We encountered an issue while pushing out our production system today. We have currently loaded around 50,000 files via a DIP import from our 20 year old legacy EDM system. We incorporate document revisioning and versioning and this is the base of the problem we are encountering. What happened is that when data from our legacy system was loaded into OnBase it took any existing titles and added them to a keyword called title. This way we could use the keyword within our auto-name string. What we did not anticipate was multiple instances of the title for a group of revisions.
So for example we have might have 5 revisions for 1 unique document (version) however, in the last 20 years, over time the title of this document was changed as it was periodically revised per OSHA and ISO standards. Our old system treated each revision as its own document so it captured these changes. So now when I go to load these 5 documents 2 may go by one title and the remaining 3 may go by another. How DIP and EDM revisioning works is that it correctly links all 5 documents, sets 4 as old revisions and the 5th as the current version. However the system does not differentiate which keywords belong to which revision. It simply loads all keywords in a "pool" for every revision.
This brings me to my issue. So we use a combination of an alphanumeric Document ID and the title in the Auto-Name string to identify these documents to our end-users. This works great except for the fact that if the legacy system contained multiple keyword instances for title the Auto-Name string simply uses the first title in the list (I assume they are put in alphanumeric order) and because of this we have no way of telling if that title is the correct title or if it was decommissioned and changed 15 years ago.
Our 3rd party LOS gave some good advice and pointed out that the keyword value in the auto-name string could actually be modified to display the other instances of the keyword (if they exist). This was done by changing the suffix of the keyword value from a ".1" to a higher number. We settled on ".4" in the assumption that we could potentially have up to 4 instances of a title somewhere in our 50,000 documents.
This idea actually works really good as the system will only display the 2nd or 3rd values if they exist. However, my issues is that it will separate the multiple instances with a "-" (Dash). Our title use multiple dashes and this makes it hard to tell if multiple titles exist or if the name just incorporated a dash. This seems to be some type of default behavior as I use the "|" (Pipe) as a separator in my auto-name string. Is there anyway to change this default behavior and change this value?
