* Citing the Times Digital Archive

[DavidNewton]

I have just gained access to "The Times Digital Archive, 1785-1985" via my membership of Cornwall County Library and I suspect that eventually I will have a number of citations. So before I get started I would welcome suggestions as to how make the citations.
My initial thought was to use a Method 2 source with the title above, particularly as I am not going to be taking screenshots of copyrighted material. But that is not set in stone.
{Added in Edit: It turns out that the documents can be downloaded in pdf format so perhaps Method 1 is the way to go}

When you look at a document, it may be a piece of a page or more than one page, you get a header describing the document

Universal Register..
The Times (London, England), Wednesday, Feb 09, 1785; pg. 2; Issue 34. (839 words)
Category: Official Appointments and Notices

and at the bottom of the document a suggested source citation

Source Citation:
"Universal Register." Times [London, England] 9 Feb. 1785: 2. The Times Digital Archive. Web. 9 Mar. 2018.
URL
http://find.galegroup.com/ttda/...
Gale Document Number: CS33685577

Obviously I would want to try to keep all this information in some way or another.

David

[tatewise]

Until you have decided between Method 1 and Method 2 it is difficult to give precise advice.
But it sounds likely that Method 1 would be better for holding the PDF &/or its JPEG converted image, plus a transcript in Text From Source.

The Repository could be The Times Digital Archive, 1785-1985 with any contact address, email, web address, etc.

I would suggest that something like this will work:
Title = The Times (London, England), Wednesday, Feb 09, 1785; pg. 2; Issue 34.
Author = © Times Newspapers Limited
Publication Info: Gale Document Number: CS33685577 http://find.galegroup.com/ttda/infomark ... ersion=1.0

But depending on the style of other Sources you may want to put the date first in the title.

[LornaCraig]

I would use Method 1. You will probably accumulate a lot of newspaper sources eventually, many of which will not come from the Times Digital Archive but from elswhere. I have several from the British Newspaper Archive on Findmypast, some from an Australian newspaper archive and some which are images of newspaper cuttings uploaded by Ancestry members and attached to their trees. In each case I attach the PDF or JPG to the source, even if I don't do a transcript of the text.

For the Title I choose something which indicates the nature of the article, such as 'John Smith: obituary'.
I enter the name of the original newspaper and publication date in the Publication Information field and record the website from which I obtained the image in the Note field.

[DavidNewton]

Thanks Mike and Lorna. I was leaning towards Method 1 when I discovered that I could download the pages. I have a favoured format for short title so I can locate sources for an individual but the title can be much more standard descriptive of the source.

Setting up a respository record here makes sense since the Database Document Reference is supplied. I assume this goes into the repository ID field, possibly together with a media type.

David.

[tatewise]

Yes, I was on the verge of mentioning the Repository ~ Identification and Media Type fields, but thought it might cloud the main points, until you had plumped for Method 1.

The only minor drawback is that Reports use the term Call Number: for the Identification field because I think it is related to LDS Family Search resources. Anyway, Call Number (CALN) is the GEDCOM definition of that field, so I don't know why FH calls its Identification in Sources. Alternatively, use the Repository ~ Note then it can say whatever you like, such as Gale Document Number: CS33685577.

BTW:
If you plan to use those fields often, they can be customised onto the Main tab of the Property Box to avoid right-clicking around on the All tab.

[ColeValleyGirl]

Call Number is a pretty widespread term used as to refer to an archive/repository's reference number.

[DavidNewton]

Thanks both. I think I'm getting there. I already have a Repository link on the Main tab of the Source Property Box and have now added a line for the Call Number and the Media Type (although frankly the choices for that are very restrictive). The source below is a sample of a Death notice. I would use my usual short title identifier and then

Title: The Times Digital Archive, 1785-1985.
Publication Information: Thursday, Apr 03, 1941; pg. 1; Issue 48891. Deaths
Repository: Gale Group Databases accessed via Cornwall Library.
Call Number: CS17250435
Media Type: Electronic
Text From Source: Web. accessed 9 Mar. 2018.
URL: http://find.galegroup.com/ttda/...
Note: Harley Walker. On March 31, 1941, the Reverend Cyril Tollemache Harley Walker, of East Hanney, Berkshire, aged 57

Just for the record I never expose Text From Source in reports, because I really do not like the Text from Source label, and so I put the transcript into the Note field.

