Family Historian User Group

When I started, I decided to name sources like this below

Marriage 1908 12 03 William McGillivray - Christina Bethune
M Cert 1897 07 15 John Clunie - Anne McDonald
Imm 1917 04 02 Howard G Barrie

This keeps similar sources together, and sorts by date.

However I've also used the same method for naming media files, and I'm concerned that as I get more and more of each, that using the same name for media and source could possibly cause confusion.

I'd be interested in how others name Sources and Media Files.

I take care to give a source and associated media file(s) the same name (Source short title, media file name) may be with a (1 of x) appended for media files if there are multiple images making up the source. It makes it so much easier on those rare occasions when I'm searching for media files via Windows Explorer (I usually do it via fH) and isn't confusing in fH as the two sets of records are distinct.

Maybe I'm not storing up probs then. Interestingly I see you do it in short title. I've never used it. What do you differentiate between full and short

In the Long Title, I put a full citation (as close as I can to Elizabeth Shown Mills guidelines); in the Short title I put a short 'fining aid' like "Burial 1871 Wincanton England, Fred Bloggs".

I use Type as a finding aid as well (to classify the kind of document involved).

I use Text from Source (for a transcription) and Note for anything that needs to be recorded about the Source document (NOT the information in it).

I don't use any of the other fields such as Publication Info, Author -- all that's in the Long Title.

Thanks

ColinMc wrote: ↑

05 Oct 2019 07:51

When I started, I decided to name sources like this below

Marriage 1908 12 03 William McGillivray - Christina Bethune
M Cert 1897 07 15 John Clunie - Anne McDonald
Imm 1917 04 02 Howard G Barrie

This keeps similar sources together, and sorts by date.

However I've also used the same method for naming media files, and I'm concerned that as I get more and more of each, that using the same name for media and source could possibly cause confusion.

I'd be interested in how others name Sources and Media Files.

If you are a "splitter" this might be a suitable approach.

If you're a "lumper" you would tend to name your source with the answer to the question "Where did you source this information?", the information being in this case details of a specific marriage and the sources could be "GRO Marriage Certificates" or "Family Search Parish Records Collection" or "Beaumont Parish Register". Within the citation, a lumper would put "where within the source" information, whilst splitters would not need that separately noted as the split source would have a more specific title including the lumper's citation details.

However even for a splitter, I would ask whether your source title adequately describes what the source is. Which is useful when trying to get back to check a particular source or reviewing what sources you have in support of a specific fact.

Saying your source for information about the marriage of William McGillivray and Christina Bethune is "Marriage 1908 12 03 William McGillivray - Christina Bethune" does not really help (it seems circular). Is the source your recall of personal attendance? - 1908 unlikely! Is it a newspaper article? - If so what paper, date, page/column?

However, "M Cert 1897 07 15 John Clunie - Anne McDonald", works as it makes it clear that it is the Certificate that is the Source of your information. (You could ask is it is necessary to distinguish between certificates from GRO or from the District Registrars - sometimes you can make out different details.)

As far as naming media is concerned, I tend to name media (within FH) with titles that say what they are. So I might have "Smith, S m Jones, P 1953-02-17 Photograph of Marriage Party".

(This is distinct from the file name - which I tend to leave well alone - despite the temptation to make it descriptive. This strategy becomes particularly effective when downloading from the likes of FMP - as you will get a warning that a media "file with that name already exists" - which is a great big prompt that you are dealing with a media record that you already have on your system! Thus you may stumble on discovering who a visitor in a census was or a possible distant cousin marriage!)

I split my data, I took that decision on day 1, having read a bit of background, and it seems to suit me fine.

As I was only giving examples for a different purpose, I did not fully explain the meanings of the naming plan.

In my case, when Source Naming, the words "Birth", "Death" & "Marriage" are reserved for those sources where I have data from an Official source. For me, where the vast majority of my records are Scottish, this means that I have printed the Scotlands People record.

Other cases where I have a Marriage Certificate for example are identified as "M Cert". I then have Emigration, Immigration, Obituary, Press etc. As I do not (and will not) be publishing or exporting my data, the method works for me. It may not be text book, but it does me.

As a result, I have a clearly defined printed list of (my) defined Words for Titles for my Sources. Following a different thread Note Record (17117) yesterday, I have a task set to add that list to my project as a Note Record.

I can easily filter source records to narrow the display down to the type I'm interested in.

I like that idea,

This is distinct from the file name - which I tend to leave well alone - despite the temptation to make it descriptive.

