Can you look back at your TMG database and check if the example of wrong source type in your screenshot from RM is actually inconsistent with what’s in TMG? If it is, there is a serious problem with the RM import.
I thought parts of it were very relevant, particularly regarding source types not importing accurately and that there are steps you can take in TMG to resolve that issue, done with iterations of imports to resolve all instances.
I think the TMG import is still available in the 32-bit version of RM9 and is in RM8 and RM7. I think you can do a TMG import with them without a licence.
From the screenshots and discussion thread it appears to me that the issues you are looking at are not new and probably occurred with RM7 or before.
To add to what Tom and others are suggesting, it looks like a significant part of your issue lies with your TMG source templates and how they imported to RM7. Something seems amuck there and you’ll want to address that first. My guess is that you’re looking at some cleanup of your TMG data and templates and then reimporting. Other than that, the screenshots you provided describe common issues with the upgrade from RM7 to RM8 or RM9.
From the 1st screenshot, which I assume is the Sources window, and the TMG screenshot, it appears that you were an “extreme splitter” of your sources. Many RM users did this same thing because there was no way to identify citations prior to RM8. That, in itself, is not an issue or, at least, it’s the same issue that many RM users have had to address with the RM8 enhancements of Citation Names and Citation Reuse.
Which brings us to your 2nd screenshot. During the upgrade process from 7-8 or 7-9, Citation Names get created based on how citation detail fields have been defined by the source template. Based on your 2nd screenshot it looks like that source template had citations fields that did not have any data. This would cause the citation field to be blank. Resolving issues with how the TMG source templates import to RM7 may well reduce this issue. However, having blank or cryptic Citation Names is also an issue that many RM users have or will need to deal with when upgrading from RM7. There’s a lot of Db cleanup involved to take advantage of these new features. This is probably best done over time rather than all at one go. On the good news front, RM9 has better color coding and grouping features that will help you track db cleanup efforts like this.
Next, it also appears that you used the ‘Merge all Duplicate Sources’ and/or ‘Merge all Duplicate Citations’ actions at some point after your upgrade from RM7. This may or may not be an issue. You should be aware of the particulars of how the RM8 and RM9 merge process works - specifically that it does not consider media or weblink fields. So, source/citations with identical data except for a URL or media attachment will be merged. The fact that your sources are extremely split most likely means that ‘merge all duplicate sources’ is safe but you’ll want to check your merged citations by looking at citations that have been “used” multiple times and verifying that they are in fact 1 citation used as evidence of multiple facts or for multiple people.
Finally, the ability to view citation “uses” have led some RM users to consider becoming source lumpers rather than extreme splitters. This is something you’ll no doubt consider as you go through the db cleanup process.
Right, the RM source type or source template cannot be changed after the fact. And, sorry, I was never a TMG user and don’t have much to offer on that front. The last page of the doc seems to to provide the only other suggestion beyond maintenance routines. Others may have additional suggestions.
Once you’re able to sort the source import issue and if there are some field(s) that gets mapped into RM citation detail fields, then you should see data in the RM Citation Name field. There may still need to be some editing to make the names to be meaningful, but it will reduce the confusion of having blanks when you look at all citations for a given source.
As I mentioned above, anyone who was a source splitter will have RM9 sources that look like citations. If your imported sources end up being split, that’s not worth worrying about. As long as your data is all there and your field labels, sentences and roles make sense, then I think you could declare success. Going forward, you may want to become more of a source lumper but it’s a lot of work to go back and change older sources.