Name:
Implementing Seamless Access
Description:
Implementing Seamless Access
Thumbnail URL:
https://cadmoremediastorage.blob.core.windows.net/18c1de51-cee0-4ade-a4a2-0711d1ca937c/videoscrubberimages/Scrubber_1.jpg
Duration:
T00H31M57S
Embed URL:
https://stream.cadmore.media/player/18c1de51-cee0-4ade-a4a2-0711d1ca937c
Content URL:
https://cadmoreoriginalmedia.blob.core.windows.net/18c1de51-cee0-4ade-a4a2-0711d1ca937c/Implementing Seamless Access.mp4?sv=2019-02-02&sr=c&sig=MsvDHF0xY46e6U%2Bvaz8UIMNBfkGkSKA3%2Bjv31ocahHA%3D&st=2025-01-15T06%3A32%3A52Z&se=2025-01-15T08%3A37%3A52Z&sp=r
Upload Date:
2024-03-06T00:00:00.0000000
Transcript:
Language: EN.
Segment:0 .
See us all coming in here now. Well, welcome to the live session, everyone. It's great to see everyone. This morning, this afternoon or this evening, depending on where you are. We have our three presenters in the live session here today, today, Heather Staines from SeamlessAccess and delta think, Julie Xu from tripoli and John feltz from Coastal Carolina University.
As I said just a moment ago, there are two goals for this live session. The first is to answer any questions and to go into more depth about the topics from the recording. The second is to explore this area of SeamlessAccess and distributed authentication and see where nice so and its membership can have an impact on the profession.
And that can take the form of educational topics, educational sessions, commissioning publications, or forming working groups for standards and best practice development. So that is what we are here for today. I'll see. Thank you for the link to our shared document.
I was looking there just a few moments ago. I didn't see any questions there and I didn't record any questions from the recording session. So I'd like to just start with opening the floor for questions. And maybe, John, since we ended with you in the recording, I'll start with you.
One question that comes to mind is, is how long does it it take for a library to implement Federated access? You said, you took it on the fast path and did it in a few months. How difficult is it and what do you think a typical implementation looks like? Well, like most of these technical things, it depends. It depends on which direction you're going to go.
I have found for libraries, you know, when you try to leverage and utilize a locally installed shibboleth instance, a lot of times you fall under the. Direction of pomposity and they use shibboleth for a lot of different reasons that we as libraries wouldn't necessarily use, you know, for accessing scholarly resources. So it can be a rather unintuitive user experience and I won't get into disambiguation and all that gets a little complicated and off track.
But we're seeing that where if you come from an institution that has like a locally installed shibboleth or that is part of, say, in common Federation and then the library is part of, let's say, open Athens, then you'll see two entries when you're in the IDP discovery, and that can be kind of confusing to users. So with that as a backdrop, I have found and it seems like a lot of library and library affiliated folks are going the OpenAthens group.
And when I first started looking and listening, going to library conferences and hearing what early adopters had in the US had said about open Athens, I mean, they were crying the blues, a lot of them not to call anybody out. But I remember a James Madison University presentation and, you know, they were bemoaning that they were on a fast track of 9 months to implement OpenAthens.
Now again, it scales an issue, but they have a whole lot more URLs to consider changing and all that kind of thing. A lot more different platforms that they have to consider, but. The longer that this relationship between epsco and OpenAthens has gone on, the easier the process has gotten. So I wouldn't necessarily recommending the Warp Speed approach that we took, but I was hell bent, if you'll allow me that, to get open.
Athens in place and rolled out by the beginning of the fall semester because that's ideal time obviously and and academic library land and it was just the way that it fit within. We were part of a consortium in South Carolina and we were part of a cohort that started in June or late May. So that gave us about three months. But I got to say, I was sweating it a lot more than it actually turned out to be, because once we got rolling there was a lot of epsco has gone to such great lengths to automate and just make the process easier, the implementation process, and it just went very smoothly.
There were ways to automate changing all of our URLs so that in our at a to z lists and our discovery platform, those were pretty easily converted away from the open. I'm sorry the easy proxy prefixed URLs into what they call an and link. So we got it all done in about three months and it was surprisingly easy and for anybody that's.
Considering making the switch from IP based authentication to Federated access, I'd highly recommend that route. And you say that users were finding the SeamlessAccess button before you really talked about it? Is that what you were saying? Yeah, that was pretty interesting, actually. I always like to share this story because I was just happened to be and we I don't have to actually.
