The University of Manchester IGeLU 2020 Presentation Transcript

Jun 24, 2020Transcriptions

Tim O’Neill:

Hello, my name is Tim and I’m electronic resources coordinator based in the Digital Systems team at the University of Manchester Library. I’m delighted to have the opportunity to share with you some of the recent digital library infrastructure improvements that we have made at the University of Manchester Library over recent times.

Tim O’Neill:

I’ll be speaking about the Third Iron integrations we’ve implemented on Primo, explaining what we’ve done, how we did it, and the impact that it’s had. I’ll talk to you about the Report a Problem button that we’ve implemented on our Primo interface that enables students and academics to more easily report an issue to our dedicated help desk service. Lastly, I’ll speak about BrowserStack, a tool that enables us to mimic different devices and different browsers, which helps to troubleshoot new resource access issues, and also to help us test interface changes that we make.

Tim O’Neill:

So the University of Manchester Library, it’s the third largest university library system in Britain, after Oxford and Cambridge. And we hold the largest academic digital library collections in Britain. You can see the figures there for the content that we have. I found out recently we have roughly 110 kilometers of shelving across all our sites, and that’s equivalent to about 350 football pitches. So we have a large physical presence, as well as a large digital presence. We welcome over three million students, researchers, and visitors per year, and we have about 40,000 full-time equivalent students studying here.

Tim O’Neill:

So these are the systems that we have in place. So, Ex Libris Alma, Primo. We’ve had Alma since 2013. Primo, pre 2011 I believe. Recently implemented Leganto as our reading list system. Authentication and access software we use is OCLC’s EZproxy, which we’ve had since 2016. We use Third Iron’s LibKey Nomad browser extension alongside Lean Library’s browser extension. We’ve also had Third Iron’s BrowZine journal engagement platform in place for a number of years.

Tim O’Neill:

So the first thing I wanted to speak to about was our Third Iron Primo integrations. And you can see on the right-hand side, we’ve got screenshots of how this integration is working. So how it looks to a student or an academic. You can see the first screenshot has a few journal contents link, which provides the link for any journal that’s available through BrowZine. It provides you with a link to that journal, so you can view the journal contents within the BrowZine interface.

Tim O’Neill:

The second one is a view issue contents link, which takes you to the… It gives you an in-context view of the articles, so it shows you the other articles that appear in the same issue.

Tim O’Neill:

And the third one, and perhaps the most impactful integration is that of the download PDF link. So this provides a single-click link using our authentication layer that takes a student or an academic from our discovery tool straight to the full text PDF where it’s available. This means fewer clicks for a student or academic to take. The traditional route would be having to navigate through the different content provider links under our Find Online tab on Primo. So this in a way, it bypasses that and gives a more direct route with fewer clicks.

Tim O’Neill:

The thinking behind implementing this was it could provide immediate benefits. It could improve the user experience of accessing our content. Implementation was straightforward too, so Third Iron were really responsive and provided excellent support on that.

Tim O’Neill:

So the first thing is we’ve using Journal Jacket images now from BrowZine, so you can see how it’s making for a more aesthetically pleasing interface. So you can see it’s improving the look and feel of our discovery tool. You can see there the before and after pictures.

Tim O’Neill:

So what does the BrowZine API provide? We do, I’m informed, have a fairly customized version of the integration with BrowZine Third Iron. And this is something we are going to look out revisiting to improve. So we did move quickly when we knew this was available, and that did result in us doing maybe a more customized version than we now need to. So that’s something we’re going to revisit. But you can see how in principle, there’s a journal lookup for ISSN, there’s an article lookup by a DOI, and you can see on the right where that is appearing and the information that’s passed through using the BrowZine API.

Tim O’Neill:

This is more detail of how the journal lookup works. So this is what’s sent from Primo to Third Iron via the API. So you’ve got the input ISSN, you can see there’s a cover image URL, and then the BrowZine web link, which is taking you to the journal in context.

Tim O’Neill:

And here’s some more detail on what’s going on with the article lookup. Again, you can see the cover image URL, the BrowZine web link, and then you can see that there’s a DOI at the top. And you can see more what’s happening there, and you’re getting the article then in the context of the issue it appears in BrowZine.

Tim O’Neill:

So this is to highlight some of the local improvements. We’re fortunate to have a software developer, well, a team of software developers at the University of Manchester Library, and they wanted to improve some of the performance or some of the loading speeds, so that they cached things like the images and they made improvements just to the loading time to try and get the most streamlined and efficient implementation as possible.

Tim O’Neill:

So this is what it looks like. When you find, say here, I have a journal I found on our Primo library search discovery tool, the view journal contents link, you click on that link. It then takes you to the journal on BrowZine. That’s straightforward.

Tim O’Neill:

View issue contents link, here I found an article in nature on graphene. The view issue contents link then takes me to the article in the context of the issue it appears then.

