Generally things work as desired now. Many thanks for your advice.
I have a question. After getting bibliographic metadata for all my files I searched for a specific journal name and found 8 relevant hits. I knew there should have been 41 hits. I looked at the Info:data for these files and the non-hits did not have any metadata entered. Now, all of the 41 files had the tag “suspect journal”, so I then put the tag ‘suspect journal’ in the Download Bibliographic Metadata program and re-ran it. I then searched MDJournal: “name of journal” and got 38 hits – still missing 3, but lots better after adding the tag. The metadata were there in those files. The only thing I did was add the tag name. The remaining 3 that did not get data were not that different, but were quite old (pre 2000) and maybe some other factor was working for them.
QUESTION: Why did this result in much better Bibliographic pickup?
So, the suggestions given were very helpful, but, as usual, lead to more questions.
Still, one of the best programs I own. Many thanks.
Regex functions are available only in the actions of smart rules, not in the selection part.
And for good reason: a RE can be sensibly run only on the raw text or a larger portion of it. That is a possibly time consuming operation – nothing you’d want in a smart group.
DT‘s index is case-agnostic because that makes sense. Were it to treat „Nature“ different from „nature“, it would find one at the beginning of sentences, the other one not. Hardly useful.
It took me a while. Please keep in mind that people on his forum are from many different nationalities and cultures. Some are rather more direct than others . An Englishman might say, “maybe you should think about this” whereas others might say “why haven’t you done this?”
Direct bluntness is not a problem. Bold and pointed presumptions are. I may indeed find use in distinguishing terms that start a sentence using a smart group search. I may accept the trade off of waiting a bit longer for a REGEX search in specific cases.
I’d also suggest that one should step onto a world platform with a geniality of wanting to learn how to be respectful in all cases, not with an open expectation to be accepted by some trick of cultural training for being obliviously rude in certain cases.
I’m not a native speaker. Where I used ”you“, I probably had better written "one“.
The indexing in DT is the way it is because they want it to be fast. And I suppose (!) that’s what many users want. Considering upper/lower case will make it slower.
And indexing does not help a bit with Regexes. That’s a full text search operation. I once tried doing that with a bunch of documents in DT (in a script), and it was not useable. But you can easily implement something like it yourself by using a smart rule that executes a script for the RE search.
. An Englishman might say, “maybe you should think about this” whereas others might say “why haven’t you done this?”