I have been a longtime user of TMG and also of RM4 and recently upgraded to RM6. Main reasons for the recent upgrade (and my move away from TMG) are: 1) RM is a modern (though not as yes as complete) version of TMG 2) TMG is discontinued (although I would have stepped over sooner if RM was available in Dutch). A word or two on the last reason: I not so much would want to have an International (or Dutch) GUI but I need to have the sentences in (in my case) proper Dutch for publishing in The Netherlands. I have had an email exchange with Bruce offering my help in translating and hope it will come about one of these days.
So much for the intro.
My main point starting this discussion is my problems (in general, but in RM as well) with sourcing. A few examples:
- Everywhere you read about genealogy, in tutorials, in webinars etc. everywhere the importance of sourcing is emphasized (and I fully agree!). But as soon as you start using genealogy software it is either absent, hidden or hard to use. An easy example in RM is that the sourcing is under a button, whilst information like birthdate, birthplace etc. can be typed in right away.
- One wouldn't want to count what has been written about splitting and lumping. Not to mention the many non-productive hours people have spend on migrating from source to detail (or is it citation?) and vice versa.
- The way sources (or citations, or references or...) should be represented in publications is comparable to religious discussions. You are either member of "citation-church" X or Y and from then on you (blindly) follow the leader.
My conclusion after many years of reading the discussions and sometimes participating is one that is close to my long time profession as data and information analyst is: the problem is in the "mind-model". In my career I have noticed the strong influence of words on our associations. Words tend to narrow our way of thinking and since genealogy has been around long before computers, digitisation and the web came along, I would argue that we need new "words" to really get rid of all the sourcing discussions.
I suggest we start with "Information" and "Information Provider". A source is a kind of combination of Information, where it comes from and where it came from. We tend to structure the (for our purpose relevant) Information in fields in our database and store both the meta-information on where we found it (where it comes from, e.g. from a website or an online database) and where it was first written down aka "the original" (where it came from, e.g. a birth certificate made up in a certain town). Sometimes we also store the name and address of where we can find the original (although nowadays the original would not be accessible, after having been scanned, due to possible handling damage).
So in our modern world the genealogy software (that teaches us Genealogy without sources is like an Elephant without a trunk) would help us with:
- start with "the source";
- type in the meta-information of the source a) where did I find it (what is my Information Provider) and where did it came from (what's the origin). Of Course it would be helpful if both of the meta-information could be "remembered" to minimize typing and errors;
- identify all the people in "the source" relevant for our research with their appropriate role and enter them as such in our database-structure. Hereby it would help if the software recognizes that most "sources" tend to describe an event where people (and property) play a role;
- identify all the relevant relations (between people, and between people and property) and enter that in our database.
With the above procedure we accomplish:
1. all information is properly sourced (we started off with a source and if one is missing then the person entering the information from memory is the source);
2. no discussion about lumping, splitting or whatever. The (modern) software could could help us with structuring the meta-information in such a way that lookup, overview and analysis is easy;
3. no discussion about citation as it is implicit by linkage brought about by the software;
4. solves outstanding issues in RM about (alternate) name sourcing and relationship sourcing;
5. kind of simplifies and cleans up the user interface of the software (see 2 and 3 above);
6. citing models could even be more automatically be filled out.
A long post and if you managed to read it up to this point: thanks for reading. I hope it makes some sense and might start a discussion about a new view on sourcing and citing.