Lately various web based accessibility enhancement, remediation, and reporting services have been made available to web content providers under the term: Overlay.
These function by adding a snippet of JavaScript into a web content provider's page. Typically, this single line of additional code functions to enable additional content processing on each page load delivered through the overlay provider's servers, before the content is delivered to the end user. While this routing paradigm is by no means unique to accessibility applications, it does introduce accessibility opportunities and challenges.
In this document we explore overlay technologies applied for accessibility purposes. This document is intended to consider:
overlay,
Web technologies support an increasingly wide variety of online services. The typical web page payload built with HTML, CSS, and JavaScript is often accessed by many different inter-process communications. Internal representations such as a DOM tree and an Accessibility tree are two primary and widely familiar examples of payload processing, but there are many others.
While perhaps best known for introducing content transformation, overlays also include the capability to apply artificial intelligence to the inspection and reporting on web content delivered to users which today is typically a composite sourced from multiple vendors, i.e. mashups.
This W3C Community Group document endeavors to inventory the various features, functions and capabilities of overlays. In doing so we introduce a new term to the typical paradigm of accessibility evaluation, edge computing.
To date edge computing has not been addressed by W3C accessibility related publications.
The term Overlay
itself, when applied to web technologies, has a history in W3C accessibility work, especially after 2009 (when WAI took on the task of helping develop accessibility support in HTML 5). The term may have seemed appropriate to name an approach that mirrored in web content the older practice where some hardware manufacturers provided (and still provide) transparent plastic sheets called braille overlays
for touchpad controls, such as are typically found on microwave ovens. These braille overlays consist of transparent plastic with a sticky backside and appropriately located braille characters on their face. With the protective covering from the sticky backside removed, these braille overlays can be affixed onto the touchpad thus affording blind users reliable access to operating touchpad controls while not hampering visibility for sighted users. Similarly, the term was used in discussions of HTML 5 media and canvas accessibility. It is also sometimes used to describe the process of overlaying WAI-ARIA markup to add semantic information to existing content which otherwise lacks sufficient semantic clarity.
Is there a better title for this section? Should we also note early use of the term in user interface design e.g., Microsoft Windows 3.0?
Overlays operate at the client side or presentation layer, effectively the “edge” of the web information flow, the final stage of the content collection and aggregation process that defines what is actually presented to end users.
As the Internet has grown, its users have come to expect speedy interaction and persistent connectivity—whether at home, or roaming in one's geographic locale, or even on international flights. The IT industry has happily promoted this expectation of persistent real-time access to data with actions such as removing memory card support from smart phones, among other initiatives encouraging dependence on cloud data storage. In turn this trend has made edge computing a core component of virtually every organization's IT strategy. While it is true that edge computing can enhance operational efficiency and automation, the need to maintain low latency for content load on user's devices heightens the demands for greater dependence on edge computing infrastructure. In this document we explore overlays as one approach to leveraging edge infrastructure for accessibility.
Section content yet to be written—and appropriately located in this document. Include
The word “overlay” has recently acquired a particular meaning, especially (and perhaps exclusively) around accessibility on the web. While formally undefined anywhere, the term has gained currency because it draws specificity by analogy to very common practices over many generations of technology, though these have gone by other names at different times and in various contexts. A partial list of examples might include:
Overlays are created with a JavaScript function within the code of a website. Any website author can add an overlay to their site, and third-party services strive to make it as simple as possible, such as this one from MailChimp.
An accessibility overlay that is intended to be visible to end users typically appears to the as a button in the corner of their browser window. The button remains in a static position as the user scrolls through the page. When it is activated, a user-interface appears with customization options that frequently include ways to change text size, text spacing and color contrast settings.
The extent of the user-side customizations available from an overlay will vary by vendor. Many accessibility overlays also perform automated functions whose complexity and actual benefit can vary widely,. It should not be surprising that poorly designed overlays can introduce more accessibility challenges for the end user than not having the overlay at all..
It is also useful to acknowledge the customer-base for
overlays. While even sizable corporations may have limited ability to employ accessibility expertise in their content development processes, frequently, the subscriber is a small business owner
with a website created by an employee with no particular content development skills by using a fill out the form
approach to content development—on a platform like WordPress, Wix, Shopify, or
one of their many competitors. Small businesses choose these platforms
because of their simplicity and low prices. Too often accessibility considerations aren'/
t front of mind for these businesses. Too often the platform providers do too little to help support accessibility; and, when they do support accessibility, obtaining an accessible outcome is too often a question of choosing the right set of templates.
