Proposed pointer definition

[DaleenRabe]

Pointer
An input mechanism that enables a user to target and interact with specific coordinates within rendered content. The form of the pointer input can vary depending on the hardware and software.

Example
The pointer can be represented by, but are not limited to:
The cursor controlled by a mouse or trackpad.
The point of contact on a touchscreen (e.g., a finger).
The tip of a stylus on a graphics tablet or screen.
The coordinates targeted by an eye-gaze or head-tracking system.

Note
The position of a pointer is used to determine pointer-specific events, such as hover, and to initiate activation of an element, such as through a click or tap.

Pointer-based interaction is distinct from the programmatic focus. While a pointer click can move the focus to an element, the focus state is an independent concept that indicates which element will receive subsequent input from devices like a keyboard. Users navigating via keyboard or certain assistive technologies interact directly with the focus state without using a pointer.

User agents determine the pointer's coordinates based on signals from hardware (e.g., a mouse, touchscreen, or camera) and the operating system.

[patrickhlauke]

Perhaps this goes beyond what's needed, but would it make sense to also include a note about certain hardware/OS scenarios that support multiple pointers - more than one finger on a touchscreen, for instance - and the possibility of multi-pointer interactions (such as two/three finger gestures)?

[DaleenRabe]

Thanks Patrick! That is a really good point. For context, the reason I was thinking about this definition was because of "Guideline 2.3.2 Pointer focus appearance". I believe that it is important to be clear about what the definition of a pointer is in context of this guideline.

It makes sense to me to account for multi-pointer interactions. How about adjusting the "Note" section to one of these:

Option 1
The position of a pointer is used to determine pointer-specific events, such as hover, and to initiate activation of an element, such as through a click or tap. Some hardware also supports multiple simultaneous pointers, which enables multi-pointer gestures (e.g., pinch-to-zoom).

Pointer-based interaction is distinct from the programmatic focus. While a pointer click can move the focus to an element, the focus state is an independent concept that indicates which element will receive subsequent input from devices like a keyboard. Users navigating via keyboard or certain assistive technologies interact directly with the focus state without using a pointer.

User agents determine the pointer's coordinates based on signals from hardware (e.g., a mouse, touchscreen, or camera) and the operating system.

Option 2
The position of a pointer is used to determine pointer-specific events, such as hover, and to initiate activation of an element, such as through a click or tap.

Some hardware and operating systems support the use of multiple pointers simultaneously, enabling multi-pointer interactions (e.g., two-finger pinch-to-zoom gestures on a touchscreen). To ensure accessibility, functionality that relies on such gestures must also be operable via single-pointer activation.

Pointer-based interaction is distinct from the programmatic focus. While a pointer click can move the focus to an element, the focus state is an independent concept that indicates which element will receive subsequent input from devices like a keyboard. Users navigating via keyboard or certain assistive technologies interact directly with the focus state without using a pointer.

User agents determine the pointer's coordinates based on signals from hardware (e.g., a mouse, touchscreen, or camera) and the operating system.

[patrickhlauke]

Sorry, late reply, but I think I'd prefer Option 1 (as Option 2 starts to sneak in extra requirements which should be handled directly by guidelines, not mentioned in passing in a note for a definition)

[GreggVan]

hmmmm I wonder if the multipoint should be thought of completely differently than pointing. Poointing is done to indicate a particular location. Almost all the multipoint gestures have nothing to do with pointing. They might more appropriately called “Muti-touch gestures” in that the word no matter where they are done with respect to the content. They have more to do with a) the number of touchpoints, b) the type of movement they do with respect to the touch area or with each other. There is a bit of pointing - in that where you double tap with three fingers or where you center your spreading fingers will indicate the centerpoint of the zoom or squash. But in that case - the multi-touch is only used for incidating the action — and the CENTERPOINT of the multi-touch gesture is used as a pointer. Does anyone know of a multipoint touch where the exact points of the different fingers are each used as a precise pointer to something? g

…

On Oct 2, 2025, at 8:15 PM, Daleen Rabe ***@***.***> wrote: DaleenRabe left a comment (w3c/wcag3#399) <#399 (comment)> Thanks Patrick! That is a really good point. For context, the reason I was thinking about this definition was because of "Guideline 2.3.2 Pointer focus appearance". I believe that it is important to be clear about what the definition of a pointer is in context of this guideline. It makes sense to me to account for multi-pointer interactions. How about adjusting the "Note" section to one of these: Option 1 The position of a pointer is used to determine pointer-specific events, such as hover, and to initiate activation of an element, such as through a click or tap. Some hardware also supports multiple simultaneous pointers, which enables multi-pointer gestures (e.g., pinch-to-zoom). Pointer-based interaction is distinct from the programmatic focus. While a pointer click can move the focus to an element, the focus state is an independent concept that indicates which element will receive subsequent input from devices like a keyboard. Users navigating via keyboard or certain assistive technologies interact directly with the focus state without using a pointer. User agents determine the pointer's coordinates based on signals from hardware (e.g., a mouse, touchscreen, or camera) and the operating system. Option 2 The position of a pointer is used to determine pointer-specific events, such as hover, and to initiate activation of an element, such as through a click or tap. Some hardware and operating systems support the use of multiple pointers simultaneously, enabling multi-pointer interactions (e.g., two-finger pinch-to-zoom gestures on a touchscreen). To ensure accessibility, functionality that relies on such gestures must also be operable via single-pointer activation. Pointer-based interaction is distinct from the programmatic focus. While a pointer click can move the focus to an element, the focus state is an independent concept that indicates which element will receive subsequent input from devices like a keyboard. Users navigating via keyboard or certain assistive technologies interact directly with the focus state without using a pointer. User agents determine the pointer's coordinates based on signals from hardware (e.g., a mouse, touchscreen, or camera) and the operating system. — Reply to this email directly, view it on GitHub <#399 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACNGDXQUPQRCZPYNG5SDTNL3VXS63AVCNFSM6AAAAACICUVGSGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTGNRUGA2TINZTGY>. You are receiving this because you are subscribed to this thread.