There was about a three week period where OpenAthens was set up on the out on the cloud on the OpenAthens side. But we had done nothing to localize it yet. We hadn't changed any of our links. So all of our links in discovery and a to z lists were pointing to or using easy proxy as the authentication method. But I would just happen to be in the admin platform on open Athens, just kind of getting myself acclimated.
And I started noticing all these successfully authenticated sessions. And it was wasn't just one or two either. I think I said there's as many as 130 Plus in one day. And I scratching my head and I talked to a colleague and was like, what in the world is going on here? And the only possible answer is, is that users were out on the open web, which is how the vast majority of them start their research experience.
They don't necessarily go through the library website anymore and they don't go through our a to z lists. Most researchers, especially higher up grad students and higher level researchers and faculty. They just start where they start, wherever that may be. And then they land on the vendor's platform. And because this OpenAthens was in place for us, we were appearing in the ADP discovery service and for instance, Science Direct comes to mind Elsevier and they had SeamlessAccess also implemented and it was working like a charm for our users.
We didn't have to give them tedious library instruction sessions. I didn't fail not one support question during that time. They just intuitively knew how to authenticate and get access to the content that they were rightly entitled. So it was thrilling in one sense, because it was like, yes, we finally have a mechanism where we can provide authentication at the point of need. But I was a little bit bummed too, because I was like, Oh my goodness, how many hundreds or thousands or hundreds of thousands of users were turned away, because there was just no mechanism in place three months ago, six months ago, a year ago, or whatever the case may be.
So I thought that was a pretty powerful message in terms of the ease of use and the user experience for Federated access. So somewhere a bit flipped or a record got imported to that distributed discovery and suddenly users were finding it. Heather does this resonate with other stories you've heard about SeamlessAccess implementations?
Yeah I don't know so much about what happens in the implementation launch window period like John is discussing. But one of the things I was thinking about is the user research that a number of the publishers have conducted. And if you've ever tried to get access to a publisher platform, you know, there's so many ways you might possibly get access. And the researchers could easily be confused.
I mentioned in the presentation, you know, that one of our publishers had up to 11 ways you could potentially get access, which is probably 10 ways to many. But when we learned about streamlining those options, making it the user journey a little bit more straightforward, thinking about where the button is going to appear on the page and standardizing it across the different services, it really did make a difference because as a researcher, you want to do your research, you're looking for your articles.
You have a million things to take precedence over the thought process of what is my log in for this particular platform? So I think in the same way that folks on the library side were just gravitating towards what seemed like an intuitive pathway for them. You know, on the publisher side, we need to kind of and I haven't been a publisher for many years, but I still put myself in that category.
We need to kind of streamline, edit ourselves a little bit, pun intended, and try to make sure that the next steps are going to be clear for researchers so they can do what they need to do. Excellent Julie maybe speaking from the publishers side a little bit, it sounds like you had a great experience getting the IEEE platforms integrated, but in the course of that and maybe in the course of talking with other publishers, what kind of obstacles do you see from the publishers side that's preventing them from implementing SeamlessAccess access?
Yeah, there could be several different types of obstacles. The first could be the publisher's general lack of understanding of SeamlessAccess to the whole idea of i.e. 21 and SeamlessAccess. It may be a bit intimidating to some small and mid-sized publishers, but the SeamlessAccess website provides excellent instructions and guidelines on all of these major elements, like the user interface, discovery and persistence.
So if someone from a publisher takes some time to read through these documentations, it should be relatively easy to digest the information. And besides, the SeamlessAccess team is always there to answer questions and provide guidance so that it's easy to overcome. Another type is some management concerns about security and privacy, and some publishers are more cautious than the others.
And so but now so many publishers have already successfully implemented seamless, SeamlessAccess. So the risks of security, of privacy seem to be relatively low. And another concern is that some publishers may have too many remote authentication options and they try to see how they can implement a SeamlessAccess access, but without sort of like hiding all the other options.
So they need to do some customizations. And so in that case, you can study different the websites from different publishers, different platforms. They more or less do some customizations. So we have to figure out what's unique on your website and on your platform and to figure out what kind of like the customization may be needed. The final thing could be the some publishers may worry about the level of work needed to implement SeamlessAccess to, but after some careful planning, we found that the work is not as daunting as we anticipated.
