Community

@TiffanyStullI absolutely agree with you on this. I do consider the presentation of GIFs to be part of Canvas since GIFs themselves are a file format rather than executable code, but I would still place proper accessible support for something that is an optional content type lower down the list of priorities than contrast in web app (or mobile app) UI. I'd like to see all of these issues addressed, but there definitely is a reasonable triage process that would incorporate that sort of prioritization. ... View more

@TiffanyStullSo this is a really good point about how Instructure has been slow to fix some of the truly simple things, but I would also say that from a web development standpoint the complexity of this request depends on the implementation. One possible solution that wouldn't guarantee accessibility but would at least be a move in the right direction would be to inspect GIFs, check the loop count, and at least let the user know if it is set to 0 (endless), and that this is an accessibility issue. Obviously Instructure is a huge company and a quick web search will show you that there are people out here who have developed workable code to handle this problem as well. I totally get where you're coming from, but it's really unfortunate that our collective response to "this slightly more complex accessibility issue needs to be fixed" is generally something like "they aren't fixing more basic accessibility issues so i guess we need to lower our standards." Edit: as far as feasibility of play/pause, this exists in MS Teams. When I drag an infinitely looping animated GIF into a Teams chat, it automatically has a play/pause button added and plays on click. I don't think there is a good excuse for Instructure not implementing features that exist on other major platforms and meet accessibility requirement. From their quarterly report full year 2022 revenue: Revenue is expected to be in the range of $471.2 million to $472.2 million ... View more

Much as the workarounds presented here are nice, they require faculty to be versed in enough HTML to implement them, and even then they compromise the desired experience of adding an animated GIF to a page. This is an accessibility issue, and I would like to see Instructure prioritize updating their code so that when GIFs are added using the RCE, they will play once and on click. Otherwise, Canvas is not WCAG 2.1 compliant, and any institution using it would need to carefully monitor every course and remediate all animated GIFs with one of these workarounds or remove them. I'm not sure if this particular issue would be a 508 compliance issue, but this is an issue with a straight forward technical fix that institutions cannot implement themselves and should be part of how Canvas works "out of the box." On a related note, Canvas images should also support the figure and figcaption tags so that accessible captions can be added in the image options. Without this support, the vast majority of images will be captioned incorrectly because very few users know or will take the time to go into the HTML to add their captions. ... View more

If that's the case, it's a little weird that new members can post to a feature user group but can't post an idea. ... View more

I believe it's now 7 months since the last update to the redesign. We could really use an update on when we can expect a new feature/fix push on this project. ... View more

This is another major issue I'm hoping to see addressed in the redesign. At the very least it would be nice if it could preview Google Drive or Microsoft OneDrive images when they're added to a discussion so institutions with those services could at least offer students a work around that doesn't consume their file storage when they're submitting required media for a discussion assignment. ... View more

@richard-jerzI would still technically classify this tool as a forum because you post topics to it rather creating channels for discussions, but I agree that the fact that this tool doesn't allow for threading is quite unfortunate. I've been off of Facebook for quite a while, but I'm pretty sure it does actually allow for threading. Social media tools are arguably more like micro blogs than they are online forums, and although I was on Twitter for years, in some ways I'm still not sure how to best characterize what exactly the platform is. The fact that posts have distinct URLs makes it decidedly not an IRC, but it's also definitely not a discussion forum. I decided that it was a toxic mess back in late 2018/early 2019 so I deleted my account. Piazza is a super powerful discussion forum tool, but it isn't intuitive for students and the learning curve it has means that it's best reserved for courses where the unique features it has (e.g. crafting a collaborative response, integrated surveys) are actually essential for instruction. ... View more

