How to edit information in Family Tree and in Indexes
Answers
-
Sorry, I've harped enough about the need for precision, clarity and preciseness when asking questions or discussing any topic here because the FamilySearch website is so complex, we don't have visual clues about what we are all talking about until we can post images again, we come from so many different places and we all have different levels of experience. This case of me being sloppy slipped past me.
@Julia Szent-Györgyi 's discussion made me realize at I need an agreed upon single definition term or short phrase for the individual partial transcriptions of documents that as a collective set make up an index to the corresponding set of documents. I don't want to need to write "so you want to edit an individual partial transcriptions of a document that is part of a collective set which makes up an index to the corresponding set of documents?" every time a question like this comes up.
"Editing an index" would refer to editing the entire index. "Editing an indexed record" is not correct if we are going to reserve the term "record" for the document itself because the indexed record is the document, not the transcription. "Edit a record's partial transcription" it too long and clumsy still.
Best I can come up with are:
- Index item
- Index entry
- Item in an index
- Entry in an index
which are all pretty synonymous. Any other proposals?
This work of creating a new language (FamilySearchese? FamilySearchian?) in which each word or phrase, such as "record" has only one meaning is difficult. In normal English to speak of a person's Family Tree record, a record in an index, and a record in an image set would all be perfectly fine.
2 -
'record metadata'?
0 -
@Gordon Collett I agree that your four suggestions are essentially synonyms. My favorite is "index entry"; it's simple and accurate, and refers to a particular entry in the index (which is what is being asked for here, if I understand correctly).
I agree with you that "Editing an index" is too broad and makes it sound like it affects more than just a single entry. "Record metadata" suffers from that excessive broadness as well. So I like "edit an index entry" better than anything else I've heard so far.
3 -
I agree with Alan that "index entry" works best. Of course, this being human language in general and English in particular, there's still room for ambiguity, because when the index includes the bride, groom, bride's parents, and groom's parents, is that six index entries or just one? (My answer is generally "six", but with the knowledge that it's always going to be a bit fuzzy.)
1 -
Dear System Admin,
Since this discussion doesn't make great sense without the context of the first couple of posts and since it is not talking about indexing at all, could you please put it back where it came from, that is Ask A Question - Family Tree?
The ongoing discussion is about editing Family Tree profiles, editing index entries (which is different than indexing records), and the inability to edit document images. If you feel it really needs a new title, then a more appropriate one would be "How to edit information in Family Tree and in Indexes."
Thank you.
4 -
Regarding the one entry vs six entry question, here is how I picture the overall process. I'm probably wrong with the fine details and the underlying process.
When indexing, the indexer effectively creates a single line in a spreadsheet for each document in a volume. At this point, there is a one to one correspondence between this original row (or column if you like since that is how the data looks while indexing) of data and the document. Each document creates one entry in the spreadsheet.
After a volume in completely indexed, post processing duplicates each single row of data into multiple rows of data in a different spreadsheet in which there is now a row for each name in the original spreadsheet. Each of these rows is then assigned a unique URL to differentiate each row. At the same time the duplicate rows of data are in interlinked to show their original relationship to each other and that there were in the same row in the original indexing. In this spreadsheet there is now a one to one correspondence between each row of data and each name.
In summary:
Indexing creates one data set entry for each document.
Post-processing creates one data set entry for each name.
1 -
It's fuzzier than that, Gordon, because "document" is itself ambiguous. If it's a baptismal register with 76 baptisms on a two-page spread (https://www.familysearch.org/ark:/61903/3:1:9Q97-Y39V-CXW?i=5), how many documents is that? One for each event in the image? 40 for the verso side and 36 for the recto? One for the verso and one for the recto? Just one for the whole image?
At the other extreme, what if the image covers only a part of a vital event, such as when marriages are recorded on recto and verso, but of course the two-page spread on the image is the second half of one marriage entry and the first half of the next? How do you define "document" then? Or what if it's one of those crazy early microfilms where the versos are on a different film than the rectos?
The thing that all of these types of originating document have in common in the indexing process is the concept of the index entry. The example image I linked above has 76 index entries associated with it. (Presumably, if they did it right. I didn't actually count.) A verso-and-recto marriage register image has one index entry associated with it. (By convention, it's the one that begins on that image.) An image from the middle of a naturalization file may not have any associated index entries, despite the fully-indexed status of the file.
Each index entry may include the names of multiple people, along with their relationships/roles in the event, and the system assigns a unique identifier to each of those people. Those unique identifiers each generate an index detail page and a search result line, and if the index entry is attached to a profile, then the unique identifier and the data it points to is used to generate the resulting source. (Which is another overused and fuzzy word.)
It's tempting, then, to consider one line on the old viewer's index tab to be one indexing unit, by whatever name — until you consider censuses, which put each new name on a new line, and group those lines into households using the data in other fields. (Or not, depending on the depradations of the various gremlins in the system.) The "one entry or six (or ten)" question becomes extremely fuzzy when you're trying to talk about both censuses and church registers.
1 -
@Gordon Collett With regard to your first terminology question, I have used "index entry" and find that to be a good choice. However, when you expand with your description of the indexing process and its results, I have to agree with @Julia Szent-Györgyi that things can quickly become muddled or "fuzzy", as she says. "Document" does not seem appropriate because of scope and ambiguity issues. What is usually being indexed is an "event entry", whether it be in its own "document", or a row, or a block of text in a register of some sort. You point out that the indexing of the "event entry" typically gets "expanded" into index entries for each person connected to the event. Are those then "event participant index entries"? And even that is not applicable to all cases….what about the relatives "referred to" who aren't actually participants or even present at the event? Do their entries become "event personal reference index entries"? We will probably lose some people along the way as to just what that means, even though we are trying to be exactly clear in our terminology. Even the term "event" is not always accurately descriptive of what is being indexed. I would note that we are not sure how this indexing data is recorded "behind the scenes". The event has data (typically type, data, place) that is "shared" among each of the individual "event entries. Do edits of this data get saved to the "event", the "individual", both? Do censuses had another level, "household" to this index entry hierarchy? It seems very difficult to come up with terminology that is both precise and practical….
1 -
I never realized that the Heisenberg Uncertainty Principle applied to language. The above are all important and intriguing comments and point out that being precise and pedantic with terminology causes its own problems for effective communication and that when trying to help people with Family Search or with anything in life, it is often most effective to let people relate a problem in their own words and then ask clarifying questions. Starting with "Tell me what you mean" or, even better, "Show me what you mean."
3 -
and sometimes when you" show me what you mean" sharing the project title with the batch code is more helpful than a screenshot 😎
0 -
But, @maryellenstevensbarnes1 we are not ALWAYS talking about a specific batch or even about an indexing project.
0 -
@Áine Ní Donnghaile True and thank you. I always read all the comments on all the topics and questions because I learn so much that way. I admire your talent and dedication. Thank you for teaching and helping all of us. Have a blessed day. Mary 😎
0