Tim O’Neill:

And lastly, there is the download PDF link. So a staff or student or an academic clicks on the download PDF link, they gets a short splash screen which displays before they then get, in most cases, the full text PDF opening in a single click. And as I say, that includes our authentication layer. So there’s no need for the academic or the student to navigate complex publisher platforms. They simply have to log in if they haven’t already logged in through our university login page, which we integrate with our EZproxy authentication and access software.

Tim O’Neill:

So moving on to one of the more interesting parts of this, is the impact that it’s had. At first sight it appears like it’s going to bring immediate improvements, but what evidence do we have to show that? So I’ve done an analysis in an attempt to get a rough idea of the impact that we were getting from this. So how many people were benefiting? Is it changing the behavior of how people are interacting with Primo?

Tim O’Neill:

So what I did, I compared click requests from Primo analytics. Click requests, a measure of the number of times at least one service has clicked within our Find Online tab. I compared this to a full text download request through LibKey and BrowZine through the Third Iron stats we’re getting. So click requests in the whole of 2019, we had 1.9 million. Full text download requests from LibKey and BrowZine, we had just under 438,000.

Tim O’Neill:

This means that just under 20% of all articles downloaded in 2019 used the LibKey technology from Third Iron. So used this download PDF integration. So that means one in every five article downloaded, there or there abouts in 2019 used this LibKey technology. That was over 438,000 individual articles. I then did an analysis to compare a nine-month period where we didn’t have the LibKey integration. So this was January 2018 to September, October 2018 to the same time period the following year, where we did have this integration.

Tim O’Neill:

What I found was that the number of click requests decreased by a similar number to the total number of increases in the PDF downloads from the LibKey integration. So this suggests that the people who were using the traditional route previously, so navigating through the different publisher links, were now benefiting from the single-click link that they were getting courtesy of the download PDF link.

Tim O’Neill:

I was interested to know, well, what kind of time-saving would that bring? And I found that on average, it takes about four clicks when you’re going through the traditional route, versus a single click when you go through this route. And to summarize, I found over the course of a year, over 100 days were saved across everybody accessing across the whole university. And that average saving every month was 73,000 clicks. So, put simply, we’re saving a lot of people a lot of time, and we’re connecting people to content quicker than they previously were being connected to it. So that to me, is a sizable improvement to the user experience of accessing e-journal content.

Tim O’Neill:

We also integrated with Unpaywall. So we did this. I think we found another institution that had done this, and we saw it bringing immediate benefits to promoting open access content that was available. So this was pulling in a link to article content that was available through the Unpaywall database. And you can see here, some of how that was doing. So a call was made to the Unpaywall API, and if the call was found a successful match, then it would pull in an open access link via Unpaywall.

Tim O’Neill:

However, we have found out that this is now included in the Third Iron integration, so we are effectively running two integrations that bring in the Unpaywall link. So this is, again, something that we’re going to work to improve because we don’t need to have two Unpaywall integrations. So we’re going to work to include the Third Iron Unpaywall PDF integration, which you can see how that appears here.

Tim O’Neill:

Other improvements that we’re going to make potentially are to look at LibKey Link, which you can see here. So this is improving that the link resolver, so particularly when you’re accessing content on an external database, you can see how that works. There’s a lot of material provided by Third Iron that goes into great detail about how this functionality works. There’s the LibKey.io tool, which enables you to search on DOI and PubMed’s ID. So that’s currently under review. These are other improvements that we’re looking at, which should make it easier to connect to content.

Tim O’Neill:

Software developers have also developed an infographic, like a live feed showing how people are interacting with our content. So you can see we’ve got a live feed showing the articles which are being accessed. So this is potentially something that could be developed further with new metrics included. So this is the kind of thing that we’re thinking about developing even further over the coming months.

Tim O’Neill:

I’ll move on now to the Report a Problem button. This makes it easier for a student or an academic when they encounter a problem with access to any of our digital library content. They get the option to click a button, Report a Problem, that enables them to report the issue directly to the dedicated help desk, which is our electronic resources help desk service.

Tim O’Neill:

Before we implemented this, a user had to manually email the help desk. We’d find often the email that they sent didn’t have key information in it. So it might be quite general, “I can’t access an article on the journal website.” So we’d have to then ask, “Well, what’s the name of the article? What journal website are you on?” And it often involved quite a lengthy email dialogue between the help desk staff and the student or the academic to establish what the issue was.

Tim O’Neill:

So these were the requirements. These are what we wanted. We wanted to display it only electronic resources, we wanted to pop-up dialog box, we wanted to capture the users details when they were logged into Primo. And we wanted the details of what they were viewing at the point at which they click the button. We wanted their browser information. And then we wanted the free text field, where they could give any additional information that might be helpful in describing the problem.

