At our November meeting, we revisited the Anthology Ally tool now that it has been more than a year since its integration into Canvas at the University. Matt Farley gave a brief demo of the tool, highlighting the fact that it serves two main functions: an automated accessibility checker and a document conversion tool. The tool allows users to enable a dashboard in their Canvas courses, providing a bird’s eye view of potential accessibility issues. The dashboard provides an overall percentage score as well as a list of items that are easiest to fix or the lowest scoring. Matt also showed how you can check any page or file in the course to view the individual item’s score and any accessibility issues that were detected. He also demonstrated how students can used the tool to convert pages and files into alternate formats such as epub, Immersive Reader, and audio files.

Next, Matt discussed what the tool does and doesn’t check. He pointed out that Ally will only scan Canvas LMS items and uploaded course files. That means if you have content that is externally linked or embedded, it won’t factor into the accessibility score. He also demonstrated that there are many potential issues that an automated tool like Ally can’t catch, such as inadequate image descriptions, pages with a lack of headings, and bad link text. These reasons and others can lead to scores that are artificially high or low, meaning that while Ally is a useful tool, instructors and designers are still encouraged to manually check their content to ensure everything is accessible.

Members of the group also shared some of their experiences figuring out the quirks of Ally. Lee Ann Dickerson noted that it flagged the same image twice for lacking alt text because the report checks for alt text in two places. She also found that the accessibility checker in the RCE doesn’t always reflect changes made in the files page, and vice versa. In addition, Lee Ann noted that RGBA color code used in her course was flagged as a problem by Ally, which suggested using a white background instead. She discovered that the issue was with the tool’s inability to read opacities. 

Aimee Brown-Shadduck discussed the limitations of the Ally tool in checking the accessibility of PDFs and images. She noted that while the tool can identify color contrast issues within text, it may not always catch issues with image resolution or readability. Elizabeth Pyatt also shared her experience with a PDF that was marked as compliant by the Ally tool but contained embedded Thai characters, emphasizing the need for manual verification.

If you would like to learn more about the Ally tool, you can check out our November meeting recording and meeting slides. Our next meeting will be on December 12 at 10AM where we will be looking back at the past year and toward the future of the AUG. For those in the State College area, we will also be meeting for lunch at Olde New York. We hope to see you there!

Our October meeting featured a presentation by Aimee Brown-Shadduck, a member of the AUG Leadership Team and an instructional technology specialist for the College of IST. Aimee was joined by several of her colleagues including Amy Gabrick, Kent Matsueda, Trisha McKee, Aaron Yorchak, and Logan Harvey to discuss their plan to ensure IST courses are accessible in lieu of the upcoming update to Title II of the ADA.

Aimee spoke about her journey of feeling overwhelmed by all the work that needed to be done and getting lost in the details trying to figure out everything at once. Kent shared his perspective, acknowledging his initial lack of understanding of the scope of the task and his realization of the need for a collective effort to meet the deadline. In talking with Aimee, he realized that the 2 year deadline to become compliant with the law meant they needed to take action sooner than expected. He began drafting a plan with the help of ChatGPT and met with Aimee and Amy G to refine and finalize a draft. They emphasized that their plan is a living document which will be continually updated as they learn more and adjust their approach. Finally, they submitted their plan to Philip Voorhees for feedback, and he gave the team lots of praise for the plan they created.

Aimee then went through the plan itself, covering some of the initial areas like identifying top enrolling courses and hiring accessibility interns to review content. She also provided a demo of the resources they created in-house, which included the IST course triage, an accessibility audit sheet, and a checklist with resources. The audit works as a tracking sheet for each course, with columns for listing issues, providing links, selecting corrections, assigning tasks, and tracking status updates. The checklist includes resources and guidance to address these common problems, with various automations set up to assign different levels of priority depending on the corrections needed. Finally, she shared the paid tools her team uses, including ChatGPT with a custom GPT she built to describe images, and ElevenLabs to generate audio descriptions from text.

From questions received during the presentation, the biggest issues they are finding so far are inaccessible PowerPoints, link text, and images of text and code. A question about the cost of implementing the plan was asked, and her team will look at the numbers and get back to us.

