How to: Deep Linking in AquaBrowser

Yesterday, I tweeted a quick post asking for assistance in creating either hrefs or URLS that could link into AquaBrowser. Specifically, link into canned searches, such as author searches, a specific subject or a combination of the two.

This is an issue that has been uppermost in my mind because of the popularity of reading lists and canned searches which we all use when linking from our websites or library blogs into the catalogue. For example, if we want to create a genealogy splash page, we want to link into key resources or pre-determined searches in our catalogues – if we can’t do this with our new catalogues, it’s a failure. One of the purposes of social catalogues, such as AquaBrowser, is to enhance user access and make the collection more visible by asking users to share and collaborate – and to make linking into the catalogue even easier! Fortunately, thanks to two colleagues of mine (one from my own library, and one from the New Brunswick Public Library Services), I’ve discovered this is possible. Not as easy as, perhaps, it should be, but we can continue creating our reading lists, resource lists and canned searches. Hopefully the “ease” in creating these links will develop as part of the vendor’s “intuitive” vision for these catalogues. From my own experience, the hands-on AquaBrowser crew are keen on new ideas and addressing shortcomings – so I think we’ll see some enhancements to deep linking in AquaBrowser.

While I’m still working out how to create complex canned searches that include Boolean – basic subject, author, ISBN and simple Boolean (with an implied AND) do work.

So, for those of you who have been trying to figure this out – I’m going to provide some examples straight out of our catalogue. Also, a nod to my colleague in New Brunswick who provided a really neat URL resource for linking into other libraries’ catalogues.

Examples for Deep-Linking in AquaBrowser
http://discover.halifaxpubliclibraries.ca/?q=caterpillars

http://discover.halifaxpubliclibraries.ca/?q=title:caterpillars

http://discover.halifaxpubliclibraries.ca/?q=author:lackey

http://discover.halifaxpubliclibraries.ca/?q=subject:cooking

http://discover.halifaxpubliclibraries.ca/?q=isbn:0671709674

If you use a MARC tag, such as the 449 tag, to create linkable reading lists, you can recall those lists as well. For example, we have a Teen Picks List at HPL. This is how you’d link to it:

http://discover.halifaxpubliclibraries.ca/?q=librarylist:Teen Picks

And finally, just some helpful tips when creating links into AquaBrowser:
• – ‘q’ initiates the Aquabrowser with a query. This can be used to build a search box on another site or page, and deep link into AquaBrowser with a query.
• – ‘uilang=en’ sets the user interface language ‘ENglish’. Also supported are ‘nl’ ‘fr’ etc. More languages can be added in language.xml.
• – ‘itemid=default::31234′ (2.0 only) gives a direct deep link to the item in the table ‘default’ and id=’31234′.
• – ‘itemid=|library/marc/howardcounty|453410′ (2.2 and up) gives a direct deep link to the item with extID=’|library/marc/howardcounty|453410′ (extID can be found in record when ?debug=true)
• ‘branch=branchcoe’ allows you to preselect locations (branches). Note that you need to branch CODE, which may be different than the name displayed. To see the branch codes use debug=true and look at the xml generated for the branch selector (under the feedback module).

This is new to me as well – so please play around with it. As you can see, it’s not as intuitive as it can be – and will definitely take some time to create. In the past with our classic catalogues, all we had to do was use the URL provided to us by our search. However, in AquaBrowser, the URL never changes or “refreshes”, so we can’t see the URL. Unfortunately, patrons still won’t be able to share more than a single item URL with friends. Hopefully, in time, we’ll see if this changes. But, in the meantime, libraries can continue to link to pre-determined searches and reading lists in AquaBrowser.

And, just for fun, my New Brunswick colleague sent me the coolest link. Here’s what she had to say:

This is my favourite type of link to do in a social catalogue: http://discover.halifaxpubliclibraries.ca/?q=md_tags:awesome

Everything user-tagged “Awesome”

Advertisements

6 Comments

Filed under Access Issues, Social catalogue, The Library Catalogue

6 responses to “How to: Deep Linking in AquaBrowser

  1. Laurel Tarulli

    I’ve been working on that today because I’m working on a set of Mi’Kmaq queries for a proposed new page on our website.

    One of the searches using Boolean would be: Mi’Kmaq AND art OR basket OR folklore OR legends OR literary OR fiction OR culture OR music OR societies OR poetry OR rites OR customs.

    After trying several attempts, I came up with this:
    http://discover.halifaxpubliclibraries.ca/?q=subject:mi%27kmaq%20art%20OR%20mi%27kmaq%20basket%20OR%20mi%27kmaq%20folklore%20OR%20mi%27kmaq%20legends%20OR%20mi%27kmaq%20literary%20OR%20mi%27kmaq%20fiction%20OR%20mi%27kmaq%20culture%20OR%20mi%27kmaq%20music%20OR%20mi%27kmaq%20societies%20OR%20mi%27kmaq%20poetry%20OR%20mi%27kmaq%20rites%20OR%20mi%27kmaq%20customs

    Extremely nasty looking, but what I found out is that for every “OR” term, Mi’kmaq must be next to it. So with Boolean, it looks more like this:
    Mi’kmaq art OR Mi’kmaq basket OR Mi’kmaq folkore, etc.

    That way, the URL reads Mi’kmaq and all the variable terms you are also looking for.

    In the end, it looks like AND is implied as long as the two terms are next to each other and the OR operator does work in the URL. Perhaps, however, I’ll be able to find an easier way as I keep working on it!

  2. hmmm. it would be nothing to code in a case-check for OR, AND and NOT.

    I did a parenthetical search within aquabrowser that produced one more result than the long url.
    subject:mi’kmaq AND (art OR basket OR folklore OR legends OR literary OR fiction OR culture OR music OR societies OR poetry OR rites OR customs) and got 120 results. The parentheses do not need to be escaped in a URL (ie. no %–) required.

    It’s good to remember that Boolean is really a system of “TRUE” and “FALSE” checks.

    AND – checks the first term (exists(x) == True) and if it returns true, it checks (exists(y) == True) – that means it’s alot faster than all other searches because if x doesn’t check out, then it can skip on to the next item. This is probably the main reason why big systems assume AND (although MYSQL’s default text search defaults to OR for some reason).

    OR – checks the first term, then checks the second term, even if (exists (x) == False). That means it is more intensive in terms of resources.

    NOT – is the worst because the system only skips if y is true. NOT mi’kmaq would produce a tonne of results, obviously. It really only should be used in tandem with an AND search. One NOT search is fine, but having a really popular set of links out there with a NOT search could have people screaming about how slow everything is running after a while.

  3. Laurel Tarulli

    Hey Ryan – this will be helpful to quite a few people – thanks! By the way, Jeremy was a great help, too!

    I think all librarians have (or should have) a good understanding of BOOLEAN. That’s not the problem – understanding URLs (which is your area of expertise, not mine) is what helps in this case.

    On the homefront – you and I may need to have a meeting about this. We need to create a simple solution/workflow between my dept. and yours so that we can manage reading lists and canned searches as they’re created throughout the library.

    Thanks again Ryan!

  4. Laurel, I think you missed the nuances of what I said about Boolean. Knowing how a Boolean search is constructed by a computer has serious implications on the resources that the search actually uses. This may not matter in the case of a single search, but if a search gets used alot (ie. a canned search on a popular website), there can be performance issues in the long run.

  5. Laurel Tarulli

    Ryan – I went back to check your canned search Boolean string and it works great as a search in the AquaBrowser Search Box – NOT as a URL – which is what I’m dealing with.

    You’ll note the URL never changes when you perform an AB search – which means you have to create a search in your URL. Without a URL, you can’t direct a blog or website to a canned search.

    If you can build me a URL search string incorporating the search terms in my example, THAT would be awesome – especially if you can clean up mine a bit! Let me know if you come up with something.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s