And so we had one or two designers working on the buttons and the other interface features, and we had one programmer working on the API and another one working on tagging the buttons. So overall, the work amount is really manageable. John's story of users just finding this is so refreshing and that that pent up demand of use that was going unfulfilled.
Are you seeing the same kind of thing from the publishers standpoint that through SeamlessAccess this is additions of people that are finding their way to the content through their institutions then otherwise might be turning away. Yeah, definitely so. Because going through library websites, going through easy proxy is always a little bit more like a less intuitive way of for the workflow.
Right so the users, if they can get to the website and see an access button there and they can directly log in, that's always a Plus. And also remember that because so many platforms are already implemented, this so a user visiting one website may already have the affiliation remembered. And so they will get to another platform. And so it's already there. And so so I think it's a great network, as Heather once mentioned.
And so that makes the user's experience. So much easier. If I can just add on to that. Peter, some of the pathways for Federated authentication on publisher sites would require the user to select a Federation before they could even see their institution. And, you know, I think that's asking a lot of users. It's a real sort of kind of inside baseball type option.
And if they have a clue about what a Federation is, that might point them. But it's really not the way you need that you want them to go. So kind of removing even that one step from them can make a big difference. good point. Yeah and I think that's how SeamlessAccess makes Federated access even better, because without SeamlessAccess access, that experience is going to be different on every service provider side, on every vendor's platform.
I mean, the button might look different. The way finder certainly behaves different to Heather's point. Sometimes you have to pick a federation, which what's a Federation to a PhD student trying to research, you know, whatever. So so it does ask a lot of the users. So, so with SeamlessAccess access, the whole point is, is a very consistent call to action, a very consistent user experience.
And from what I've seen so far in the various integrations that various vendors, service providers have used is a very consistent experience. I feel like I've been monopolizing the conversation here. So I'm going to pause for at least 15 seconds, maybe 20, for people to unmute themselves and ask a question or to put it into chat.
If you do put it in the chat, I can relay it verbally for the recording. But others that are here. What questions do you have about SeamlessAccess from a publisher standpoint, a service provider standpoint or a library standpoint? Hi this is Mike.
Pam go ahead. I work for U. L standards and engagement company. So we're an sto publishing our documents and stuff. And I was just curious. New to the whole SeamlessAccess platform and what looks to be done for like universities, libraries and stuff along those lines. One of the questions, we're always asked and other CIOs are always asked is from our users or stakeholders, is there an easier way?
Can't you all play nice together and share information an easier way. And one of the things I was looking at, could this platform of the SeamlessAccess do something to where users wouldn't have to log in to each of the actual sites of the cios, but maybe we could do something along the lines to where it's that SeamlessAccess to go across the multiple platforms and allow users easier access to multiple sites with that single sign on aspect.
I mean, the taller order that everybody wants is one site for all still information to be used, but that's a way taller order. Is this something that could bridge some of that gap? If I may, I want to expand the sdoh acronym. I believe that standards, development organizations. OK Yeah. Sorry so, mike, I can jump in here.
This is Todd carpenter with miso. It's a really interesting question. You'll be interested to hear that many of the ceos, the larger stock associations like I, tripoli asked me, et cetera ACM, are actually already involved in using or implementing SeamlessAccess. The challenge is with many of the implementers in that community, they don't run the same sort of Federated identity systems that some do.
But not every, every corporation will be running a Federated identity system in the same way that a University would. So it's not and actually one of the. Driving forces behind the launch of SeamlessAccess back in the day. Six, seven years ago, I think, was actually the pharmaceutical industry who face a lot of similar challenges that, say, a large manufacturing company might when implementing Federated identity management and access control across a multinational organization.
It's been a while since SeamlessAccess has had much interface. I mean, heather, you might know how much interface we've had with the pharmaceutical industry since we launched as a production service. But recently, you know, we do hear from other types of organizations, non academic, non scholarly communications organizations about whether or not the functionality would work.
And of course that's a possibility. But I can't recall a pharma approach since I've been involved, which is about 2 and 1/2 years now. So I mean, it's something that we certainly. Many of the larger publishers, as I said, as the organizations have been, they do this for their journals, publishing program. Right whether or not they're doing it for the standards, I mean, you'd think that the access control system would be the same on other side, but sometimes it's not.