Aimee shared their team’s IST Accessibility Planning files which includes the PowerPoint presentation and plan document. We also encourage you to check out the October meeting recording if you missed the meeting. Our November meeting will be focused on the Ally tool and how to better interpret your course’s accessibility score.

At our September meeting we were joined by Maggie Kwok, disability coordinator for Student Disability Services at World Campus. Maggie discussed a number of topics and answered questions related to different types of accommodations, including split exams. She pointed out that students often need to have their exam split into multiple parts due to a variety of disabilities that may result in them not being able to sit at a computer or focus for long periods of time. There are several ways to approach this accommodation depending on the nature of the exam and the student’s needs, and there is a documented process for splitting exams.

Maggie also addressed situations where a student “may need” a specific type of accommodation, such as live captioning. In these cases the student is approved for the accommodation, but they may not always want or need them for each class. Maggie said the best thing to do here is to reach out to the disability specialist for more information in order to confirm what steps, if any, need to be taken to ensure the course is accessible for the student.

In general Maggie emphasized that she and the rest of her office are here to help instructional designers and faculty successfully implement student accommodations, so to never hesitate to reach out and ask questions. If you have questions, you can reply all to the email notification that gets sent out for some types of accommodations, or email their office directly at sds@worldcampus.psu.edu

Speaking of which, she also addressed the discussion around the official letter of accommodation vs. the advanced email notifications. At World Campus, email notifications go out before the semester about certain types of accommodations such as split exams and captioning. That is because these can take time to set up, so the design teams and faculty are given advance notice. But the letter of accommodation is the official document that outlines a student’s accommodations, and this should be provided to the instructor by the student.

Another topic discussed was students in courses that use Honorlock for proctoring. Maggie mentioned that accommodations should be entered into the notes field, and that the burden should not be on the student to inform the proctor of their accommodations.

Lastly, Maggie shared some stats with the group:

• 706 students are receiving accommodations – 33 are provisional status; 11 temporary status
• 240 students are in “pending” status – they are either scheduled for an upcoming registration appointment with a Disability Specialist or their office is waiting for supporting documentation from the student

There was also a great discussion in the chat with several other questions asked and answered, so we invite you to view the meeting recording which also has the chat included as an attachment. Please stay tuned for news about our upcoming October meeting!

At this month’s meeting we were joined by Philip Voorhees, Manager of the IT Accessibility Team, who provided the group with an overview of the recent DOJ ruling around Title II of the Americans with Disabilities Act which specifies requirements for making web content and mobile applications accessible. Philip provided several key details about the ruling, which clarifies specific rules around how government and government-funded organizations (including universities like Penn State) need to provide accessible content. This Mid-Level DOJ resource goes over the timeline for the ruling, what it covers, the exceptions, and the impact it has on our work.

During the meeting, Philip also answered several questions from group members. The following is a condensed portion of the meeting transcript presented as a Q & A.

Q: What would happen if someone is linking out to 3rd party video? So they’re putting a link to a YouTube video in a course rather than embedding the video itself in the course. Is that a workaround, or would that video still need to be accurately captioned?

A: Anything we link to that is a part of our services, programs, or activities needs to be accessible.

Q: How does this apply to resident courses as opposed to a fully online course?

A: If there are digital materials used in resident courses, then those digital materials need to be accessible.

Q: If a video has no audio and it’s only animation, what would you recommend to the faculty?

A: In that instance, if the video has no audio, then it’s just a visual element. Depending on what the faculty wants out of that or what the student needs to get out of that visual element, that needs to be done in a descriptive fashion.

Q: Is there a distinction between required and optional content?

A: No

Q: Is there any consideration for work that we do for external organizations, such as grant work, research partnerships? Is there an exception for that type of work, or is it just a blanket statement that anything Penn State’s name is attached to needs to be accessible?

A: Anything that is part of our program, services, or activities needs to be accessible.

Q: Is President Bendapudi and the various colleges aware of this? What communications have gone out, and will there be any broader support or funding around this?

A: Bendapudi and everybody, I mean all of the executives know about this. I update the President’s council every week on this particular issue.