One more question. Is there any way to extend the list of repository media types?

David

[tatewise]

No, that is why I suggested the Repository ~ Note where you can say what you like.
Rather like you prefer the Source ~ Note over Text From Source.

Like Lorna, I would probably finish up putting the person's name and date of death in the Title using a similar format to a Death Certificate but making it clear it is not.
The Times Digital Archive, 1785-1985 I'd relegate to the Author, Publication Info or Repository Title.

[AdrianBruce]

For what it's worth (and we all seem to do these things differently), because it's a scan, I effectively start by describing the source as if it were the original, and only swerve off at the last minute to include details of the image of the source in the publication data. In effect, it's still The Times, it's just published differently.

For instance:

Title: "Arrival of the RMS Iberia", The Argus (Melbourne, Victoria), Thursday 18 September 1884, page 6

Short Title: ... (my own format) ...

Type: _Newspaper; scan

Publication Information: Digital image of original published in Digitised newspapers, TROVE (index and images), accessed 26 February 2016

Actual Text: <See attached file>

Repository: Trove web-site (National Library of Australia)

This assumes that you don't really need to use the publication details at all for the original newspaper. If you do (as you would for a book) then the publication info is a composite of both.

[DavidNewton]

A lot of good ideas here and as Adrian says we each have our own way of doing things but it is instructive to see how others do things.

I have tried using the Repository Note and agree it gets rid of the Call Number label, which is good, but in a report the Note starts a new line after the Repository name which is not so good. It did make me look again at the final report format and I realised that the Media Type could be combined with the Call Number, CS17250435 (Scanned Image), and achieve the same report result as having them separately and opening an opportunity to describe the media type in your own words.

I like the idea of including the Accessed date, if available, in the Publication Information but this is one of those things which really have to be recorded at the time.

I am still uncertain about including the content details in the title although with Census sources my title includes both the address and the principal householder (who may or may not be the Head of Household). My method 1 short titles conform to a pattern which keeps together the method 1 sources associated with a given individual and in date order but are skimpy on the actual source material.

I tend to be indecisive, which means lots of editing if I change my mind, so just once I would like to get a format that I like on the first run.

David

[DavidNewton]

Still dithering and trying to put together the suggestions above! I have no difficulty including all the information that I have within a source record and so basically all that matters is how is looks as a citation in a report. So I examined that a bit more closely. My latest effort is to ditch the repository and call number/notes and put that information into the publication details. I expanded the author field a bit and put the source date, page number, Issue number and category into the title. The end result gives me a citation as follows:

The Times Digital Archive 1785-1985, © Times Newspapers Limited, "The Times, Thursday, Apr 3, 1941, p 1, Issue 48891. Deaths" (Gale Group Database via Cornwall Library Online, Document Number CS17250435, Downloaded 9 Mar 2018.).
Harley Walker. On March 31, 1941, the Reverend Cyril Tollemache Harley Walker, of East Hanney, Berkshire, aged 57.

The only thing missing from this is the URL and I am happy for this to be hidden in the Text From Source field. I'm still not sure about the author information. Should the copyright notice come first?

David

[AdrianBruce]

Some thoughts...
1. Despite it being in my example, I'm not convinced I need the download date. It's definitely needed for a transcript since the content of the transcript could be "corrected" between downloads, but the content of a digital image isn't going to change (though it might get sharper, etc.)

2. The URL is probably not necessary. Unless Gale Group commit to permanent URLs (e.g. ARKs?) then the URL of the page will change sooner or later. The URL of the site as a whole might be more useful but even then I'm not sure if it is any use as I won't access the GALE Times via your library but through mine....

3. Frankly, I wouldn't bother with the copyright notice. It's still copyright (or not) whether you add that bit or not. Feel free to show it with the image if you print it.