Welcome @JDPearson! In addition to my role as my college's Distance Education Coordinator, I've also been working with the California Virtual Campus's Common Course Management System Advisory Committee. That group works to represent the interests of all 116 California Community Colleges in relation to Canvas, and we have a discussions workgroup. Many of our members are off for the summer now, but I sincerely hope that we will collectively have an opportunity to meet with you sometime late this year. I'm also more than happy to meet with you sometime this summer as I'm in a 12 month position. As to some of the discussion in this thread, I feel the need to point out two things: I see the comparison to Slack and I need to state that this is the problem for most of us using discussions. Slack is essentially a modernized IRC. An IRC is not a discussion forum, it's a chat platform. Trying to make discussion forums have an IRC like interface is fundamentally misguided. IRCs rely on channels rather than forums. You can close an IRC channel, but they're intended to be open ended communication tools. They're great for project groups and live communities to discuss projects or activities. They're not good for medium to long form posts which is the primary use for discussion forums in educational settings. If we want a consumer or productivity product that's an analog for LMS discussion forums, Slack is not the place to draw on as an example. Reddit or Quora both present better examples on the consumer side as analogs, and I have trouble coming up with an analog in the productivity software space because most organizations lean on things like email for long form communication. This is pretty trivial, but the term redesign isn't really problematic here from my perspective. Even if 100% of the code for the discussion forums were new (which, it's now being implemented as an LTI so I'm sure a lot of it is), we're talking about a feature set that's part of a larger platform. This isn't a new version of Canvas, it's just a rework of a part of Canvas. Best, -Moses ... View more

@AllisonPittler You've asked for specific examples of where students are struggling with the redesign, but I want to emphasize that at my campus I have told faculty not to turn on the redesign because the UX issues with the slide out tray are so severe that it doesn't take forcing our students and faculty to struggle with it for me to know that they'll hate it, and that it will quite probably negatively impact our outcomes. I can't emphasize this enough. Before I turn anything on in Canvas at El Camino, it is my duty as Distance Education Faculty Coordinator to make sure that it's not likely to make students or faculty struggle, and I have enough UX know how to take one look at that tray and know that it would present a real risk to student success in some courses, not to mention create a huge headache for faculty that will slow down their workflow and diminish the quality of classroom discourse. The bottom line on the workflow issues with the tray are really straighforward. A discussion in a course is in most cases, implemented to create back and forth communication between and among the students and faculty. There are other possible uses, but that's the big one. Hence, any changes in design need to support the UX patterns that we already know define the positive parts of discussion forum interaction for learning not just in LMS's, but in any other place where a forum style of communication is used and learning takes place. The slide out tray unfortunately violates user expectations and disrupts the patterns that effectively support teaching and learning in an online space. Say I'm a student being asked to post once and reply twice. I'm not the first student to post and some responses already have replies. With the current UI, I only see top level comments and an indication that there are replies. If this discussion is working the way most faculty want it to work, there might be two or three replies to a post, and then at least one level down from that more responses if the students are motivated and engaged with the topic. As a student, I need to see everyone's responses, and the responses to those responses before I post my own. I need to not have to store that information in my head as I click down into each response and sub response. If I'm forced to do that, I've expended a huge amount of cognitive effort managing that task, and that means  I have less cognitive load for actually framing my response and thinking about the subject matter of the course. It's part of the most fundamental principle of UX for education software: Your interface needs to require the bare minimum of cognitive labor because you need as much cognitive capacity as possible available for learning the subject or topic that the class is about. Then there's the role of the instructor. If I've had a successful set of interactions between my students, I'm going in to do two things: Reply including possibly providing formative feedback, summing up the conversation, advancing the conversation further, etc. Grade This UI is terrible for both of those things. I can't see the whole flow of the conversation at once. I can't search the page effectively. I can't tell whether there are common trends going on in my student responses without taking very detailed notes. If I see something that reminds me of a student response to another student, I need to track down the initial post and then click into the responses until I find the post I was thinking of. I can't even tell at a glance if some students are dominating the conversation in a way that's destructive to classroom discourse. tl;dr The slide out tray completely misses the most fundamental practices that define a useful online discussion forum in a course because it chews up cognitive load in managing the interface taking away from the users capacity as teacher or student to engage in the topic. ... View more