[patrickhlauke]

multiple points can also be used independently - for instance, touchscreen sound/lighting consoles where you can operate many faders/controls simultaneously, but independently from each other.

resizing and rotating / directly manipulating graphical objects, where the actual points that you touch are locked in, so the object then adapts (scales/rotates) to make sure those points remain in the same relative position on the object. even pinch-zooming on maps follows the same principle. it's not just doing a gesture, it's actually tracking the two points and making sure they remain on the same location on the map, zooming and rotating the map accordingly.

and from a technical point of view, authors will receive coordinates for each individual touch point.

they may then interpret these how they see fit - extrapolating a gesture from them. but that is not necessarily the norm.

so i would not go down the abstraction route here. this just talks about the actual input mechanism.

if anything, gestures as a concept deserves its own separate definition (including single and multi-pointer gestures).

[patrickhlauke]

in light of the confusion, possible even better to generalise the proposed

which enables multi-pointer gestures (e.g. pinch-to-zoom).

to

which enables multi-pointer interactions (e.g. two or more fingers simultaneously on a touchscreen)

[DaleenRabe]

Patrick, thank you for clarifying your preference for Option 1 and for the excellent point that a definition shouldn't contain normative requirements. This makes sense and I agree we should avoid that. Your latest suggestion to generalize the wording is a great way to refine this further.

Gregg, you raise a good point about the distinction between pointing to a specific coordinate and multi-touch actions that are more gesture-based. It's true that many multi-touch actions (like a three-finger swipe) are about the gesture itself, not the precise coordinates of each finger. However, as Patrick noted, there are scenarios in direct manipulation, like rotating an object or using a virtual mixing desk, where the specific coordinates of multiple pointers are tracked and used simultaneously and independently. From a technical standpoint, the user agent is still handling multiple, distinct pointer events.

Given this, I think Patrick's suggestion to move away from the term "gestures" and focus on "multi-pointer interactions" is the right approach. It's a more accurate and encompassing term that avoids the ambiguity of what constitutes a "gesture." It keeps the definition focused on the input mechanism itself.

Here is a refined draft incorporating the feedback so far:

Pointer
An input mechanism that enables a user to target and interact with specific coordinates within rendered content. The form of the pointer input can vary depending on the hardware and software.

Example

• The pointer can be represented by, but are not limited to:
• The cursor controlled by a mouse or trackpad.
• The point of contact on a touchscreen (e.g., a finger).
• The tip of a stylus on a graphics tablet or screen.
• The coordinates targeted by an eye-gaze or head-tracking system.

Note
The position of a pointer is used to determine pointer-specific events, such as hover, and to initiate activation of an element, such as through a click or tap. Some hardware also supports multiple simultaneous pointers, which enables multi-pointer interactions (e.g. two or more fingers simultaneously on a touchscreen).

Pointer-based interaction is distinct from the programmatic focus. While a pointer click can move the focus to an element, the focus state is an independent concept that indicates which element will receive subsequent input from devices like a keyboard. Users navigating via keyboard or certain assistive technologies interact directly with the focus state without using a pointer.

User agents determine the pointer's coordinates based on signals from hardware (e.g., a mouse, touchscreen, or camera) and the operating system.

[patrickhlauke]

last minute side point: I think in WCAG 2.x we've tried to avoid "i.e", "e.g.", and so on in favor of clearer/simpler language, so one last tweak perhaps: replacing

(e.g. two or more...)

with

(for example, two or more...)

[GreggVan]

+1 to @patrickhlauke ’s suggestion to swap out i.e. and e.g. to “in other words” and “for example” an amazing number of people do not know what i.e.d and e.g. mean g

[DaleenRabe]

Thank you! I will keep this in mind going forward. For now I have adjusted in the version below:

Pointer

An input mechanism that enables a user to target and interact with specific coordinates within rendered content. The form of the pointer input can vary depending on the hardware and software.

Example

• The pointer can be represented by, but are not limited to:
• The cursor controlled by a mouse or trackpad.
• The point of contact on a touchscreen (for example, a finger).
• The tip of a stylus on a graphics tablet or screen.
• The coordinates targeted by an eye-gaze or head-tracking system.

Note

The position of a pointer is used to determine pointer-specific events, such as hover, and to initiate activation of an element, such as through a click or tap. Some hardware also supports multiple simultaneous pointers, which enables multi-pointer interactions (for example, two or more fingers simultaneously on a touchscreen).

Pointer-based interaction is distinct from the programmatic focus. While a pointer click can move the focus to an element, the focus state is an independent concept that indicates which element will receive subsequent input from devices like a keyboard. Users navigating via keyboard or certain assistive technologies interact directly with the focus state without using a pointer.

User agents determine the pointer's coordinates based on signals from hardware (for example, a mouse, touchscreen, or camera) and the operating system.