i>Clicker Integration with D2L/Brightspace

I’m sure i>Clicker won’t be particularly happy with this post, that’s ok, because I’m not happy with them. The one thing as an LMS administrator that you should feel very, very sanctimonious about is sharing data. We have an internal policy that we’ll never share student numbers. I also think that we shouldn’t share usernames, however some folks feel that’s ok. I guess it depends on your username convention. First dot last as a convention is great, except in these cases when you’re trying to maintain privacy. Now, if you’re making a single sign on connection through LTI, you can have the originating LMS username obfuscated, which essentially boils down to a paired database table that has two usernames in it – the one handed off to the third party via LTI, and the one in the LMS. Typically, I’m a little bristly about this as well, because you become reliant on the system not changing how this is handled, and if there’s an HTTP call mixed in there, it still could be sniffed out while in transit… but that’s not what this post is about.

Since last year, I’ve been bugged by i>Clicker to do an integration to make sign-up with their service seamless for students. There has been some requests for integration between i>Clicker and the gradebook in D2L from some of the more heavy users. The first request, I’ve always put off because, frankly, I don’t do anything a private enterprise wants me to do, they aren’t my boss, nor are they a member of my community. They serve one purpose, and that’s collecting money. The second, however, does benefit a select user group on campus. We were thinking about this since last year’s summer of integration, and April had a couple minutes free (more on that in forthcoming blog posts), so we scheduled some time to finally make it happen.

We schedule a call, to walk us through the integration and see if there’s anything that we have questions about. I don’t need anyone to walk me through an integration, but I often like to raise privacy, data collection policies, and other awkward questions to the poor sucker who’s on the end of the other line. Typically they are ill informed. I didn’t even get to the awkward stage, as a request was made that was frankly shocking. I was asked to turn on passing the Org Defined ID – or our student number – to facilitate the connection. Not just at the tool level, but at the configuration for the Tool Information. See below:

config_tool_consumer (1) Now if I understand this panel correctly, it not only changes the configuration settings for the tool in question, but for all the tools. ALL the tools. So not only i>Clicker, but anything else you have connected through using the External Learning Tools administration panel. I asked our technical account manager about changing it, and he basically said, “yeah, that’s not good.”

So in the middle of this exchange where I explain how we don’t pass the student number under any circumstances, the i>Clicker representative seemed to be a little miffed about my protests. He wasn’t particularly nasty about it, but certainly didn’t seem to understand why this was an issue at all. Looking at i>Clicker’s website, students are asked for their ID. It’s different asking a student to give them their ID (consent) and setting up access to everyone’s ID (no consent).

What makes me wonder is, how many other institutions even give this a thought? Surely we can’t be the first person to balk at the idea of handing over student data like this? Or maybe we are being too paranoid? I mean, I guess there’s people who have faith in the third-party vendors, but I’d prefer having a license, stating exactly what they are doing with data, how they’re using it, how long it’s retained and an agreement signed between the two parties. That way if the external party violates the agreement, the institution can hold them liable for the data breech – something a little stiffer than “oops”.

Fusion 2014 – Unconference and Day One Recap

Instead of a big post I’m going to break my experiences up into three distinct posts because a) it’ll get me to post more frequently, b) that’s something I want to do and c) no one wants to read a monolithic block of text.

So I flew out of Buffalo, and it was an interesting time crossing the border where I got the fifth degree about where I was going and what I was doing. I think they thought I was being paid to speak at a conference, next time I’ll have to change the language I use to say something like attending a conference. After the border and the pornoscanners at the airport. I arrive in Nashville. Now, I’m not that worldly, but I’ve been to a few places. Nashville is not one of my favourites, not because the city is particularly terrible, it’s not particularly walkable, and it has well, public transportation issues. Outside of those quibbles (which are big problems for me) it’s a fine city with some fine people.

The Unconference

One of the best things that happens at Fusion for the last 5 or 6 years is the Unconference. I missed the first few because I was never able to actually get to Fusion, but the last couple of years I’ve been able to go, this was the event kick-off that was fun, social, and often leads to previously undiscovered ideas and new ways to break D2L. I didn’t stick around for the full discussion because I was a bit tired, but the one thing that I did learn was that VHS (Virtual High School) use Javascript to develop interactive elements of courses. Now that’s not a particularly shocking example, but combine that with the Valence API and maybe you could do some in situ testing and push results to the gradebook. Later a few of us went out for a nightcap and a good time was had by all.

Fusion Day One

Typically the first day has a ton of beginners and introduction sessions in the morning, so I ended up meeting with my co-presenter to go over our session the next day. The sessions I did attend were incredibly useful for me and I learned a ton about how other places develop in-house solutions. In fact most of my attendance was in sessions that were around External Learning Tools or the Valence API.

Keynote

So John Baker’s keynote threw the audience a little, the big takeaway was that Desire2Learn is now D2L, and the Learning Environment, or the LMS, is now called Brightside Brightspace. I guess there’s a thinly veiled jab at the competitors being the dark side, but I can’t say that I understand the need to change names. Lots of people at the conference have suggested that Desire2Learn seems a very 1990’s thing, reminiscent of the dot com boom/bust. I can’t say that they’re wrong. However, it would’ve been nice to have been told that officially. I’m a bit of a smartass when it comes to names, so my immediate nature is to shorten this to it’s logical shortform, BS. Not necessarily flattering. I don’t think D2L is big enough to have gotten out in front of it to shorten it to B, which in and of itself is not a good acronym either (B product? B movie?).

I’m not the only person who’s looking for a short form for it either. Considering I don’t know the difference between Brightspace and the LE (so is the new version called Brightspace version 1?) or if the existing products are called the LE 10.3 still… so many questions. None of them answered.