Will think about that

I keep the same name for Source records and any related Media records / Media file names. Unlike some others who have responded here, I consistently start the record name with the persons surname and keep it fairly simple - viz. "Scott, John - 1901 Birth record", "Jones, Peter J to Smith - 1854 Marriage record", "Jones, Paul - 1894 Birth Certificate", etc. I utilise the Source Type to further separate, say, different types of marriage records (eg. GRO record, Marriage Certificate, Parish transcript, etc).

This approach allows me to easily find within both FH and Windows Explorer, all source records / files for, say, "Jones, Peter", etc. It also works well when referencing / displaying source records on my website.

That looks similar to my method but with the order of the naming reversed.

Thanks for the comment

I use a 'splitter' scheme very similar to Brent.
To answer the question of "Where did you source this information?" the details are in the Author, Publication Info and Repository.

Mike;

Could you give an example of your filenames and the corresponding media and source names?

I believe I may have landed on a similar approach.
However; I'd appreciate seeing an example of one that's has been kicked around a bit and is still working well.
The last thing I need is something that doesn't always result in unique names.

Here is what I have at the moment...

For naming media files you could always include an individual or family ID number in the filename.

It may not be immediately obvious who they are from the number but very easy to look up.

Thank you, "lax_gen".
I thought about the ID numbers, but wondered how stable they are.

I use a system similar to laz_gen, but I differ in that I give everyone a unique Custom ID, which I then use to prefix all specific individual media files. ( NB Custom ID is stable and safe to use)
Like J0256ab Christening of Fred Bloggs, or J0256ab service record for Fred Bloggs 1 of 6.
NB I think adding dates to the filename is redundant, as it is already found in the fact.

For more general census media, I use a standard format, such as 1861 2924 F246 P03

Gary_G wrote: ↑

13 Apr 2023 18:23

Thank you, "lax_gen".
I thought about the ID numbers, but wondered how stable they are.

If you dont use the tool to renumber records, they're verybstable - FH never reuses an ID within the same record class (and explicitly discourages you from doing renumbering because it can break e.g. some custom diagrams.)

Thanks, Helen.
If someone exports the data and imports again, are the id numbers retained or does it renumber?

Gary,
Not too sure, that’s why I stick with my Custom_ID. I don’t trust it not changing

I was thinking of trying to use something like I indicated in one of my posts, which might allow me to relate my source names to my image names.

eg. Démoulin (Ulysse): 1884-10-17 Birth [Archives départementales de l'Oise: 3E259/2: 64-65]

That is;
The items in square brackets are; the repository: Call_no: image(s).

This will usually ensure that the text outside the square brackets is always unique. However; when doing some British research a few years ago, I came across 5 siblings baptized at the same place and date. Consequently; they were in the same image. I guess that the only way to make it ABSOLUTELY unique is to add the entry number.

This is getting far too complicated. There must be an easier way.

Gary_G wrote: ↑

13 Apr 2023 19:09

If someone exports the data and imports again, are the id numbers retained or does it renumber?

When you import a GEDCOM, it will import the ID numbers in that file. If it is the same file you have exported, they will remain the same. However, if another person or a software application, has decided to change anything in the file while it is out of your control, clearly FH will import the changed ID.

I have never had a problem with the numbers changing in the 20+ years I have been using genealogy software. The big no-no, as Helen points out, is manually resequencing the IDs.

Gowermick has explained his ID system elsewhere and it is quite ingenious, however maintaining unique IDs is too much for me, I would be guaranteed to mess it up at some point, and then I would need to change them ... back to square one!

The only software I have seen that does not import and preserve the Ids from GEDCOM is Heredis.

Thanks, everyone
It sounds like this is still something I need to think about some more before committing.

IMO exporting your data and importing it again is red herring.
Whatever product you export to can potentially upset Record Id, Source Titles, Media Names, Custom Id, etc.
You cannot rely on anything being retained perfectly unchanged unless you carefully vet the other product beforehand.
Why would you do it anyway?
IMO genealogy should not be migrated backwards and forwards amongst products.
Stick to one product to hold your master database and only transfer it out to other products, but never back again.

FH Record Id are never changed unless you deliberately use certain tools to alter them.
See FHUG KB Record Identifiers.

Gary, there is little point in me posting my record naming conventions because they only relate to generic Sources.
IMO if a few Source records or Media records have the same Title it is not a deal breaker.
They are linked together back to the particular person and event involved and so are easily differentiated from others with the same Title that are linked to a different person and event that happen to have the same Name and Date.

Mike;

I've seen Heredis mess up ID numbers when I send a friend a GEDCOM.
Thankfully, we both have stayed clear of that program ever since.

I think I might just use the person ID after the name. Does the image import also assign a unique ID? It is possible for me to have two records of the same event in my image files and I need to differentiate those. Actually; this is starting to look very much like what I do under my current naming convention.

Please read the KB article I quoted as it explains the answer to your question, which is YES.