Q: Is there a tool that can help with checking accessibility in a content management system (CMS)?

A: We have yet to be able to get our crawler to deal with CMS materials as well as public websites. I’m hoping that that’s gonna happen sometime soon.

Q: Would a transcript work for a video with only instrumental background for emotional effect?

A: If the audio doesn’t mean anything, and it acts like a graphic that we just mark as null because it’s decorative, then a descriptive transcript would be enough.

Q: Will there be any broader support or funding?

A: The biggest funding component of this is dealing with captioning and audio description.

Q: Can an individual be held responsible if someone files a complaint against the University about inaccessible content?

A: Oh yeah, there’s a couple of different cases out there where faculty were directly held responsible.

Q: Who would be responsible for enforcing or training faculty about this information?

A: So enforcement is really an interesting word. You know it depends on how the executives want to manage that enforcement, so I am uncertain exactly how they’re going to want to manage that process. The reasonable and easy way to do it is, you know, guys, this is part of your contract in order to use or work at Penn State. You have to know your responsibility.

Q: Could you explain a little bit more about what the top 30 course project is?

A: Yeah, so basically, we identified what’s called top 30 that have identified as being the highest impact courses at the university. In other words, they have the highest attendance and have the highest percentage of risk as far as individual, because a lot of these courses are also required of individuals with disabilities attending them and the materials not being accessible. We have traditionally gone through the process of the 504 model of reactive accommodation of which that doesn’t meet accessibility requirements. In reality it should be the other way around.

For more information about the Top 30 course project, you can view this list of the top 30 courses. We also invite you to check out the August meeting recording for more of this great discussion. The meeting chat is available as an attachment on the Kaltura video.

As a reminder, a new group topic ideas document is available for group members to add their own ideas for potential meeting topics. Please stay tuned for news about our next meeting in September.

Our June 2024 meeting was a celebratory event in that it marked the 10th year anniversary of the Accessibility Users Group. Sonya began by detailing a brief history of the group’s origins, and revealed the original charge of the group which was the following:

“To bring together those working in online education at Penn State who are doing the hands-on accessibility consultation, remediation, and training for their unit in order to find solutions or share solutions to accessibility-related problems and to create some documentation around best practices so that we can benefit from each other’s knowledge and experience and possibly share that knowledge with the wider Penn State community and beyond. Doing this would save us all time by allowing us to benefit from solutions others have found and increase our effectiveness at finding good solutions by benefiting from the different strengths and perspectives of the group.”

In the beginning the group consisted of less than a dozen members, met face-to-face, and primarily used Yammer to communicate. Since then many things have changed, least of which is the change in group membership. There are now more than 70 members and we have expanded beyond Instructional Designers in online education to also include multimedia specialists, programmers, IT specialists, and people in other roles across the University. We also made the shift to meeting online before the pandemic, and over the years have created a number of resources like our website and Teams space.

One of the biggest developments to come out of the AUG was our working groups that focused on a number of different topics. Lots of great work was done in these groups, and many AUG members helped create the Canvas Styles resource.

While so much has happened in the past decade, with the recent DOJ ruling and increased awareness about accessibility the next 10 years of the AUG may be even more eventful. The majority of the meeting was spent by group members sharing their stories related to their history with AUG and accessibility in general, and what they envision the future landscape to look like, so we highly encourage you to watch the June 2024 meeting record. We’ve also included the meeting chat so you can read all of the great conversation that was happening there.

Loading...

Taking too long?

Reload document

| Open in new tab

Download

As a reminder, a new group topic ideas document is available for group members to add their own ideas for potential meeting topics.

The May 2024 meeting of the Accessibility Users Group featured a panel of group members discussing strategies and tips for utilizing artificial intelligence tools to describe images. Sonya Woods kicked off the meeting by providing a brief introduction to the concept of image description in general. She emphasized that providing text descriptions to visuals supports Universal Design for Learning considering that descriptions can benefit people in a wide variety of situations whether they can’t see or even just have a poor internet connection. She also provided an example description for a Venn Diagram, pointing out that AI gave her some useful ideas to enhance the description she had written on her own. General tips for image description include remembering context is key, consider your audience, and be concise.