And there's sometimes the publishing is running on a different technology stack. I'm looking over the attendee list to see if any of the CEOs are. Here, and I don't. A quick perusal of the attendee list I don't see anyone from. And on this particular session, whether I see anybody from the SDO community here.
Yeah I didn't see anybody quickly like that either. We might put that down as something to take forward and explore as an organization. That being our second goal, in addition to information sharing and asking questions, is where there might be areas that meso can help extend some of this work. I see in chat, there's a question. Is there a way for SeamlessAccess to work with shibboleth in the future?
Anyone feel comfortable enough to tackle that? I will take a stab, if nothing else. No one's going to rescue me. OK so was that you, john? Are you just waiting to see what I pull together? Well, at the last second there, I was like, oh, what's Peter going to say?
Yeah, I wasn't sure if that was specific to the disambiguation problem or just the way that SeamlessAccess works with or federate with Federated access and how Federated access many times utilize the shibboleth. Yeah so I work with a service provider and so I am working with institutions that have shibboleth on their campus.
And we, we make bilateral arrangements between the, the identity provider and the service provider. And that's being done with shibboleth. Seamless access is something that you can layer on top of that shibboleth infrastructure that you get yourself registered in a federation, you make yourself known. Suddenly you can, as John was describing, a peer in a list of schools for which that institutional access button is live and something that's workable.
So something like, like shibboleth or Azure Active Directory or Google directory is, is some of the foundational elements that you need that, that then something like SeamlessAccess layers on. So if you have shibboleth, you're already, you're already getting started. And I see a comment.
Thank you, Tracy. Seamless access is simply the identity discovery service that sits on top of shibboleth. A very concise way of summarizing my random ramblings. Thank you, Tracy. So I am looking at the clock and we have about 5 minutes until the end of this session.
I want it to leave this little bit of space for us to reflect on hearing about SeamlessAccess the program, SeamlessAccess from the publisher perspective and SeamlessAccess from the institution subscribing institution perspective. What kinds of things would it be advantageous for an organization like ISO to take on?
Is is there some more educational elements? Is is there work that nice? So could champion champion with SeamlessAccess to make it work even better? What are your thoughts on that? And again, I'll open it up for discussion from anyone. So I have two things I want to talk about.
The first is many libraries in the us, Canada, they already have shibboleth, write or open others. But somehow these libraries don't want to approach publishers to set up their service. To work on that. They That they say that there is only for internal use. Yeah enough for library resources and they some I've talked to many librarians.
They don't want to talk about it and and so how can to help educate or push or persuade these libraries to open up their mind and to put library resources under shibboleth and open lessons so that end users can benefit from SeamlessAccess and benefit from Federated authentication. That's number one.
Number two is that there are still many libraries, especially in other countries. They don't have shibboleth yet and don't have open ATMs yet and for cost for many other issues. Right and some of them still use easy proxy. And so in that case, how is it possible for us to make easy proxy, let's say seamless, like SeamlessAccess. And so because we do receive requests from libraries saying that I use we use only easy proxy.
Can you put our easy proxy log URL into your institutional signing box? And we tried and it worked. But understand, this will not be remembered by SeamlessAccess. And so we have to figure out how we can remember it in ourselves by putting it into the user's browser. So but if an organization or someone else can help make this work for users other than just open that and ship love, that'll be even great.
That'll be great. Wonderful I'd just like to add that education is always beneficial, spreading the word. You know, not every country or portion of a country has a Federation that covers its institutions. And so we do hear from service providers that are working in areas that don't have a strong national identity provider that's going to cover them.
There are also publishers who work with, say, hospitals, and they're not going to be within the educational consortium dropdown list. So I think as we move forward and think about how else the other ways that SeamlessAccess can be of a benefit, we're going to want to look at some of those nuances. And complexities of access, either by global region or by industry segment that make things a little bit more complicated.
It great. Any other closing thoughts? Well, we have reached the time to close out the interactive portion of this session on collaborating to implement SeamlessAccess. I want to thank our panelists heather, Julie and John for sharing their knowledge and experience with us.
And thank you to the participants here in the interactive session. If after the session closes, you have new ideas on the topic. Feel free to contact me as the moderator or any one of the nice host staff that you see around, and we'll make sure those thoughts get to the right place. And with that, we will call it close. Thank you, everyone, and have a great rest of your day.