Jerry Bryan

Jerry,  You have outlined one of the things I have against Evidence Explained.  It creates a lot of “templates” or use cases, and it can be very hard to decide.  Don’t get me wrong, EE is very good to use and understand the importance of sourcing the right stuff, but can be overwhelming.

I agree. I do use Evidence Explained all the time, but it's 1200 pages long. That's just too many pages and it's totally overwhelming.

There is also something else wrong with Evidence Explained that I find hard to describe in general. I can only describe it by specific examples. For example, you will search in vain how to cite obituaries anywhere in those 1200 pages. Rather, you will find how to cite newspaper articles and you are expected to cite an obituary as a special case of a newspaper article. These days, obituaries sometimes only appear on a funeral home Web site and sometimes never appear as newspaper articles. So citing obituaries using the newspaper article model sometimes doesn't work. But to me the deeper issue is that it is the obituariness of obituaries that is important, not the newspaper articleness. In contrast, Evidence! is an earlier and much briefer version of Evidence Explained, and Evidence! does explain how to cite obituaries as obituaries. So I think there is something about what the true essence of evidence is that I think that Evidence! gets right that Evidence Explained gets wrong. I suspect that if this mysterious essence of sources could be gotten at accurately, that a comprehensive guide could be more like 100 to 200 pages than 1200 pages. Indeed, in a certain sense I'm describing Evidence! rather than Evidence Explained.

For example, I'm not persuaded that Web sites are ever sources. I'm not sure what they are exactly, but it seems to me that Web sites are more like repositories than they are like sources and that the underlying database or record set on the Web site is the actual source. So it may not make much sense to describe how to cite Web sites as sources. Rather, it might make more sense to describe Web sites as repositories.

Jerry

So, the interview templates are useless to me. I will have to make my own Interview template. Thank you for clarifying that in my mind.

It took me rather a long time to get my own source templates into pretty good shape. One of several advantages of having my own is that I could change them and improve them. You can't change RM's templates in any way. You can copy them and change the copies, but users don't tend to start out that way. Then having chosen a particular template for a source, they are stuck with that template without a major do-over of that source.

I don't have an Interview template. Rather, I have a Person template. I use the Person template if I interview a person face to face, if I interview a person on the phone, if I receive an email from the person, if I receive a postal letter from the person, if I receive a text message from the person, if I receive a message from the person on Facebook, etc. Hence, email is not a source. My telephone is not a source. Facebook is not a source. Etc. It's always the person who is the source.

I suppose that if you carry the Person concept to an extreme, you don't even need a Book template because you could use the Person template with the author of the book being the Person. But I do have a Book template. I have something like 20 or 25 templates, rather than the several hundred templates supported by RM. I find one of the hardest things about using RM's templates is knowing which one to use, coupled with the virtual impossibility of changing templates without completely redoing sources based on a particular template. So where RM has many Book templates, I only have one Book template so I know which one to choose. RM's Book templates have different options depending which template you choose. When I need options, I try to incorporate them in the same template. I do have to choose options sometimes, but I seldom have to choose templates because there is only one choice.

Jerry 

You could have made your sentence template be "Eye color. [Note]" without the quotes and just left your information in the note field. Except that RM includes the note field automatically and does not actually have a [Note] variable. Therefore, your sentence template could have been simply "Eye color." without the quotes. The sentence template looks funny without the reference to the note field, but it does work.

I have a number of "note only" fact types myself, and they do work. However, they do have one very ugly characteristic. Namely, the citation superscript appears after the body of the sentence and before the note. So you get such things as Eye Color:²³ blue rather than Eye Color. blue²³.

I use SQlite scripts to supplement my work with RM. Were I in your situation, I would have written a very simple SQLite script to move the Note data to the Description field for the specific fact types that needed that kind of data manipulation. If you are up to a little programming, then a simple script can sometimes save you dozens or hundreds of hours of data entry.

Jerry

Your analysis is correct in that converting sources as you describe is an incredibly labor intensive task. I have been working on the task in my database for years, and I am not even close to be being done.

In my case, the issue is not converting from a third party software's version of sources to RM's model. Rather, it is converting from RM's Free Form sources that I had already entered by hand into RM to using RM's source templates. Part of my project has been to move all sourcing data into RM's Master Source - the yellow area on RM's data entry screens for sourcing data. I'm using only the yellow area rather than the green area so that any correction I make will be applied immediately to all the citations for the source rather than requiring me to chase down each citation and make the same correction multiple times. To that end, I developed my own source templates, similar to RM's except that they place all the data into the yellow area of RM's data entry screen for source templates.

RM8 will include a feature which I think will be called Shared Citations, although I am not sure of the official name for the feature. The feature should allow a given citation actually to be shared multiple times rather than being Memorized and Pasted multiple times. After RM8 is released, I will need to examine this feature very carefully to see if I wish to start using it or if I wish to continue doing my sources the way I have been doing them.

Jerry

I would go even further than KFN in one respect, although my understanding of the genesis of the GEDCOM model may not be 100% accurate. I would suggest that the GEDCOM model for sources was set up primarily for books, where title, author, and publishing information was the "source" and the page number was the "citation". Indeed, the GEDCOM tag for the "citation" is PAGE. Users of genealogy software were then stuck with mapping other sources such as birth certificates, newspapers, and census records onto that model. There is more than one reasonable way to accomplish such a mapping and different users and even different experts may come up with different but reasonable ways to accomplish such a mapping.

I have come to think that a "citation" is really a footnote sentence as a whole, whereas a footnote sentence is usually viewed as footnote sentence = source + citation. So there is an immediate conflict where the citation is either just a part of the footnote sentence or else is the footnote sentence as a whole. I am struck by the fact that many genealogy sites these days will provide you with a complete footnote sentence that you can copy and paste into your genealogy software, except that if the software follows the footnote sentence = source + citation model then there is no right place into which to paste the completed footnote sentence. This kind of copy and paste certainly doesn't work with RM's built-in source templates, and it doesn't even work very well with RM's Free Form source template.

What the original question is asking is about the rationale for entering data into what RM calls the Master Source field (the yellow area of the data entry screen for source information) vs. entering data into what RM calls the Source Detail field (the green areas of the data entry screen for source information). There is not general agreement about the answer to this question. If you use RM's source templates, they answer the question for you and you may or may not like the answer. If you don't like the answer, you can define your own source templates. Or you can use RM's Free Form source template, but even using RM's Free Form source template constantly begs the question of which of your sourcing data should be entered into the yellow area of the data entry screen for sources and which should be entered into the green area. For every expert about how to do this, there is another equal and opposite expert.

Back to my suggestion that a "citation" is really footnote sentence as a whole. I would similarly suggest that a "source" is the book or piece of paper that you can hold in your hand, or images from a book or piece of  paper that you can hold in your hand. Whether and how to break the footnote sentence as a whole into smaller parts then becomes a question simply of how you want to manage your list of sources so that you can find things easily and so that you can reuse parts of existing sources for new sources without having to retype everything from scratch.

Jerry