williwa:
Your need to create custom Source Templates is fortunate. So many use built in templates and then discover they don’t format quite right or some other limitation. Because the built in templates cannot be modified, it is in your best interest to create custom templates for everything, even if it is just a copy of the built in because you will be able to modify it if you wish.
This is just me, but every time I need a new template, I will copy a built in template and prefix the Name with a tilde (~). Then I modify it to my liking. Whenever I need another template, the template list is populated with all the prefixed tilde items at the top, making it easy to find the ones I am looking for.
If you are looking for advice, and it sounds like you are, I would also recommend parsing your Source Citations into as many fields (items) as you possibly can. This will give you maximum flexibility in the future. You may want to consider if you want to be a Splitter or a Lumper in the near future. Here’s the way I differentiate between a Splitter and a Lumper. If you make a list of all the ITEMS you want to cite, in the order they will appear in your citation, you might have a list that looks something like this:
US
census
1920
population schedule
jurisdiction
enumeration district
page
dwelling #
family #
line #
person of interest
Somewhere in that list will be a dividing line between the Source Name and the Citation Detail. Where the template places that line, which is your choice, determines whether you are a Lumper or a Splitter. If the line is near the top, say after 1920, you are a Lumper. You are lumping every citation to the 1920 US census into one source of the entire 1920 US census. Some might prefer the Source to be at the Jurisdiction level and would be Lumping to each County in the 1920 US census. You can easily see that this person would have a longer Source List by Lumping at this lower level. The further DOWN the list that the split occurs, the more your Source List grows, and that frightens some people. 
Personally, I am a Splitter. I am an Extreme Splitter when it comes to single event documents like a birth or marriage certificate. I do not Lump those with anything about them. When it comes to the census example I consider the Source to be the Page I am looking at, but consider each individual’s enumeration a separate event. Therefore, my line goes right after Page, and the items below Page go in the Citation Detail section. Many people just use a text field called Citation Detail, CD, and enter the data as text.
The Lumper might look like this:
“Morgan County, Ohio, enumeration district (ED) 23, page 33B (written), dwelling 137, family 144, line 16, Israel Embree”
Mine, as a Splitter would look like this:
"[dwelling#], [family#], [Line#], [Person of Interest]’
Now there are two things going on here. I am placing less information in the CD AND I am parsing the information into data fields. Whether you become a Lumper or a Splitter, I highly recommend parsing as much as possible. There is a reason behind this. Let’s say after many years of data entry, there is a new Citation Guru that has come up with a BETTER genealogical citation for census records. The only problem is that their format requires the person of interest to go before the line number instead of after. For those who have the first CD above as their format, they are going to have to go to EVERY census Citation and edit that information, one at a time. If you have parsed it, as I have suggested, you only need to edit the Template and all is corrected. I speak from experience.
Let us know if this has triggered any other questions.
Best,
Paul
