Create Temporary (Stopgaps)

Stopgaps are intentionally temporary, partial measures introduced by builders when full accessibility is not yet feasible.

Stopgaps occupy a crucial and sometimes debated space within the accessibility lifecycle because they arise when foundational work has been delayed or neglected.

They are positioned in the builder-side care space based on two key factors: 1) who implements them (the builders) and 2) the builder's intent to use them as a bridge while fuller accessibility work is underway. A measure does not become a stopgap just because it is partial. The builder must also mean it to be temporary.

They are not polished or equivalent alternatives to accessibility -- they are acknowledged as incomplete. Yet they demonstrate care by reducing harm in the short term. Stopgaps do not and should not replace the obligation to fully remediate. They buy time, reduce harm, and signal accountability.

Builders create stopgaps to be replaced by more complete acts of care. Stopgaps need end conditions, replacement paths, or explicit removal plans. If a builder intends a stopgap, incomplete by nature, to serve as a long-term fix, the measure is a manifestation of neglect.

not support channels

Unlike support channels, which can operate as standing infrastructure, stopgaps are defined by intentional temporariness and an explicit path to replacement. A stopgap can be a temporary interface, alternate workflow, manual service, or human-mediated route, but it must be openly partial and scheduled to be replaced by fuller access.

Role in the ENABLE Model

In the ENABLE model, stopgaps are builder-side interventions that arise late in the process -- often during iteration or as a last-minute response to accessibility issues. While not perfect, they represent care in motion: a refusal to launch without any accommodation.

Why stopgaps matter

Failing to do anything until a complete solution is ready places all of the burden on end-users in the mean time. Stopgaps shift some of the burden away from navigator-side by creating something usable now -- however flawed -- with a clear path toward long-term remediation. They admit imperfection while centering immediacy and dignity. A partial measure without intentional temporariness does not do that category work.

Grounding

Builders already create stopgaps in practice, even when they do not call them that. Universities commonly use Equally Effective Alternate Access Plans when a site, platform, or digital tool cannot yet meet accessibility requirements, and those plans spell out how access will be provided until conformance can be met.¹²

Federal enforcement uses similar logic. When the Justice Department moved to intervene against Miami University over inaccessible educational technologies, it argued that the university still had to provide accessible materials and equal participation rather than leave disabled students waiting for the core systems to be fixed.³

W3C does not name a general stopgaps category, but it does recognize partial conformance claims when third-party content sits outside an author's control. In those cases, authors still have to identify the non-conforming parts for users rather than pretend the page fully conforms.⁴ Partial conformance alone does not create a stopgap. The builder's intent to use a measure temporarily is the additional condition this category names.

Regulators and practitioners also document what stopgaps are not. The FTC ordered accessiBe to pay $1 million after claims that its widget could make websites compliant with WCAG.⁵ WebAIM's practitioner survey found that 67% of respondents rated overlays not at all or not very effective, and 72% of respondents with disabilities did the same.⁶

Kyle Wiggers reported that accessiBe marketed overlays as products that could make sites compliant, but those products could not cash out those claims in practice.⁷

In the news

FTC Order Requires accessiBe to Pay $1 Million for Deceptive Claims (January 3, 2025)
-- Federal Trade Commission

• The FTC ordered accessiBe to pay $1 million for falsely claiming its AI overlay widget could make "any website compliant" with WCAG. The order bars the company from making such claims for 20 years. According to UsableNet, 25% of all digital accessibility lawsuits in 2024 targeted websites using overlays -- cited as barriers, not solutions. This enforcement action confirms that overlays are not stopgaps when marketed as permanent fixes; they are manifestations of neglect and disability dongles.

CVS MinuteClinic Kiosk Settlement via Structured Negotiation (May 16, 2024)
-- Law Office of Lainey Feingold

• After the National Federation of the Blind sued over inaccessible check-in kiosks at CVS MinuteClinics, CVS agreed to remove the kiosks over 18 months and ensure any replacement technology works for blind people. This settlement demonstrates what a legitimate stopgap pathway looks like: acknowledging the barrier, providing interim alternatives, and committing to full remediation -- not deploying an overlay and claiming the problem is solved.

Examples

• An Equally Effective Alternate Access Plan for an inaccessible LMS or vendor platform until remediation is complete
• A manually updated HTML list of video links and transcripts while an inaccessible video library is rebuilt
• A temporary text-only booking or checkout path while an inaccessible widget is being rebuilt
• Publishing alt text as a separate document when it is not yet integrated into the CMS
• An alternate accessible format or human-mediated completion route while a required digital workflow is being remediated

Overlays?

Are overlays stopgaps?

Maybe, but only under certain conditions, which are rarely met.

To qualify as a stopgap, an overlay must:

• Be deployed by builders, not users (this is true for builder-side overlays, but not navigator-side tools)
• Be reactive, addressing a known, current harm or accessibility failure
• Be intentionally temporary, with a stated end condition or plan for full remediation
• Not obscure or delay true accessibility work (unlike Disability Dongles)

In reality, most overlay implementations fail this test. They're often:

• Used proactively as a crutch in place of true accessibility
• Marketed as complete solutions, not temporary ones
• Known to introduce new problems for assistive tech users

Therefore:

In theory, a builder-side overlay could be a stopgap if it's deployed in response to a specific harm, with transparency, and a plan for removal or replacement.

In practice, overlays are deployed as accessibility solutions -- so they're not considered stopgaps in the ENABLE model. They're more often manifestations of neglect or performative care, which is not care.

Care sounds like

“We know this is a partial solution -- here's what we can offer right now while we work on a real fix.”
“It's not WCAG-compliant, so we've allocated additional resources to ensure you can get the information in the meantime.”
“We didn't want to delay access entirely, so we've posted this stopgap with the current limitations clearly stated.”

Neglect sounds like

“Accessibility is coming later. Users will just have to wait.”
“No one should be using that part of the system yet anyway.”
“We're focused on the main experience -- the edge cases will have to wait.”

Real-world Scenario

An airline launches a new booking system with a flashy design -- but screen reader users report that the calendar widget is inaccessible. Instead of forcing them to call support or abandon their purchase, the dev team pushes out a basic text-based booking form within days. It doesn't have all the bells and whistles, but it works. The team labels it “temporary access form,” collects feedback, and commits to full remediation. This stopgap keeps users in the loop, and out of harm's way, without pretending the problem is solved.

---

Footnotes

1. Harvard Digital Accessibility Services: Creating an Equally Effective Alternate Access Plan (EEAAP) ↩

2. Boise State University: Equally Effective Alternate Access Plan ↩

3. U.S. Department of Justice: Justice Department Moves to Intervene in Disability Discrimination Lawsuit Alleging that Miami University Uses Inaccessible Educational Technologies and Course Materials ↩

4. W3C: Understanding Conformance ↩

5. Federal Trade Commission: FTC Order Requires Online Marketer to Pay $1 Million for Deceptive Claims That Its AI Product Could Make Websites Compliant with Accessibility Guidelines ↩

6. WebAIM: Survey of Web Accessibility Practitioners #3 Results ↩

7. Kyle Wiggers, TechCrunch, FTC orders AI accessibility startup accessiBe to pay $1M for misleading advertising ↩