4. Depending on what you do with the copyright notice, you probably have one too many "The Times" in there. I was impressed when reading Elizabeth Shown Mills, the guru of (basically) American citations and source analysis, when she said that a citation needn't repeat the same thing - how many references to "Ancestry" do you need in a citation, she asked, tongue in cheek, before someone realises that the source comes from Ancestry?

[ColeValleyGirl]

Re download/accessed date, I always include it because it's not unknown for sites to remove data sets or for whole sites to disappear, so it can be useful in reassuring me I'm not going mad

If I've accessed the site via the Wayback machine, I make that clear as well. And I note the form in which I consulted the source (image, transcript etc.)

I do include the website name and top level URL for the site but not the specific item (unless the site has committed to permanent URLs). And I don't include details of intermediate sites.

I don't include copyright notices -- why that and not (for example) the terms of service of the providing website? As Adrian says, it's better attached to any image that you might republish than to the source.

For newspapers I include the publication place as well to disambiguate papers with the same title and I give a column as well as page number and sufficient identification of the text in question to make the relevant article/whatever easier for people to locate. If the paper produced multiple issues on the same day, I'd include the issue information as well.

So my citation for that source would be:

London, England. "The Times" 3 April 1941 (digital images), <website name> (<url goes here> accessed: 9 Mar 2018). Page 1 Column <column number goes here> headed Deaths, entry for the Reverend Cyril Tollemache Harley Walker, of East Hanney, Berkshire, aged 57, died 31 March 1941.

I can't swear my punctuation would satisfy the most punctilious of the citation police, and it certainly isn't fully compliant with ESM, but I reckon it answers the questions: what did I consult and where and when, and how can others consult it? It doesn't presuppose that others will follow exactly the same route to the end point as I did, but does allow them to assess how likely it was that what I consulted is worth tracking down.

I'll also add I put all of it in the Title and don't use any of the other fields -- I never managed to devise a way of using them that catered for the breadth of the source types available.

[DavidNewton]

Thanks for the comments Adrian and Helen.

I was interested to see that Helen writes her entire citation within the title. It certainly allows you to do it your own way. However, for me I can see a problem as I sometimes produce reports with reduced citation references by excluding the content (Text From Source and/or Note)
I'm beginning to think that copyright notices are possibly contentious. I always include them if I find them just as an acknowledgement to the copyright holder. The terms of service of a website publishing copyrighted material seems less relevant and does anybody actually read them?
I agree with Adrian that Times probably occurs too often, but within my context of copyright holder, database name and newspaper reference I can't see how to get rid of one.
It has certainly given me things to think about, as I hoped it would.

David

[ColeValleyGirl]

David,

The ToS of websites can be very relevant -- for example, they can publish material which is Crown Copyright (e.g. Census Images) but restrict the use you can make of it... Copyright rules may say you can publish it but ToS prevent publication. I acknowledge that many people don't read ToS but some sites can be very strict about enforcing them...

Not sure about

a problem as I sometimes produce reports with reduced citation references by excluding the content (Text From Source and/or Note)

I do the same -- the report just shows the full citation -- i.e. enough info for others to locate the same source -- and exclude the text from source (which I use for transcriptions) and notes (which I use to explain any discrepancies I've identified between that source and others -- for example "This MI says Ann died on 21st December 1882, but the burial register says she was buried on 20th December 1882. I believe the burial register to be more accurate, as the MI will have been created at some time after the death/burial." -- or to otherwise highlight problems with the source or the reason I've interpreted the information in it as I have.

[Mark1834]

Interesting food for thought here. My approach as been to consider two separate but related questions - what detail do I need for myself or another researcher to check the information quoted, and what is simply convenient to have in some circumstances.

For example, for things like census returns, parish registers, GRO indexes, probate calendar etc all I think you need are the image file and the core reference. Given the PRO citation, anybody can look up the census in any one of many repositories. Given the name of the parish, date and type of event, again anybody can check the information. In both cases, knowing exactly when I downloaded it and whether it came from Ancestry, FMP, or somewhere else is usually just extra detail and not relevant to the veracity of the record. It may help to find it again if anything happens to my stored image file, but that is simply convenience. Any doubts about the data would be recorded as a note.

