Three times in the last week I have been caught up in the paper vs. digital controversy. My perspective is that the last time I calculated it, one paper copy of my primary data base would run over 80,000 pages! Of course, there are errors and all sorts of duplicates in this large file, but the errors and duplicates are nothing compared to what they would be if I had done all that work on paper. As an example, my Great-grandmother worked on genealogy for about 30 years and in examining her files, it is apparent that she duplicated all of her work about three times.
The duplicates in my Great-grandmother's files are not a reflection on her ability, it is simply impossible, absent some sort of card index, to manage any genealogical research once the number of people in the file reaches into the thousands. In my opinion, if you have from 0 to 900 or so individuals in your family file, it makes no difference at all whether you keep your information on paper or in a genealogical database program. I realize my critical mass number is totally arbitrary. In reality, I just picked it out of the air. I admit that if you are super organized and meticulous, you may be able to manage a paper system with more individuals. But if you are like me, anything over 10 is a challenge without a computer.
By the way, in either system, the key to organization is the system. If you have no systematic way of entering your individual's information or for associating sources, then your lack of system will break down and you will end up wasting a huge amount of time and effort. The challenges with a paper based system are the "one-to-many" and "many-to-one" problems. I have written about these before, but they bear further analysis. An example of "one-to-many" is a U.S. Census record listing one or more related families. That one record has a relationship to more than one individual i.e. one document to many individuals. The opposite is having a number of documents that relate to one individual. Unfortunately, you can also have a mixture of the two, multiple documents that refer to more than one individual.
In a paper system adequately addressing the one-to-many and many-to-one problem is part of the reason why computers exists today. Genealogy is fully as complex as any corporate organization chart, government bureaucracy or any other complex organization.
If you try to address these problems with a paper system, you can have a separate file for documents and you can even make a source reference for each individual to the paper document in your "U.S. Census Records" file. You can also have a unique document number on each paper document. Now you remember the 80,000 plus pages I reference above, how many documents would I have if every one of my families also had a couple of source documents?
So you see, the paper vs. digital issue is really like quantum mechanics, it only works at very small levels. If paper works for you, then fine, I think that is great. But my challenge has been amassing huge amounts of paper without making the time to catalog all of it and associate the facts with the individuals either on paper or in the computer. In theory, whenever you found a source you would immediately enter the full citation of the source into your database and link the source to all of the individuals mentioned in the document. I say in theory, because you may get documents in the thousands like I do with almost no hope of doing the cataloging.
Now, where does the paper system start to break down? Even assuming numbers are not the issue, individual paper systems become very redundant. In the one-to-many and many-to-one issues, you will begin to see the inefficiencies of the paper system almost immediately. As an example, how do you show that a relative appears on several different contradictory source documents in a paper system without resorting to a manually created timeline or a narrative report? On the other hand, most of the genealogical database programs today will create a timeline and incorporate data from multiple sources.
I also realize that the propensity to prefer paper systems is based on emotions and not reason. It may also have a large measure of techno-fear at its base. I will come back with some ideas about meta-organization of genealogical records in the near future.