Meeting minutes
Collaboration Tools Accessibility User Requirements.
Janina: has started working on revisions, but they aren't yet ready for review. Janina has found additional collaborative environments to be used as examples.
Janina: Not all collaborative environments are text editing applications.
Janina needs to find a few IDEs.
Janina asks whether COGA have been informed of our problem - that the document is concerned with the collaborative aspect of any given task. Janina planned to raise this with COGA yesterday.
Janina plans to foreshadow it before it arises in a COGA TF meeting.
Janina: this wasn't possible in yesterday's COGA TF planning meeting.
Janina expects it won't take inordinately long to update the document with examples and clarification.
subtopic+ Issues 33 & 34
<DavidSwallow> User Need X: As a user who finds some web sites hard to use and struggles with remembering and following instructions, I sometimes need in-page instructions so that I can do the correct task.
<DavidSwallow> REQ X: Instructions from the author are easy to find. For example, an icon and link to a readme document is always available on the main toolbar.
Janina: it's the usual issue of being out-of-scope - not specific to collaborative environments.
subtopic+ Issues 12, 13, 14, & 15
<DavidSwallow> User Need X: As a user with a cognitive or learning disability and who likes to browse on the Web, I need the structure to make sense to me, so that I find what I am looking for, without looking in the wrong place.
<DavidSwallow> REQ Xa: Allow the user to determine the storage location of the collaboration space.
<DavidSwallow> REQ Xb: Allow search through a page, whole document, a group of documents (such as a project or directory).
<DavidSwallow> REQ Xc: Allow the user to structure the collaboration space.
Janina: whether you can structure collaboration spaces depends on the capabilities of the tool (e.g., whether hierarchies of content are supported - e.g., directories/subdirectory hierarchy).
Janina: It isn't clear there's a requirement specific to collaboration.
subtopic+ Issue 37, 38, 39, & 40
<DavidSwallow> User Need: People with cognitive disabilities need to be able to follow comments and replies without becoming overwhelmed with many levels or needing to navigate between the levels can be difficult
Janina: notes this is addressed in the text.
Janina: strategies include being able to focus on specific comments without interruption by other comments.
<DavidSwallow> REQ Xa: The structure of comments and conversations are visually Clear. For example, by using icons and different colors to emphasize the different levels of comments
Janina: any specific strategies should be identified that we don't already include.
<DavidSwallow> REQ Xb: Multi levels can easily be expanded or closed.
Janina notes we can act on the color/icon suggestion.
david: e.g. color to distinguish
david: expandable/collapsable
<DavidSwallow> REQ Xc: The current draft of wording should be easy to see at the top of a thread or in the text.
jasonjgw: unclear what level means in this context
Janina suggests we find the existing discussion and ensure the appropriate strategies are addressed as described in the issues. We should be cautious not to require color alone, however, as WCAG guidelines have long required.
Janina: we should include the strategies as examples.
jasonjgw: Will look
jasonjgw: We'll make sure we have appropriate examples before closing as fixed
subtopic+ Issues 16 & 17
DavidSwallow: End of Sec 5.2
<DavidSwallow> User Need X: As a slow reader I need short summary for long pieces of content or an option for an easy to understand language version.
<DavidSwallow> REQ X: Provide summaries of changes. Examples: list of comments and revisions they made (for use in meetings, for example). Also, items marked resolved: these can be difficult to track, understand how/why they were resolved.
Janina: asks how this could apply to audio editing, visual/graphical editing (e.g., charts).
<DavidSwallow> Previous explanation of Issue 16 from GitHub: w3c/
<gb> Issue 16 Coga Suggested User Need As a slow reader I need short summary for long pieces of content or an option for an easy to understand language version. (by jasonjgw) [COGA]
jasonjgw: A commit log might be useful and could drive an AI summary
Janina notes this is an authoring requirement that may be satisfiable by an AI intervention, in which case the summaries are likely to be available more often. Human authors are seldom likely to write summaries of changes in many scenarios.
jasonjgw: Ability to group changes and note that AI to describe would be useful
raja: Was discussing descriptions yesterday
raja: Noted it's subject specific, must know what you're asking for
raja: Describing noise, or sonic event may be interesting in one way to some people, but differently to others
jasonjgw: Picks up wonderfully on my point re AI to summarize -- seems similar.
jasonjgw: Good summary needs to know context and what is of interest to collaborative authors
jasonjgw: Not optimistic that AI is up to this in many cases
raja: agree
jasonjgw: Perhaps a mechanism present to add summary
Janina: refers to the suggestion to add a summary. Janina notes the HTML table (now deprecated) summary attribute.
jasonjgw: Different from Details/Summary -- a note on some span of changes
jasonjgw: Similar to what's available in revision control
jasonjgw: Would be in scope
Janina: the ability to identify a span of revisions and create a summary of them would be useful.
Janina: if there are accessibility-related reasons for providing the summaries, they are more likely to be provided in the social context of the collaboration.
Janina notes that summarizing changes is a skill in itself that not everyone has.
Janina: the proposal under discussion is to ensure a mechanism is present for identifying the span (of revisions) from here to there, and then the author(s) can include the summary. Users and authors are the same group of people for purposes of the collaboration.
Janina: if there are accessibility requirements related to summaries, the collaborators are more likely to provide th summaries.
Janina: refers to discussion of the idea that descriptions of images or auditory content may need to be different for different users according to their disability-related needs. This is different from providing a description of, e.g., visual content.
Janina: refers to prposals to create a list (taxonomy?) of description types.
jasonjgw: Notes state of the art in AI supplied descriptions is an interrogative process where the generated description can be refined
jasonjgw: Because large language models have general info, that info can be applied about an object; e.g. a well known person or location, that info can be sourced
Janina: this returns us to the issue of summaries, and AI could be engaged through a dialogue.
Janina notes the benefits of refining one's prompts to a large language model to obtain satisfactory results. Even then, human intervention in the output is often desirable.
Janina: we can note the value of AI/ML in automatically generated summaries - for purposes of this document.
Janina: notes the varying capabilities of human authors in summarizing changes.