In the main the classic paradigm of a web domain owner serving content directly to end users across the Internet has long ago become a historical artifact.
Today's typical web page is often a dizzying composite of multiple content streams injected from various sources, including content unique to the user's locale and even to the specific user herself. Content delivered to a user today may be a unique one-off
composite that will change if the page is refresheda According to the 2021 Web Almanac notes more than 20 third party injection streams for today's typical web site, with 10% containing over 90 separate content injections.
We need a good graphic here.
The source of the typical website today is actually many sources.
Even third-party plugins aren't always created by a single source. Few developers exclusively use their own code. They rely on libraries, components and frameworks to build their plugins. Edge technologies may actually be best placed to help overcome the problem of substandard accessibility in composite content delivered to end users by identifying precisely the accessibility challenges that end users may experience and helping remediate them where it matters most—at the user level, for the simple reason that technology based tracking and reporting can examine far more end user payload deliveries than human testing will ever cover. Whether the resulting remediation is an inprocess transformation, or an upstream recoding, gathering and reporting many samples to upstream end points is arguably the most efficient way to spot anomalous patterns (and a great application for A-I analysis).
Ensuring accessibility of websites and applications will ever remain a challenge. While everyone agrees that accessibility is best baked-in at the beginning of the digital content life-cycle, achieving accessible results will ever remain a work in progress simply because the ever more dynamic web will itself continue to be a work in progress.
This section needs more work.
With the proviso that we are assuming well crafted approaches, some useful applications of overlays include:
both andopportunity here may well make the sum of the parts far more valuable than the traditional approach requiring independent settings on each individual device. we may do well to recall that
Backbuttons did not appear in web content because the user agents of that day lacked menu-driven mechanisms (or keyboard shortcuts) enabling the same feature.
EDITOR'S NOTE
This section needs more work.
It is to be noted that poorly designed overlays can actually work to impede accessibility. Some common antipatterns found in overlays include:
Explore a standardized, secure (and privacy preserving) cross browser and cross platform mechanism to iteratively enable users to build their own personal preference profiles. Consider how such profiles might be securely shared across a user's multiple devices.
This section needs more work. In particular we should seek to harmonize with the emerging Maturity Model (W3C Working Group Note track document) now in First Public Working Draft (FPWD) publication in the W3C Accessible Platform Architectures (APA) Research Questions Task Force (RQTF).
It has previously been noted that edge computing hasn's figured in the web content accessibility paradigm, though many remediation providers are increasingly testing with end users at the content delivery point rather than at the content server source. While testing with real users is clearly valuable, it can arguably be made even more valuable by supplementing with automated reporting. Meanwhile, identified problems may be remediable as noted above. Most important, however, is the careful consideration of how accessibility might leverage emerging industry application of W3CD authentication technologies to carry user content rendering settings across all user devices.
Consider efficiency trade-offs. What efficiency is being discussed, efficient for whom or what, and maximizing the efficiency of available components. Examples:
Following flow needs further discussion and development
Though not all overlay solutions provider operate from the same paradigm, the following is offered as a high-level overview of content creation, updating, and delivery today along with some suggestions of additional ways overlays might function to benefit users.
We emphasize that the use of overlays does not necessarily remove human expertise and oversight from a remediation process. But rather than perform the tedious and time-consuming task of identifying and remediating challenges by hand, humans can use overlays to make the traditional remediation process more efficient and comprehensive. This is already current practice for some overlay solutions providers today.
This document exists to explore both appropriate and inappropriate application of overlay technology from a technological viewpoint. We do, however, acknowledge that overlays have become highly controversial among accessibility professionals. Therefore, for ethical considerations we refer readers to the IAAP who have published a position paper regarding Overlays.
The following terms are used in this document:
Edge computing is a networking philosophy focused on bringing computing as close to the source of data as possible in order to reduce latency and bandwidth use. In simpler terms, edge computing means running fewer processes in the cloud and moving those processes to local places, such as on a user's computer, an IoT device, or an edge server.