What is the actual goal of citing the "access date/time" of when you access a URL? For example:

Bruce Bower (July 17, 2010). "Serbian site may have hosted first copper makers". ScienceNews. Retrieved 22 April 2017.

Or sometimes thanks to web.archive.org, you get some that archive the URL:

Allen, Michael J.; et al., eds. (2012). Is There a British Chalcolithic?: People, Place and Polity in the later Third Millennium (summary). Oxbow. ISBN 9781842174968. Archived from the original on 2016-10-05. Retrieved 2016-02-02.

If URLs can change willy-nilly, what is the point of referencing the access date? The URL could become dead at which point the source/citation seems useless, or the content could be changed and the original content at the retrieval date can no longer be found. So I don't understand why this practice is in place, and yet I don't have a better alternative.

  • 7
    “Half” is inaccurate. All the references will be dead sooner or later. We are in fact in the classical/mediaeval environment where you refer to a previous writer and the only MS of the cited work disappears. Many pre-Socratic philosophers are known only from quotations in later works. If a printed source exists, always cite it - in addition to an online version, if you like. Much of the purpose of the date is to help the reader decide whether there is even any point in chasing up the reference you have given. Commented Sep 5, 2022 at 7:11
  • 4
    Well, in a way this "timestamping" of citations is indeed irrelevant - it simply doesn't matter, as your paper is timestamped anyway. It is expected the source is valid at the time of submission and during review process - if it changed before submission or during review, citation is worthless and should be removed or replaced. ((of course, if you yourself have a type of media that can change eg a webpage, then it does make sense to timestamp accesses to other transient stuff)) Commented Sep 5, 2022 at 7:31
  • 9
    I don't really get your question. Yes, stuff pointed to by URLs is incredibly unstable, even if the link itself stays up. We know that. But you already know about a tool to combat that: the waybackmachiene. With a fixed date and an url, you can - with some luck - get the exact version of that website that was cited right there. Whats your question?
    – Hobbamok
    Commented Sep 5, 2022 at 10:53
  • 4
    You go to web.archive.org/save and let the Archive save the page, then use that as the timestamp that you cite. While there's a low chance that it might still get removed etc., others will then have a good chance of getting pretty much exactly the version that you looked at.
    – nobody
    Commented Sep 6, 2022 at 12:16
  • For the record, good URIs never change. Commented Sep 12, 2022 at 11:46

3 Answers 3


The point of referencing the access date is because URLs can change willy-nilly. It's true that it might be hard for a reader to access the specific version of the website you're citing if the website has since been updated or deleted. But at least it gives the reader some idea of what's going on if, for example, the reader finds that the information you cited has been changed or is missing.

Incidentally, this is why you should try to avoid citing websites which are liable to change in the first place, if you can avoid it. Instead, aim for sources with greater permanence. Ideally published, but at least archived with a permanent identifier, DOI, etc. If you really can't avoid citing a website, and the website later dies without any kind of archived version, then that's unfortunate but there's not much you can do.

Note that in your second example, you're actually citing a published book. In that case, you just need to cite the book itself, not the particular website you might have used to access it or the date you did so (the book is already in a fixed, published form, it's not going to change based on the date you access it). As long as a reader has sufficient information to find the same version of the book (including e.g. the particular edition number you used, if applicable) then that's sufficient.

  • 28
    'there's not much you can do' Except there is: anyone can save the current iteration of a web page in the Wayback Machine. Commented Sep 4, 2022 at 15:10
  • 34
    Assuming the Wayback Machine never goes down or runs out of funding!
    – Lance
    Commented Sep 4, 2022 at 18:12
  • 8
    the idea that this is much you can do, does not assume that
    – amara
    Commented Sep 4, 2022 at 22:15
  • 5
    @DanielHatton There's been instances of the Wayback Machine removing archived pages at the request of the site owner or some other authority.
    – walen
    Commented Sep 5, 2022 at 9:35
  • 2
    @user253751 source?
    – walen
    Commented Sep 5, 2022 at 15:45

URLs change or stop working, and the contents of webpages also change.

The date enables readers to view the exact version of the page that you viewed, even if the page was subsequently changed and even if the URL no longer works - at least, it enables them to do this if the page is archived on archive.org.


You cite with a date for your own veracity. At the time the citation is made it is (almost always) checkable. The page may change or disappear, as you note, but it existed in the form you claim it did on the date you claim - or so you claim.

Without the date, nothing is really checkable. So, you are making a claim of honesty about what you say and giving proper attribution.

And, we do the best we can with the tools we have. The internet archive provides an improvement over a situation where sources can change without comment or ability to verify. Not a perfect world, but it is the one we live in.

Note that when you cite a book, you cite the specific edition. And when an archive captures various versions of a web page the date points (hopefully) to the version you accessed.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .