Thursday, May 22, 2014

Types of librarian expertise - are they getting easier to acquire for non-librarians?

I have been recently thinking of the types of expertise academic librarians have and how recent trends in academic librarianship have made things harder.


1. Domain (basically knowledge of the research area)
2. Systems (how to actually use the search interface)
3. Information seeking (more on the structure of information and how to construct searches, etc.)

She goes on to mention a 4th type "interactional expertise", all of this is very interesting.

What I would like to consider is a) The difficulty to acquire such expertise and b) sources of such expertise, do academic librarians have a competitive advantage there?

To be precise, can someone who is not an academic librarian (here defined as someone who does not work in a academic library) easily acquire the same type of expertise academic librarians have assuming time is no object?


Domain expertise

The importance of domain knowledge for academic librarians is a well debated area (eg  Should libraries hire phds rather than just MLIS holding academic librarians?) and is I suspect the hardest expertise to acquire as it is the most specific of the three expertises.

Academic librarians definitely do not have any competitive advantage here and probably are not expected to. After all the typical academic librarian needs to have domain expertise in the area of library and information science as well as the discipline he is serving as a liaison, so it is not realistic to expect much.


Systems expertise

My understanding of  "Systems expertise" may differ from Christina, but I am of the view this type of the expertise is probably the easiest to acquire and getting easier with time.

Most databases (Scopus, Web of Science) come with extremely detailed help files, the typical academic librarian learns about them through trial and error and as a last resort referring to help files, same as any other user.

There really isn't any inherent difficulty for someone bright to sit down, read the help files on Boolean operators, proximity and field searching. The concepts maybe a bit alien if you have never encountered it before but once you got the basic concept down it's a matter of just "button hunting" on different platforms really. (Or am I undervaluing this expertise?)

Arguably a lot of systems require "Systems expertise" simply because they are so poorly designed and not due to any inherent difficulty.  

Improving user interfaces make systems expertise easier to acquire

While user interfaces of typical library systems are still poor, they are improving all the time and as someone remarked on Twitter, Google (or search technology in general) is probably not going to get any worse, so I expect in time, it will be easier for someone who is not a librarian to acquire this expertise. 

One thing that occurs to me though is librarians generally do have a competitive advantage here because they typically have direct contact with the library vendors supplying databases, discovery services etc. A minority of academic librarians like myself have additional expertise in terms of troubleshooting discovery systems, link resolvers etc because we are the ones who decide on the settings that may affect results.

So for example, if I need information on a certain feature in Scopus that isn't covered in the documentation, my Elsevier rep would just be a email away to answer my questions as we are direct customers. (Note : That said my experience is often you get very obscure or worse outright wrong answers from support staff of such companies because the developers are usually shielded)

This is where it occurs to me the typical competitive advantage exist enjoyed by librarians over our users lies. That and the fact that database vendors love to keep changing their interfaces and no-one but a librarian has the time to keep looking at them. :) 

But our competitive advantage is diminishing 

The typical researcher doesn't know who to contact if they have a question on say text datamining, though I would add in recent years this advantage has diminished because there are signs that publishers, database vendors are trying to reach out and engage directly to users not least by setting up Social media accounts like Facebook, Twitter as well as offer other direct services to users.

A interesting question to consider is this, when JSTOR or some popular database is down, do people complain to your library's social media channels or do they do it on say JSTOR's facebook page?

Why should our users ask us when they have a specific question about JSTOR , when they can get their answers from the horse's mouth?

Users are increasingly using systems not under our direct control due to cloud services

Increasingly as libraries adopt cloud services such as LibGuides, Summon, some next generation platforms in the cloud, there is an interesting side effect that libraries are becoming even more of the middle man.

For example take the library catalogue, in the past it would be something locally hosted so if anything went wrong, only our library would be affected and only our library could fix it.

But now we use Summon as our main search, and when Summon is down, every Summon using library is affected , this means ANU, Duke etc. Libraries that use Primo Central hosted in the cloud would be in the same boat.

In such cases, going direct to Summon people would be far more effective than going through the library because effectively we can't do anything anyway.


Users are increasingly using systems by parties we do not have privileged relationships with 
Another thing to consider is that increasingly users are shifting to systems that are not provided by the library (even as a intermediary). 

We are talking about Google, Google Scholar, Zotero, Mendeley (unless you get the institutional version maybe) etc.

Despite intense interest by researchers and librarians on Google Scholar (5 out of 10 of the most hots articles in LIS field for April 2014 are on Google Scholar!), librarians just don't have any privileged access to Google

Google is known for poor customer service, and  "if you are not paying, it means you are the product" also means libraries can't really demand answers, though I have found recently Google Scholar does respond to questions about entries from institutional repositories, where libraries serve as a source of information. In hindsight, this shouldn't be surprising as Google Search responds to webmasters as well but not users.

Academic librarians like myself who are tasked to be "knowledgeable" on Google Scholar, are reduced to reading up whatever literature exists by researchers who themselves spend time figuring things out by trial and error.

This may make you more knowledge than the typical person who has not read the literature, but anyone even a non-librarian who bothers to do that can achieve the same level of knowledge = you have no competitive advantage. 


Information seeking expertise

Christina's defines this as "more on the structure of information and how to construct searches, etc." It's a little vague and may blend in with systems expertise but I presume this refers to knowledge in general of the scholarly communication cycle and how it affects search.

This can be anything from knowing typical sources, how the fields and information are typically structured in general (e.g controlled vocabulary) .

In some areas, the typical researcher may actually have a competitive advantage over the librarian. For example, after all he actually does the research and knows what sources to use and what type of searches work.

It would be a very bold librarian indeed to suggest to a distinguished eminent professor that the librarian knows more about scholarly communication and information seeking in his area of expertise! 

It could be argued that academic librarians do have some advantage in that they are typically aware of new products first, though even this may not be as true anymore as mentioned above.


Other expertises

Of course searching is increasingly becoming a smaller part of the academic librarians skillset. We are told we need to learn how to support all stages of the scholarly communication cycle such as
  • Supporting grant proposals
  • Research data management
  • Reference management
  • Bibliometrics 
to name just a few.

Perhaps this is wise course of action, as searching is only to get easier.

Of course, even if the typical librarian has no competitive advantage over the typical researcher, and the typical librarian expertise is becoming easier to acquire, it doesn't mean academic librarians are doomed.

I assumed that "time is no objection" but of course it is!

First of all, beginning researchers would be much weaker in the 3 expertise mentioned, though I suspect most will eventually acquire them on their own even without librarian guidance.

Also just because a distinguished eminent professor could devote time learning about the ins and outs of Google Scholar, doesn't means that he will. 

They have their own fields of expertise after all.  

Add the synergistic effect of a bright academic librarian who has a unique blend of the 3 expertises... this is where our value lies....














Wednesday, May 7, 2014

Lazy Scholar - Interesting Chrome extension - a review and comparison with other library find full text options

Every librarian worth his salt knows that despite the rise of web scale discovery services, Google and Google Scholar are often the go-to tools of researchers.

In particular, while we prefer to direct our users to the official published version, we know that any free copy will often work in a pinch and Google Scholar in particular is the #1 tool out there to look for free copies floating on the web.

With the rise of Open access particularly Green Open access, with researchers depositing preprints into institutional repositories and discipline specific repositories (not to mention sites like ResearchGate, Academia.edu which may or may not be legal etc) such a strategy of searching Google Scholar for "free" copies is getting more and more important.

I recently discovered this Chrome plugin called Lazy Scholar that automates this step of searching for free copies via Google Scholar.

This plugin appears to be created by a Phd student Colby Vorland and does not appear to originate or is influenced via LibraryLand (Thinks to Chris Bourg for drawing my attention to it via Twitter) , so it is interesting to see how this stacks up with the Libx plugin  which is "is a joint project of the University Libraries and the Department of Computer Science at Virginia Tech."

In libraryland, we have basically solved the issue of users searching Google Scholar and linking to full text via Google Libary Link Program


Google Scholar Library Links


But what happens if users just Google (or link via other means) and land up on the publisher page, or some other indexing service that has no full text like RePec, or PubMed?

Our solutions tend to be either adding the proxy (via Proxy bookmarklet is most popular though there are many many ways), or the more powerful Libx plugin , which allows among other things to leverage unique identifiers like DOI, PMID and use the library's link resolver to find the appropriate copy.

Both these solutions focus on getting access to the official published copy with searching Google Scholar for free copies as a secondary thought. (though the link resolver might have a link that you can click to search for a copy in Google Scholar as a secondary method if it can't find a subscribed version).


Library's Link resolver provides a last ditch effort to search Google Scholar by article title, when there is no known full text via library sources


Lazy Scholar's approach is different. It attempts to locate the free copy first, though it does give you the option to add the proxy similar to proxy bookmarklet as well as leverage the link resolver via Scraping library links in Google Scholar. Arguably, in many ways, we can see how this plugin reflects the mindset of a researcher and how this plugin reflects the mindset of a researcher. Why go through the library with complicated passwords when you can get the free copy first?


Basic functionality

By default, you need to click the Lazy Scholar button, and it will attempt to scan the page you are on for an article and it will display a notification if it detects an article where full text can be found (via Google Scholar).




What happens if Google Scholar can't locate free full text? Lazy Scholar will display the below up to 2 other options.




You can always add your library proxy to the current page, but that basically duplicates stuff like the proxy bookmarklet.

Of course, adding the proxy directly to the page often doesn't work, because either (1) You may have full text elsewhere rather on the current page, or (2) It is a page that has only the abstract and no full text itself (eg Pubmed, Repec, repositories that list metadata no full text etc)

This is where the useful "I noticed you are signed on into institution login ....." link comes into play. Where does that link go to?

What happens is that Lazy Scholar will check the Google Scholar result to not only look for free full text but it will also see if a library link to full text is available -
This is the normal Findit@.... entry you find next to Google Scholar, if you have set up library links (see above).



This link will be scraped and added to the "I noticed you are signed on into institution login. Click here to go there" link.

Note : There is currently a known bug if you are using Google Scholar outside the United States. Typically when you use Google Scholar you will be redirected to a country specific subdomain, like in my case, I am always sent to Scholar.google.com.sg rather than Scholar.google.com and if you set up library links it will be on the country subdomain version and Lazy Scholar won't be able to detect it.

The workaround is to go directly to scholar.google.com/scholar_settings (notice the lack of the .sg) and set library links there as well.


You can also turn on auto-detect in the options (right click on LS icon and select options), though you have to set up permissions.



If these recommended settings are set, when you visit any page that it detects as an academic article, and if it detects free full text via Google Scholar, there will be a popup on the top right and clicking on it will send you there.




If no free paper is found on Google Scholar, you will see a different popup.




Could be wrong but the autodetect doesn't seem to give the option of "I noticed you are signed on into institution login. Click here to go there" link.

In any case while I haven't done a full test, Lazy Scholar seems capable of recognising titles from a wide variety of sites including but not limited to

  • Sciencedirect
  • RePec
  • Emerald
  • PubMed
  • ACM Digital Library
  • Taylor & Francis
  • Oxford Journals
  • Science AAAS

JSTOR doesn't seem to work at all and Wiley works sometimes (there is a bug for some). I am unsure how detection works (Libx uses COINS)

In any case, Lazy Scholar's greatest benefit is when used on pages that do not host full-text themselves.

Just for fun, I tried it on our Dspace institutional repository, which currently consists mostly metadata of articles published by our researchers. Lazy Scholar works beautifully (though where it links to is interesting)




My testing shows auto-detect can be a bit buggy, it may take a while to pop something up and sometimes it is faster just to click on the button manually. Even manually clicking on the button will occasionally be slow.

That seems to be the main function finding free full text via Google Scholar, but there are other features.


Display times cited and altmetrics 

Lazy Scholar also tries to be helpful to assist in assessing the quality of the paper.