Similarly, GRO Indexes are now readily available, so probably all you need to record is the degree of validation of the transcript. "Strong" validation would be checked personally against the original image, or validated FreeBMD (bold, double keyed), etc, while "weak" validation would be simply accepting an unchecked transcript such as non-validated FreeBMD. As long as you have that level of detail, anything else is superfluous and forms more of a research diary rather than strict citations.

Where you are relying on transcribed material that cannot readily be checked against the original (e.g. FreeReg, FamilySearch, etc), I agree that it is good practice to note when the record was accessed, as they can and do change with time.

For your original example, I'm sympathetic to Adrian's approach - it is basically just the Times, so as long as you have the date of original publication and a bit more information to find it within the paper, that is all you actually need. Website addresses, copyright notices, etc are optional detail.

It therefore follows that the more obscure and specialised a source becomes, the more detail you need to record to locate it again.

This is related to how I treat image files. I always just link, and never copy. My image of the 1881 Census is no more a FH record than it was an Ancestry or FMP record. Other family history programmes are available, and I want my legacy (with a small l) FTM database to still work, and I want to be free to play with other programs such as Gramps without copying gigabytes of files or referring to files within FH folders.

Clearly, we all have different needs, but I suggest that what you actually record is driven more by how you want to present the data (just in the FH interface, printed reports, etc), rather than what is strictly needed for good research practice. After all, when writing a research report in my old day job, I would cite the various documents referred to, but not when I read them and where I obtained my copy from.

[tatewise]

Mark, I understand your rationale for linking FH Media images externally rather copying into the Project internal Media folder. However, just bear in mind that it complicates migrating FH from PC to PC, because the linked Media File Paths are unlikely to be the same in the new PC, resulting in broken file links that will need to be repaired.

You can mitigate against that problem, either by using a standard PC folder that exists on all PC such as C:\Users\Public, or at least by keeping all your Media within one root folder or its sub-folders, then on the new PC the Tools > External File Links buttons should easily repair them.

[DavidNewton]

Mark

After all, when writing a research report in my old day job, I would cite the various documents referred to, but not when I read them and where I obtained my copy from.

I agree, in the days before the internet this was generally enough and was certainly my own practice. But now with many sources online, and some jpurnals exclusively online, different methods have been adopted. For example the Harvard Citing and Reference Guide suggests for electronic sources that the reference conclude with Available from: URL [Date of access] . I am not suggesting that this is what we should all do. I think it is probably a bit of overkill for genealogy reports and, as we know, internet sources have a habit of disappearing or being relocated which suggests that such attention to detail is irrelevant and likely to be misleading. In the end it is a personal choice.

With regard to images, again a personal choice, but I copy because I find it easier to rename the files to my format choice in the copying process and I often copy the same file more than once with different filenames. Obviously this can easily be done in file explorer. The one thing that bothered me a bit was that if I discovered later that I had the wrong image then deleting the media record did not delete the image file (solved with a simple plug-in). Although I have not yet moved FH to a different computer I take Mike's point regarding using the standard Media folder.

David

[Mark1834]

Indeed - this forum provides plenty of evidence that copied files are the easier option for users who neither know nor care what is happening “under the hood”, but if you are familiar with folder/directory structures and working at the command line it is easily managed. I routinely use FH on two different systems, but they are set up with the same folder and drive names so everything is transparent. It was a simple one-off change in FTM when I upgraded from W7 to W10 and started using OneDrive a couple of years ago, and similar in FH should it change again in the future.

[Mark1834]

Agree - that’s why I differentiated between generic sources such as the census that are available in many locations and more specialist material that may only be available on one site. In that latter case you definitely need more details about where it comes from.

For images, my practice is to rename as I download, so it is always immediately clear what the file is (e.g. “Census 1881, Parish Name, PRO reference.jpg”. It is slightly annoying if I have to adjust the name to accommodate limitations of the filing system (e.g. “...RG12_1234...” instead of “...RG12/1234...”) but I have systematic methods for that. I organise folders hierarchically, so census, baptisms, death certificates, military records etc all have their own logical place in the structure.