There's been some effort to address this through communication (which I appreciate), but I think there was a lot of confusion on this because the Discussions Redesign was made available with a few key features not having parity with current Discussions. This feature is a pretty prime example.
I would just like to offer, for future reference, that based on my experience with this redesign, assignment enhancements, and New Quizzes/quizzes.next, it's problematic for the community to have a feature option that can be turned on in a live course when it lacks parity with core functions of the feature that it's replacing. Let me break this down with some considerations of the big 3:
- New Quizzes/quizzes.next: The list is really quite long of features that didn't work at some point in the process when campuses were being encouraged to turn this on. The biggest problem by far has been with banks and groups not being ready for import from Classic, but even something like matching questions not having feature parity for grading was a huge issue for many of my faculty until it was finally fixed.
- Assignment Enhancements: As things stand, if I were to choose to make this option available for my faculty it would create a totally different UI for students depending on the type of assignment. It's a great UI improvement, but we're not turning it on unless it includes group assignments and peer reviews because providing a consistent experience to our students is really important to us. With this one, I at least felt like there was very clear communication about why I wouldn't want to turn it on.
- Discussions Redesign: Obviously there's this thread. Beyond that and without getting into the "focused" 1 reply debacle, this feature redesign rolled out without the next/back buttons which is one of the most essential/core functions of Canvas. It's clearly deeply ingrained in the product design philosophy which puts modules at the center. Having those buttons is essential for the new feature to work within the way that every teacher/designer who is trying to create a clear/intelligible path through their course does design for their students. I will add that at least the messaging about this product was much more cautious that what I remember with quizzes.
I know in theWhat is the feature development process for Instructure products? post it states that, "For Canvas, all features are available for testing in the Beta environment unless otherwise noted. The beta environment allows users to explore new features without affecting their live data." but, there are a few problems here:
- Most users who might want to try features don't have access to beta instances.
- Features like the Discussions Redesign don't actually test well in beta where most of us have a very limited number of users with access.
- Most importantly: The way Feature Options works isn't limited to beta, so if you choose to enable a feature option, you're turning on something that is incomplete for potential use by your whole population, or at least a section of it.
I understand that Canvas is taking a very Agile approach to product development, but I find that the specific implementation of Agile feels like a mismatch with user needs. We're talking about the LMS with the largest share of the global market rather than the scrappy ed tech start-up company Instructure once was. In that context, users need stability in the product, and if there's a new version of a feature that can be turned on, it needs to at the very least have parity with the feature it's replacing. Otherwise, it's a really bad idea for us to turn it on because we can't reasonably help test it without essentially compromising the learning experience and outcomes for our learners. In some instances, those learners are paying a lot of money for the learning experience. In other instances, it's deeply tied to equity in the populations that our institutions are serving. In all instances, it touches on the users' time (admins, faculty, designers, and students) which is the most valuable resource any of us have.
I respect the intention behind how the Agile design process is implemented for the Canvas product, and please believe me when I say that I understand how SaaS is different from old school SaaP development, and why Agile is the right fit. I just think that given the scale of the product, it's absolutely essential that a Feature Option is at the very least in full parity with all of the features users expect in relation to the feature it's replacing.
Without ensuring that's the case, it feels like the product is being rushed out in an incomplete state, and this leads to a whole lot of unnecessary user frustration. Speaking personally, I'd be much happier seeing videos released of incomplete features in development and literally only having those feature available in beta and being tested with groups that aren't live classrooms, than I am with the current approach. Then once those features have full parity with the feature they're replacing, roll them out at scale and we'll understand if there are still some bugs as long as they don't mess with students' submission process or grades.
Just to end on a positive note, I can point to the new RCE, Gradebook, and Analystics as great examples of this. By the time they were really being promoted, they were feature complete and had a lot of great improvements. There were a few fixes they still needed, and there are ways that I certainly hope they will both be extended, but they all stand as really successful examples of Instructure rolling out a product improvement to a core functionality without making an incomplete feature available that can lead to user frustration.
Thank you for coming to my TED Talk 😝