Definition of "set of Web pages / set of software programs" in mobile context

[JJdeGroot]

We need to figure out how we deal with the definition of set of Web pages and set of software programs in the context of mobile apps.

Referenced in:

• 2.4.1 Bypass Blocks
• 2.4.5 Multiple Ways
• 3.2.3 Consistent Navigation
• 3.2.4 Consistent Identification
• 3.2.6 Consistent Help

Also related to our discussion about 2.4.2 Page Titled

[AlainVagner]

This issue is quite similar to these issues from the WCAG2ICT project: w3c/wcag2ict#261 and w3c/wcag2ict#308
This point has been discussed recently (2023) in WCAG2ICT and a decision has been taken not to adapt the definition of "set of software programs".
We should probably stay consistent with WCAG2ICT and consider that these SC are not applicable to a set of screens in a mobile app, or do you see any possibility to adopt another approach here?

[AlainVagner]

We have discussed in our latest meeting to investigate the possibility of a definition of a set of screens.
I propose to start with a naive approach where we slightly adapt the definitions of set of web pages and set of software products and see what fits and what needs to be refined. We could iterate on these definitions.

From the definition of a set of web pages

I just replaced here "Web page" or "page" by "screen" in the definition of "set of web pages".

collection of screens that share a common purpose and that are created by the same author, group or organization

Examples include:

• a publication which is split across multiple screens, where each screen contains one chapter or other significant section of the work. The publication is logically a single contiguous unit, and contains navigation features that enable access to the full set of screens.
• an e-commerce app shows products in a set of screens that all share the same navigation and identification. However, when progressing to the checkout process, the template changes; the navigation and other elements are removed, so the screens in that process are functionally and visually different. The checkout screens are not part of the set of product screens.
• [3rd example deleted, because it uses concepts such as subdomains, which do not apply to apps]

Note
Different language versions would be considered different sets of screens.

Questions and remarks:

• Should we mention that the screens should be part of the same app?
• the 1st example seems to apply, even if such apps are not so common, the 3rd example does not seem to apply, has no equivalent in the app world.
• should we have an example of other common set of screens, for example in which we can navigate thanks to a tab bar?
• Regarding the note: apps usually follow the language of the operating system. Apps being able to show different language version of screens without changing the main language of the OS are not so frequent, AFAIK. Should we keep this note?

From the definition of a set of software programs (WCAG2ICT)

"software programs" or "programs" become "screens", "distributed together" becomes "part of the same app"

set of screens: collection of screens that share a common purpose, are created by the same author, group or organization and that are part of the same app and can be launched and used independently from each other, but are interlinked each with every other one such that users can navigate from one screen to another via a consistent method that appears in each member of the set

Note 1
Although "sets of web pages" occur frequently, "sets of screens" appear to be extremely rare.

Note 2
Redistributing or bundling previously distributed screens as a collection does not constitute a set of screens.

Note 3
Consistent does not mean identical. For example, if a list of choices is provided it might not include the name of the current screen.

Note 4
If a member of the set is separated from the set, it is no longer part of a set, and would be evaluated as any other individual screen.

Note 5
Any screen that is not part of a set, per this definition, would automatically satisfy any success criterion that is specified to apply to “sets of” screens (as is true for any success criterion that is scoped to only apply to some other type of content).

Note 6
If there is any ambiguity whether the group is a set, then the group is not a set.

Note 7
If there is no independent method to launch the screens (as is common in closed products), those screens would not meet the definition of a "set of screens".

Note 8
[vocabulary, does not apply here]

Example: One example of a set of screens would be a group of screens that can be launched and used separately but are part of the same app and all have a menu that allows users to launch, or switch to, each of the other screens in the group.

Counterexamples: [do not apply here]

Questions and remarks:

• here the main blocking point is the possibility to launch the screens independently. This is clearly not feasible with a set of screens in a mobile app. This point is mentioned in the definition itself, in the note 7 and in the example, which mentions the concept of closed products. A closed product software is defined in WCAG2ICT by a product where there is no Assistive Technology to communicate programmatic information. In the case of a mobile app, there is clearly an accessibility tree, and the possibility to use assistive technologies with them. Should be remove this condition about being able to launch the screens independently? Should we remove the mention of closed products?
• notes 1, 2 does not seem to apply to set of screens

[JJdeGroot]

Discussion during MATF meeting on June 5, 2024

Source

JJ shared the issue #11. dealing with the definitions of 4 SC, 2.4.1, 2.4.5, 3.2.3 and 3.2.4

Jamieshared more light on Sets of software programs. #308

Jamie set of webpages refers to website as a concept. while entire app is not a set of screens, but actions are performed across the screens.

