the source have been found.
I have no doubt about that.
Once that clear, I have to adopt webtrees (or Gedcom) constraints
To give a source of a "Document" implies to follow : REPO->SOUR->Source_Citation : TEXT : Media Object

To the definition:
REP -> the place where the source was found
SOUR-> Book, Volume, Document, Album .. (The Source points to the element that interests us among all those found at the REP)
Citation details -> the reference point at the Source (Page Number, Paragraph, ..)
TEXT: Is optional for me -> Verbatim information of the event
Media Object -> Is mandatory : A full copy of the Text in order to keep a backup

So, reading all your proposals, the solution that I will adopt for the SOUR issue, is to include all the books highlighting a particular event such
* Wedding
* Birth
* Death
without specifying a date.

Reason to not specify a date : because next year, our government will add an additional year to the registers which can be consulted online

Conclusion overview of the solution :

REP: "Archives Départementales de Soultz-Les-Bains" with the URL link to Web location
SOUR : Soultz-les-Bains – Etat civil – Registre de naissances
Citation details : Registre de 1895 N° 34
Media Object : Paul Klugesherz 26/08/1895 : Registre de Naissance

Same in English
REP: "Departmental Archives of Soultz-Les-Bains" with the URL link to Web location
SOUR : Soultz-les-Bains - Civil status - Birth register
Citation details : Register of 1895 No. 34
Media Object : Paul Klugesherz 26/08/1895 : Birth register

Conclusion for now: I have once again to modify my webtrees rules highlighted in Wordpress ...

Don't hesitate for any comments
And once again Thanks for your hep