It also displays Google Scholar Times cited and Web of Science times cited scores(available if you are in-campus at an institution that has Web of Science) - scraped from Google Scholar.

Altmetric scores are also included not sure how useful this is, but it was probably added because it was easy to do, though you do have to opt in via options.





This plugin is pretty cutting edge, with support for the new PubMed Commons, so you can see if there are any comments in this new system.






Citation/Reference Management

As the focus of Lazy Scholar is to make locating pdfs for download easy, it naturally has features like auto-renaming, ability to export references to EndNote, Google Scholar Library etc, or just copy and paste citations in APA, MLA etc.




Other ways to get papers....


What happens if the paper you want is not free and adding the ezproxy does not work?

Most researchers know that a workaround is simply contact the author direct. LazyScholar, makes this easy by trying to locate the author email and displaying it for you.




Slightly more controversial I think from the librarian point of view is the support for #icanhazpdf





In case, you are unaware this is the practice of tweeting for articles you want behind paywalls you can't access with the hashtag #icanhazpdf (Some analysis here)  and most of course there are debates on the morality and legality of this. 

Some librarians would much prefer researchers come to us in the libraries (what about those who don't have access to academic libraries though?) so we can do document delivery or even consider subscription (if demand is high enough), but from the point of view of researchers this might be too slow.

Of course, this feature is optional, but it does reflect the mindset of researchers who just want access to what they need as fast as possible.

I guess if you are concerned about support for #icanhazpdf, the following possible feature in testing mentioned in the  blog post  PDF exchange: need beta testers might concern you even more.


Conclusion

I find LazyScholar a useful plugin that I can expect to use (though there are bugs such as slowness and on one of my systems LazyScholar just refuses to run despite starting a brand new profile), but more than that I see it's development as a fascinating insight into what researchers want. 

LazyScholar definitely resonates with what researchers want because I was retweeting this and the next thing I know, one of the researchers at our institution who are following me picked it up and blogged about it

Compared to Libx plugin  a library backed plugin, Lazy Scholar has less integration with library systems (besides inserting the proxy, which I suspect is independently discovered by researchers all the time without library input) and with a focus towards getting the full text by the fastest and any means not necessarily through the library.

Lazy Scholar focuses first and only on Google Scholar to search for free copies, a strategy that will pay off as time goes by, their own statistics (though it is early days) show about 1/5 of articles can be found that way and only them tries library access via Ezproxy (not link resolver)

Again this differs from I suspect most library specific efforts where searching Google Scholar for free copies is the last resort.

That said the scraping of library links from Google Scholar is outright brilliant.

Ideally on any page with article metadata, the ideal way is to somehow evoke the library link resolver to be sent to the appropriate copy or place where you have access.

Adding the proxy doesn't solve the appropriate copy problem so while it works most of the time, it is not the most accurate and may fail even if the library has access.

Both Lazy Scholar and Libx provides ways to activate the link resolver but in different ways and degrees.

Libx supports COINS as well as autolinking of unique identifiers like DOI, ISSN and leverages basically the link resolver, with no reliance on Google Scholar (the "magic button" function is the only thing that usesGoogle Scholar).

On the other hand, Lazy Scholar relies almost entirely on Google Scholar, pulling in free text or via library links (using the librarylink resolver).

This caters to users who Google (not Google Scholar) or otherwise managed to get to some page with article metadata and don't know where to get the full text. They can now use the link resolver (in a way) to get to full text.

The main weakness I can see is if something isn't covered in the index of Google Scholar, Lazy Scholar can't do anything beyond adding the proxy. This is of less concern then you think because Google Scholar has one of the largest if not largest index of Scholarly material, so almost everything you come across in other sources would probably be indexed in it.


It is probably the librarian in me talking but LibX still feels better to me since it handles books etc, it will be interesting to see if Libx can incorporate this particular feature, though philosophically you can see the difference between the two.

That said, if you are not affiliated with any academic library, Lazy Scholar is by far superior as it automatically gives you free full text.


Share this!

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License.
Related Posts Plugin for WordPress, Blogger...