So you have found a great resource to bookmark, and you click the bookmarking extension. Within a few moments you see the metadata for your resource on the Edit Bookmark screen, but where has this data come from?
Step 1: What data can we find on the page
The first source we use to find metadata is the page itself. After clicking the bookmarking extension, the URL (Address) of the page is sent to Reading Lists. This is compared to a long list of known websites, each of which has a custom set of rules designed to read that page.
If a match is found then the rules are packaged up and sent back to the page being bookmarked so they can start gathering as much metadata as possible. The metadata will usually be written onto the page itself, sometimes visibly, sometimes hidden. Occasionally the rules may make a call to a separate webpage to get richer data.
Once the rules have completed, and have gathered all of the data they can find, then we are ready for step 2.
Step 2: Can we expand on the data we have found
At this point you will find yourself being redirected to the Reading Lists bookmark editor, but before we reach that point we still have a few more things to do.
The data we have found is a great starting point, but may or may not be trustworthy. The next step is to see if we can improve the data with some more authoritative sources.
The first source of truth we search is your institution’s own catalogue service. If the found metadata contained a Local Control Number (LCN) for your catalogue then we use this to match an exact record in your catalogue. Usually this will be via a z39.50 lookup, but different catalogue may use other lookup mechanisms. If we get an exact match with a record in your catalogue, then the metadata found here will replace the metadata which was read from the bookmarked page.
If no LCN is found, but we instead find an ISBN, then we will use this for the z39.50 lookup against your catalogue. If an exact match is found then, as with the LCN match, we will use the metadata we find in your catalogue in place of the data found on the page.
If we are still without a match for an authoritative record then we still have two more options to try...
If we have an ISBN which failed to match against your catalogue, then we will instead try it against the Open Library. Again, if a match is found then we will use the metadata from here to replace the data found from the page.
Lastly, if we have found a DOI, we can use Crossref to try to find a match. As before, if an exact match is found then we will use the data here in place of the data which was found on the page.
If at this point we still do not have a match then we will proceed with trusting the data which was lifted from the page by the initial bookmarking rules. In any event, the data we have procured, by whichever method, is now passed on to the Bookmarking editor.
Step 3: Bookmark editor
At this point you are transferred to the familiar bookmark editor, which is pre-populated with the metadata we have found.
The initial set of rules which were matched when reading data from the page is named above the editor. Keep a look out for something which looks like “Bookmarked from Amazon”. In this example the rules which were matched were for a page on Amazon. If you worry that the results being returned may not match what you are expecting then this is a good first place to check to see if something is not going to plan.
Step 4: Back to the page
Once you are happy with the metadata which has been discovered for the resource you are bookmarking, click to create the resource. Reading Lists will then set about creating your new bookmark and, optionally, adding it to a list for you. While this is happening it will transfer you back to the page you tried to bookmark to allow you to continue where you left off.
To see a webinar on this also visit: How the magic of bookmarking works?
Frequently Asked Questions
Q. The bookmark editor is describing my resource as a webpage, and only contains a title and a link, why is this?
A. If we are unable to detect any metadata for the resource you are bookmarking, then Reading Lists falls back to bookmarking it as a web page resource, simply linking back to its origin. If the page contains a genuine resource type to be bookmarked, such as a book or article, then it may be that a new set of custom rules need to be created to read that website - please contact Talis Support if you think the website may need adding to our bookmarking parsers.
Q. When bookmarking from my catalogue, sometimes I will get the physical copy of a resource when I am bookmarking the electronic copy, and vice versa.
A. If you are a Primo customer and your catalogue separates physical and electronic copies into different records then you may wish to turn on a new feature called "Bookmarking Disambiguation". This will attempt to detect when a single resource - eg. edition of a book - exists in multiple records in your catalogue, and it will provide an extra screen before the bookmark editor where you will be able to select the specific edition you are seeking.
Q. Does the z39.50 lookup only look at the first ISBN in our catalogue records, or all ISBNs?
The z39.50 lookup will look at all ISBNs in your catalogue record when attempting to find a match.
Q. I've bookmarked an ebook, however the web address isn't coming through to the bookmark, so it's not displaying as an online resource.
The bookmark has identified the ISBN of the ebook and used the z39.50 lookup to search your catalogue. In this lookup it has located a match in your catalogue and used the information from there to populate the bookmark. If the web address isn't in your catalogue record for the book (specifically the 856u field), the web address won't show your the bookmark.
Note that for ALMA Primo customers where there is no 856u in the catalogue record, the bookmarking extension also checks the Uresolver for additional links and adds those into the bookmark.