Tim O’Neill:

This is what’s happening behind the scenes in the Primo app studies, the details, the bibliographic data that we wanted to be passed through.

Tim O’Neill:

It was built using Angular Material user interface. So here you have how that appears on the front-end. And there’s an image there showing what the Report a Problem button looks like for a member of staff or a student or an academic.

Tim O’Neill:

This is the item-level information that’s passed through. You can see that on the slide, and the result. I was hoping that that would display, but unfortunately it hasn’t. So what would happen, you click on the links available online, so any item that is an electronic resource will give you the option. You click, you get the pop-up, and that then enables you to fill out the details for, well, any extra information that you want to provide to the help desk. If any of you are interested in implementing this, then you can get the code and the information that you need on GitHub, and I put the address up on there.

Tim O’Neill:

So what has the impact been of this? Well, we noticed an increase in the number of inquiries that the help desks were getting. And I think that is because we’ve made it easier for people to contact us than previously, as we had a button appearing at the point at which a user needed to report an issue. It ended up being our busiest year in three years just after we implemented it. And I looked at our current inquiries that we have in our ticketing help desk system, and two in three of them had arrived through this channel. So that’s a good proportion of our inquiries that we’re getting are from this route.

Tim O’Neill:

So lastly, on the BrowserStack emulation tool that we’re using. We pay for a license for this tool and it allows you to mimic different operating systems and different devices. You can see here we have what our library search would look like on a particular device, and here it’s an iPhone. You can see a Nokia phone on the right where we’re looking at a content provider platform.

Tim O’Neill:

So this really comes into play when we have complex or complicated technical access issues to our e-resource content. These three are real examples that we’ve had reported to the help desk. So HTTP status error, server error application, 504 gateway time-out. Often this is could be a browser issue. So it’s really valuable to have access to a tool that allows us to replicate the browser that the user is using and even the operating system or the device that they’re accessing on. And that helps us then provide more information to the content provider’s technical support team, which will hopefully enable them to pinpoint the problem in a quicker and more timely fashion.

Tim O’Neill:

We can also use this tool to test any Primo interface changes that we make, to check that there are no issues with how they appear on different devices. With the Third Iron integration, we found that the view issue contents and the download PDF wasn’t necessarily staying on the same line, so we were able to add a bit of padding around the characters to then make for a better displays for it. We were only able to do that by the use of BrowserStack. The only other way of doing it would have been to have a physical device for all the different tablets, smartphones. And that’s, practically speaking that’s not going to happen. So having a virtual emulation tool is very useful.

Tim O’Neill:

So to summarize, I think we now can enable students and scholars to access e-journal content a lot quicker than they used to be able to, courtesy of the LibKey integrations that we’ve done. We now offer a quick and easy way to report problems to any resource content on Primo through the Report a Problem button. And we can now troubleshoot access issues by mimicking different devices and setups on BrowserStack and testing different devices too.

Tim O’Neill:

So ultimately, we’re saving a lot of people a lot of time, and hopefully people are connecting to our digital library content quicker, and hopefully spending more time doing the important research and less time having to navigate our interfaces. So hopefully getting them to a citation, to an article for instance, to the full text much quicker than they previously were able to.

Tim O’Neill:

Thank you very much for listening. I believe there’ll be opportunity now to ask questions. So I’ll handover back now to the moderator of today’s session, but thank you for your time.

Speaker 1:

Hi. I hope you can hear me now. Let me stop the share, sorry, quickly. Well, let me get a question asked first. How many developers do you have in your team, and are they working for the library only, or are they part of university developers team?

Tim O’Neill:

We have a dedicated team within the library, and there are, if you include the manager of the team, there’re a total of six full-time staff in that team.

Speaker 1:

Can you tell… Same one. Does BrowserStack notify you of any issues that you need to check? Or do you need to check every environment to analyze the problems?

Tim O’Neill:

We tend to check the most popular devices. So you have to manually go in and pretend you’re accessing on the physical device in the virtual environments. So to my knowledge, there isn’t a way of, say, testing the same development across a number of devices. Although, that’s perhaps functionality that they may introduce at a later date.

Speaker 1:

[inaudible] demo, that demo that didn’t play during the presentation to ProposalSpace. So that should be available on the IGeLU website. If not now already, soon. And one last question, what were the costs of the various extensions? That’s a challenge in one minute.

Tim O’Neill:

The costs, so the price. I think for the latest, you’re probably best talking to the respective vendors. I know that they have both frozen the price from the previous year. So I’m optimistic given the impact that I’ve shown that they both have, that we’re going to continue with them beyond this year.

Speaker 1:

Okay. Well, thank you, Tim. That was a great presentation.

Tim O’Neill:

Thank you.

Speaker 1:

And thank everyone else for attending. I’m going to end right on time, and I’ll see you all at the next one or somebody else will see you. Bye.

Tim O’Neill:

Bye. Thank you, everyone.