Change is not universally good, and in this case we have virtually every community member reacting negatively to the primary change in presentation not because it's different, but because it's a genuinely bad user experience. The slide out tray arguably violates heuristics 2, 4, 6, 7, and 8 of the Nielsen Norman group's widely accepted 10 usability design heuristics. That's half of the heuristics that UX designers employ to make sure that interface design is likely to be good before you test it, let alone role it out for general availability. The tray design is not a good case for giving change a chance. There are plenty of other aspects of  the redesign that are, but this one is just bad design that doesn't fit user expectations or needs. ... View more

For this feature to work in a way that meets user expectations of how @mentioning works on other platforms, it needs to generate a notification including an email notification with a deep link that takes you directly to the part of the discussion where the mention took place. Ideally, users need to be able to respond to the @ mention from their email because again, that's the expectation based on other online platforms. If there is no deep linking to the part of the discussion where the mention happened, then the feature name is misleading in relation to user expectations. This is also currently the case with quoting since it doesn't work in any way like quoting does on other platforms. ... View more

This post appears to have found the wrong topic. This is the discussions redesign. I'm guessing you were looking for New Quizzes to post in? ... View more

I hope this gets seen and gets a response. This is another feature parity issue. Even though autosave is a newer feature, discussions is the biggest part of online courses where having an autosave feature matters. I have spent years telling faculty and students to compose responses outside of the LMS. With the advent of autosave in Canvas, I was finally able to not push this approach and let them engage with the platform in the way that's natural for the users. Given user expectations for autosave to work in Canvas, it's absolutely essential that this feature is supported before I tell anyone on my campus to use the redesign. Needless to say here (but I'll say it for emphasis), they will also need to have a view that isn't the focused/slide out view before I will be telling any of my faculty to use this version as well. ... View more

I'm not exactly sure how Instructure handles the Discovery phase, but Discovery is typically a pre-Design phase and Design should be a pre Development phase. What does that translate to? Discovery is the phase in which you do pre-design research, scoping, and planning. This typically includes defining the user group, collecting available quantitative data about the users, doing qualitative research (e.g. interviews, focus groups), and ideally doing both mock-up and prototype testing. Then you can define the scope and develop a product roadmap. In Agile development, you aren't locked into the results of your Discovery phase, because in principle Design involves more user testing, and then ideally you do distinct alpha (incomplete product) and beta (minimum viable product) testing before rolling out the software, and of course you'll continue to not just support but enhance and/or modify the product which is why the term is SaaS (Software as a Service) rather than SaaP (Software as a Product). The part about what constitutes a minimum viable product can be tricky. This is part of why I posted about the importance of feature parity in this thread. From my perspective, even a product in beta that's a replacement or improvement needs to have feature parity with all of the features you're planning on keeping. If it doesn't your users are going to be confused and angry that the product is being made available for use in their actual work. It's also worth noting that although beta testing can be public or private, it's really not supposed to be the same thing as going live in the production environment (e.g. where the users are doing there real work). In the context of things like productivity software and educational software, the stakes are way too high to roll out applications that are incomplete or aren't well tested for public use. In entertainment software your product can fail and you can get a lot of angry reviews, but in software people use to work and learn, the consequences of the software not doing what it needs to do for the users is a lot higher. Sure, it's not as bad as the potential life and death consequences of incomplete or buggy software in the healthcare or defense sectors, but it's still going to have consequences that are much greater than just angry users. ... View more

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 😝 ... View more

@aring  This has been fixed. It was included in the August 21st 2021 release change log. It was included under feature parity, but it would have been nice to have it recognized as a bug fix since none of us knew it wasn't working when the redesign was first rolled out, the checkbox was still there but it didn't actually work, and it was discussed as being broken in this thread. From a transparency angle, I feel like having a table like New Quizzes has showing the current state of features in comparison with existing Discussions would be really helpful. I had to do some digging to come up with the answer to this one. ... View more

I could see the rationale for both truly anonymous discussions and discussions that are anonymous only on the discussion board. The latter would allow students anonymity in interacting with in the course which could de-risk interaction, but could still allow a faculty member to use those interactions for grading or attendance. Ideally it would be great if there was support for both types of anonymity, but at the very least I agree that we need clarity on how the feature is supposed to work. ... View more