I’m sure many will talk about Chris Hadfield’s inspirational speech, it was great and all, and I certainly appreciate what he’s done. I just don’t see the connection to the conference that he brings.

Integrating Neat Tools and Activities into your Course through LTI

This session was all about External Learning Tools – which we’ve had a summer of dealing with so far. This particular session talked about integrations between SoftChalk, SWoRD and TitanPad. I’m familiar with SoftChalk through a series of courses I’m taking at Brock University and I can say that I’ve never been particularly impressed with the product – perhaps that’s the way that Brock is using it, or the way the course was developed, or a limitation of Sakai, Brock’s LMS. Either way, this session demonstrated the connection between SoftChalk activities hosted in Content then connecting to add grades into the Gradebook – certainly a more interesting way to deal with whatever you design in SoftChalk.

SWoRD was a particularly an interesting case – although I don’ t know how robust or deep the integration was (I suspect D2Lwas merely passing enrollment data to SWoRD). SWoRD is a peer assessment tool that might be an alternative to something like PeerScholar.

I’m always happy to see Etherpad clones, and TitanPad was used as an example, but if you’ve hosted an Etherpad clone at your institution you can pass user names to the Etherpad for auto tracking in the document. I’m not sure how robust Etherpad is for say, classes of 600+, but that would be an interesting experiment.

One thing the D2L presenter said was that in the configuration of the external tools, when you check the option to send User ID, it means sending the anonymized version of the username, which is interesting because the language used in the external tools dialog would benefit from adding this tidbit – we’ve turned it off in most cases (and seem to have no issue with students/instructors logging into the external tool) because we thought it would violate our University privacy rules.

The Secret to APIs

The second session of the day for me was around the use of Valence (D2L’s API) to create personal discussions in a course with enrollment of one student and the instructor. The big takeaway for this was that in courses that have enrollments set, you can save a ton of time by writing a script to do the repetitive boring stuff like create a group of one, enroll a student in it, then create a discussion topic and restrict it to that group. Was interesting to see C# used as the middleware programming language – I thought that C# was out of favor but maybe not? PHP would’ve been easier, and PERL/Python might’ve been faster to complete the task. Either way, this is the work that earned Ryan Mistura the Desire2Excel award in the student category. Cool stuff

Solution Spotlight/D2L Year Recap

Nick Oddson and Ken Chapman handled the recap of the D2L year, focusing on the extensibility of the platform. They did point out that there is a 40% faster time to resolution because they’ve increased their support and SAAS service teams. Which is good, because their service was slow before. I have noticed that their support turnaround is probably the best it’s been in years.

The looking forward part of their talk was interesting – it seems like they talked a lot about either 10.3 improvements (that were already announced last year, and available now), or stuff that we can’t see yet. Perhaps a chart:

10.3 Feature Unreleased to the Public
  • Wiggio
  • Discussions (Grid View restore)
  • Binder – Windows 8 support
  • Quizzing UI/UX improvements
  • Content Notifications
  • Student Success System
  • LEaP – Adaptive Path learning
  • Course Catalog (currently being used on Open Courses)
  • Visual Course Widgets (customization, I presume at a cost)
  • Built-in practice questions in Content (contextualized learning)
  • Gamification built into the Learning Environment (I assume 10.4/LSOne/Brightspace)

I suspect that the amount of talk about predictive modelling is something they want to build primarily for remedial use, and for online courses primarily. As a market strategy, that makes some sense. Some of D2L’s bigger clients are primarily online universities.

Blackboard Collaborate Integration with Desire2Learn, Uhh D2L, LE uhh Brightspace 10.3

I think I did that right?

Back in June we took a few weeks and integrated Blackboard Collaborate (our web conferencing tool) with our instance of the Learning Environment (Brightspace just doesn’t feel right). We are currently running 10.2 SP9 of the LE.

Reflections? Well, for such a simple integration (and really the D2L interface is waaaaay better than the Blackboard Collaborate interface) it took a hell of a long time. We had to purchase and get D2L to install the IPSCT pack – so if you’re entering into an agreement with D2L and may way to do this later, definitely spend the cash up front. From start to unveil it was over six weeks – now that’s not solid work on just this. After D2L installed the IPSCT pack, we had to contact Blackboard support to get our credentials. Seeing as we’ve had total turnover in who supports Blackboard Collaborate, our new Collaborate support person was not on the list of approved contacts – which is funny because she’s the one who does all the tickets. So we contact our account manager. No response. It turns out that well, they are no longer our account manager, that’s why we haven’t heard from them in over 9 months. Great. So support can’t do anything, neither can our phantom account manager. Finally we get to the bottom of who our new Blackboard account manager is, they straighten out the mess and our person is now an approved contact. After that it still takes a week to get our credentials for test and prod.

Configuration on test went smoothly enough – if you’ve ever worked with External Learning Tools in the LE, it’s the same as any other configuation in that tool – have the address to make the connection, secret key and password, check a few more boxes, and then off you go. Now everyone who gets enrolled in the LE gets a Default Role at the org level, and then gets assigned a more applicable role at the course offering level, which means for us, you have to go through not only the Instructor/Student and TA roles, but the Default Role as well. While this is a pain to do, it’s often easy to forget to do it – and that’s what we promptly did. A day or two was spent tearing what’s left of my hair out, until the lightning struck and it sparked the engine enough to get it firing again.

Fast forward a couple of weeks and we get some time to implement it on prod, we yet again forget what we did to make it work. A week later we said something to the effect of  “Fudge, Default Role…” ran off to the LE and fixed our error. Sometimes it’s not the technology that fails you…