Scroll down to see the Zotero Forum.
Do we need access dates for web sources?
Picking up on an off-hand mention of this in a previous thread, I would like to broach this with the group. Zotero's CMOS style is programmed to follow CMOS's newly revised guidelines on access dates. While it once required the date you accessed an internet site, it now says:
An access date—that is, the self-reported date on which an author consulted a source—is of limited value: previous versions will often be unavailable to readers; authors typically consult a source any number of times over the course of days or months; and the accuracy of such dates, once recorded, cannot readily be verified by editors or publishers. Chicago does not therefore require access dates in its published citations of electronic sources unless no date of publication or revision can be determined from the source (see also 14.13). [From CMOS online at https://www.chicagomanualofstyle.org/book/ed17/part3/ch14/psec012.html.%5D
Zotero lets you put an access date on your record, but it is only included in a formatted citation if there is no other date in the record. EE still reflects the older CMOS style. If we are to create a revised Zotero style to handle our work, it would be good to hash this out ahead of time. What are your thoughts about CMOS's assessment of the usefulness of access dates. Do we need them?
----
It would be especially helpful if any of you have known of situations in which the access date made a difference in your ability to analyze a citation or its evidence.
"Do we need access dates for web sources?"
Unequivocally Yes with a few caveats. 🙂
As genealogist's and family historians access dates are not just for readers, editors or publishers. They are beneficial to us as researchers. They can act as a research log of sorts and help in our analysis of the source.
That being said. If I am accessing a previously published book or a traditionally published newspaper online I do not use the second layer in EE where you list the website and access dates. I might make a note that I used an online version but my final citations do not include websites and access dates. Nor do I use website and access dates for FHL microfilm. I use the basic FHL microfilm citation. This is primarily because you can always locate the microfilm through the catalogue and its links.
The other source I am not sure about using websites and access dates is for census records. I am not 100% convinced the added information really matters as long as I have the NARA roll numbers or FHL microfilm numbers. The website and access dates seem redundant to me.
Ancestry is a different animal. They continually change the names of databases and what they contain. Makes me nuts. They have also removed older databases from the website. I have seen entire collections disappear. Because of these reasons, my Ancestry citations do not begin with Ancestry's name of a database. I place the Ancestry information in my second layer. My first layer is written as if I was looking at the original record either at a courthouse, church or state office. The second layer is Ancestry's title with an access date and the third layer is where Ancestry acquired the data. I do have a few exceptions to this. For example the Pennsylvania death certificates. Ancestry scanned the originals in colour. These scanned images are available at the state archives and on Ancestry. My citations for these do not include Ancestry. I use the citation format as if I obtained the certificate from the state.
I find CMOS's statement "will often be unavailable to readers" misleading and not entirely accurate. For example, Wikipedia logs every edit. You can view each and every revision made. Another example is the tree on FamilySearch. Every edit is logged and can be reversed. If I am using a clue I found on the FamilySearch tree I always make a note of when I accessed it and include it in my citation. I am not familiar with wikitree but I believe they track the last date modified.
One other reason to log an access date. If I find a case study, or perhaps information about laws or context on a blog or other website and it disappears if I have the website and an access date I can use the Wayback machine to locate the information. I have already done this in a number of cases.
Excellent feedback, Ann. I also wondered by CMOS hasn't brought up the Wayback Machine. The constant reconfiguration of Ancestry databases is something I hadn't noticed, but certainly does raise a need for documenting where things were at the moment you got them, since the Wayback Machine can't reconstitute the contents of a webpage that sits in front of a database. (And it's a warning to walk away with images of what you extracted--not expecting a URL to always get you back to it.)
Donna,
Here is a classic example of Ancestry changing the name of a database and it happens every couple of years!
This is today's title:
U.S., Social Security Death Index, 1935-2014
Back in 2003, it was:
U.S. Social Security Death Index, 1935-2000
Why they can't just make the title U.S. Social Security Death Index and use an edition number is beyond me.
On 18 December 2011 Ancestry redacted all of the SS numbers for anyone who had died in the last 10 years. Prior to that, you had the information. The irony is State death indexes often include the number without Ancestry redacting it. Knowing when you accessed this database can be extremely helpful. When Ancestry redacted the information it could still be seen on other sites version of the database. I don't know if this is still the case or not.
When they redacted the SS number in 2011 anyone who died after January 2002 their SS number disappeared. If you accessed the data in November 2011 you got different information than you got after 18 December 2011. Since it hasn't been ten years since Ancestry redacted data you could have information that is no longer available and will not be available for several more years or at least until Ancestry purchases a newer edition.
Here is one I accessed 7 December 2011
Albert Leick
SSN 404-38-2102
Last residence: Melbourne, Campbell County, Kentucky, 41059
Born, 12 October 1917,
Died 8 January 2006
State/Year issued: Kentucky before 1951
This is today's entry:
Albert Leick
Last residence: Melbourne, Campbell County, Kentucky, 41059
Born, 12 October 1917,
Died 8 January 2006
State/Year issued: Kentucky before 1951
Great example, Ann. I see what you mean!
Great discussion. For my own notes, I do use a date the record (online or not) was accessed. This helps me later when I'm working through a problem to recall what I knew when.
I agree with Ann's assessment of the need (or lack of) to use the date in a citation for a manuscript found online. I create my citations similarly, as if I used/viewed the actual thing. However for dynamic websites, for the reasons Ann also described, I do use the date accessed in the citation and very often download a pdf of that page for my own use in my notes.
Thanks, Donna. It does appear that genealogists are hitting on a more fluid experience of online databases than is common in the scholarly world. I think it's because we want new material rolled of as soon as it's ready--not once every ten years, when some great glob of new data is included.
For all it's drama, I like the way we do it. Give me the records as soon as they're ready.
I too like to use the date of access, that way if I go back to repeat a search, I know when I did it previously.
Thanks, fhtess65. That could be another tweak to a CMOS variation, if we have general acceptance of the idea. I think some don't want to have it in their citations, but it will not appear if it is not keyed in. So everyone could get what they want.