The panel began with Wade Shumaker, an Instructional Production Specialist with Teaching and Learning with Technology. Wade demonstrated some of the ways he has used AI to aid in improving the accessibility of digital documents. Wade underlined a couple of key points that he’s learned from his experience using AI for image description like the fact that many AI tools can recognize and provide text found in images. He also shared his process for transforming an image of a table into an accessible HTML table that retained all of the text from the original.

Next up was Elizabeth Pyatt, an IT Accessibility Consultant with the IT Accessibility Team. Elizabeth highlighted some of the limitations with current AI tools, specifically that they are still not great at translating mathematical equations and recommended using a tool like EquatIO instead. She did show that AI can be effective for describing complex graphics, including organization charts, though stressed you should often consider the results a first draft that need to be reviewed and edited for accuracy. For maps and geological data AI was also able to identify and explain the information in the key.

The last panel member was Heidi Sporre, an Instructional Designer for the Dutton Institute. Heidi took a slightly different approach to AI and image description by using ChatGPT to train the AI on the type of information it should prioritize in a description by feeding it government accessibility guidelines. By giving ChatGPT specifications on the ways it should describe images, she was able to get better results than by simply prompting the tool to provide a description.  Heidi shared a document with the group that outlines the specifications she used to guide ChatGPT toward providing better descriptions, and you can access this and all of the resources shared by our panelists in our AI Tools for Image Description resources folder. Additionally, you can check out our AUG May 2024 meeting recording.

Other updates from this month’s meeting include:

• A new group topic ideas document is available for group members to add their own ideas for potential meeting topics
• Our June 2024 meeting will celebrate AUG’s 10 year anniversary. We invite you to share your stories about the group and accessibility at Penn State. Look forward to more details about an in-person get together after the Zoom meeting on June 13th from 10AM-11AM.

At our April 2024 meeting we were joined by Benjamin Hammel, an Instructional Technology Specialist at Penn State Lehigh Valley. Benjamin provided a demonstration of several techniques that can be used in Photoshop to enhance color contrast when working with charts in Photoshop. The primary purpose of Benjamin’s demo was to show how the color contrast can be improved for low contrast charts when you don’t have access to the data and can’t recreate it. If you work with images in Photoshop, we encourage you to view the April 2024 meeting recording to review the steps Ben takes to improve chart visibility.

For the remainder of the meeting, several group members shared some of their favorite tools that they use to check color contrast when reviewing or creating online content. Matt Farley began with a demo of the Color Contrast Analyzer, a standalone application that is available for Windows and MacOS. This tool can be used to identify the color contrast ratio for colors anywhere on your computer screen. It also gives you the ability to pick and choose new colors, and conveniently lets you know whether or not different color combinations meet various WCAG standards.

Next Aimee Brown-Shadduck showed off the WebAIM Color Contrast checker tool, specifically the bookmarklet. By adding this bookmarklet to your browser, you can use it to check the contrast of colors anywhere on the web. Kerri Weitzel introduced Adobe’s color contrast website, which includes several features like the ability to design a color theme, have colors with good contrasted recommended to you, and view color combinations that are color blind safe. Last but not least, Meltem DiLeo shared the Tanaguru Contrast Finder website, which allows you to check color combinations, set your preferred minimum ratio, and be provided a range of accessible colors.

While wrapping up Sonya Woods provided a brief update on the DOJ ruling regarding accessible web content and mobile apps. This expansion of the ADA will go into effect on June 24, 2024. You can read the full DOJ ruling for more details. Our next meeting will take place on May 23 from 10-11 EST and will feature a panel discussion regarding the use of AI to assist with image description.

For our first meeting of 2024 we were joined by Danielle Steinhart, a Copyright Specialist from the Office of Scholarly Communications and Copyright at Penn State. To begin with she outlined the definition and aims of copyright protections, specifically related to copyright law in the United States. She also covered some distinctions between what type of material can be copyrighted and what can’t, such as a generic list of restaurants versus a blog post covering an individual’s favorite restaurants. Copyrighted material is also not protected forever, as works before certain years (currently 1925) have entered the public domain.