Jamie set of web page aligns with the concept of native apps

julianmka agrees with Jamie

JJ agrees that set of pages maps to native pages

AlainVagner what is screen in the mobile apps and how do we deal with the side effects of adopting this model

Jamie MATF and Web accessibility relationship?

Jamie if we are a subset, it can be more clear how the main document influences our document.

JJ explained we are at the same level as to WCAG2ICT

JJ we need to maintain the guidance consistent.

JJ we can discuss the same with WCAG2ICT participants and they can take into consideration

JJ will be attending in person meeting in September and discuss collaboration with other workstreams

Jamie The older questions from the WCAG2ICT pertaining to mobile should be added to out github

Jamie Maybe reopen the relevant issues

JJ needs investigation to understand how the issues with closed and then reconsider the solutions

JJ to seek the possibility of group meetings to collaborate

AlainVagner has never seen set of software programs in the mobile world. This issue needs to be addressed

@JJ shared notes where it is mentioned that set of software appears very rarely

Gleindson in brazil there are two different apps, reporting to same bank, and communicating with each other

Jamie The set of web pages definition in WCAG2ICT seems like an almost 1:1 alignment with screens in native apps in Zoom chat

[JJdeGroot]

Related issues: w3c/wcag2ict#308 and w3c/wcag2ict#261 and w3c/wcag2ict#225 (comment)

[JJdeGroot]

Judging from the related issues for WCAG2ICT, the main issue is that they did not succeed in defining view/screen.

Only because we could not find an objective - non-slippery or foggy definition of "views" which would be more equivalent to pages

Couldn't figure out how to define a VIEW in any stable fashion, if you can figure that out — there are a lot of people on your side

WCAG 3 intends to use views instead of web pages, definition:

Views include all content visually and programmatically available without a substantive change. Conceptually, views correspond to the definition of a web page as used in WCAG 2, but are not restricted to content meeting that definition. For example, a view could be considered a “screen” in a mobile app or a layer of web content – such as a modal.

Perhaps we can build on this for our definition in the context of mobile apps.

[julianmka]

Would aligning with WCAG 3's definition of view create an awkward dependency if it changes as WCAG 3 continues to develop?

[AlainVagner]

To my mind, the definition of view is technology agnostic. They mention "screens" for mobile apps. We could have a similar definition but without the mention of web pages.
For example:

Screens include all content visually and programmatically available without a substantive change. For example, a screen is what is displayed after tapping on an item in a burger menu or in a tab bar, it could be also limited to the content of a modal.

I don't know if we should refer to WCAG3 in this definition, like:

A screen is a particular type of view, as defined in WCAG3, dedicate d to the specific context of mobile applications.

[JJdeGroot]

This topic was discussed today, see: WCAG2ICT meeting minutes

[jamieherrera]

Would aligning with WCAG 3's definition of view create an awkward dependency if it changes as WCAG 3 continues to develop?

Only if we describe it as dependent on WCAG 3. Otherwise as a concept the idea is helpful to break away from current dependency on WCAG2ICT's weak sauce reasoning

[jamieherrera]

I am reviewing my company documentation for hybrid app content; this too would need to be defined better in both WCAG2ICT and our documentation. WCAG2ICT should have a note for how hybrid (web) content plays with "sets of native apps" vs "sets of web pages" (if the content was in a website this wouldn't be an issue so why make it this "rare" obscure concept?).
Realistically, if we hold to "set of views" within an app, which I very much think we should, we should make a note that the expectations do apply to hybrid as well.
It seems that WCAG2ICT haven't tried applying "set of screens" to native apps after reading the issues mentioned above.

[jamieherrera]

@AlainVagner 's approach above is essentially what many companies are already doing for the WCAG SC referencing sets of web pages

"I propose to start with a naive approach where we slightly adapt the definitions of set of web pages and set of software products and see what fits and what needs to be refined. We could iterate on these definitions."

I don't feel that the WCAG2ICT team has a strong enough argument for requiring their definition of software applications apply indiscriminately to native apps.

Let's make sense and not try to shove these SC into "sets of apps". This definition doesn't do any good for users

[AlainVagner]

No problem, I am totally fine with trying another approach. Thanks @jamieherrera !