When it comes to the intersection between copyright and accessibility, Danielle made the point to say making sure something is accessible can sometimes trump certain copyright considerations, but accessibility does not give you an excuse to ignore copyright law. For example, if you don’t already have permission to use copyrighted material in a course, such as a document or a video, making the content accessible such as remediating the document or adding captions to the video does not give you permission in and of itself to use it. However, if you already have permission to use certain materials, altering them in ways to make them more accessible may be considered fair use. But she could not give legal advice and emphasized that every situation may be different and should be evaluated on a case by case basis.

After the presentation, there were several great questions and discussions had between Danielle and group members asking about different situations where copyright law might clash with accessibility. If you would like to learn more about how you can ensure you are offering accessible materials while following copyright law, we encourage you to view the February 2024 meeting recording, and you can reach out to the Office of Scholarly Communications and Copyright if you have any questions.

For more information you can also check out Danielle’s slides.

Our December meeting was a look back at both the history of the group and the last year of meetings, as well as a brainstorming session to discuss which types of topics we’d like the group to cover in the future. First Sonya Woods shared some statistics and fun facts about the group, including:

• The group began in 2014 with 11 members and now has more than 70
• The first topic the group covered was about PDFs, and PDFs was still our most popular meeting topic in 2023
• We’ve had several working groups that have created useful resources for the university, including Canvas Styles and Split Exams

In 2023 some highlights include the introduction of the Ally tool, conference share-outs from Axe-Con and CSUN, updates about the captioning policy, designing accessible videos, and PDF accessibility.

For the second half of the meeting we discussed topics that group members would like to see covered in 2024 and beyond. Some of these topics included the following:

• Automated document accessibility tools
• New Quizzes in Canvas
• Concerns around copyright and captioning
• Accessible practice for people with cognitive and learning disabilities
• Accessibility at the library

These are just a handful of several ideas people thought of, so to see more and add your own please visit our AUG 2024 Topic Brainstorming document.

And if you couldn’t make it to the meeting we invite you to check out our AUG Year in Review slides and the December 2023 meeting recording. Thank you for a great year and we look forward to seeing you in 2024. Happy holidays!

At our October 2023 we were joined by Elizabeth Pyatt, Jennifer Wagner, and Angel Peterson to talk about different aspects of PDF accessibility.

First they reintroduced the currently available courses in the Accessibility Learning Path and previewed an upcoming course focused on PDFs. Jen began by discussing some of the differences between saving a document as a PDF and a scanned PDF that is an image, and the various accessibility challenges that come along with each. She pointed out that if you have the original document you will be able to make improvements at the source before turning it into a PDF, versus if you have the PDF only you can improve it by doing things like tagging, ensuring it goes through the Optical Character Recognition (OCR) process, and checking the read order.

Jen discussed some overall strategies for ensuring newly created PDFs are accessible, such as creating a style guide so everyone involved in the creation of documents will use the same styles for headings, tables, and others elements. Also establishing a strategy for describing images is crucial, especially if multiple people are involved in the creation and remediation of a PDF, so that the final experience is consistent for the reader.

Angel brought up copyright challenges with certain publishers where legally they are not allowed to change colors or add descriptions within the text itself, basically anything that would change the viewable content in the document. In some cases she works with the authors to design an accessible template, or ask for the PDF to be provided in another format such as HTML. Jen added that sometimes specialized content such as Math will ideally need to be rendered in a format like Word.

Angel also discussed some other topics covered in the upcoming ALP course on PDFs, including how to check and add tags. Specifically she talked about the difference between manual and auto-tagging, how to add alt text to images, and explaining what a typical tag structure looks like and how to navigate it. She also showed how to use the Reading Order pane to make sure a screen reader will read all of the elements in a PDF in the correct order.

Elizabeth went over some services and tools for OCRing documents, including Sensus Access, Read & Write Gold, and Anthology Ally, all tools which are currently available at Penn State. She also pointed out that in some cases if you have a scanned document you can search and find the original source on the web in an already accessible format with real text.

For more information about PDF accessibility, we encourage you to watch the October 2023 meeting recording and check out Elizabeth, Jen, and Angel’s slide deck below.