Published run · 16 August 2026

Forty-six seconds, one question, and nothing for a phone.

A clinical literature-search tutorial from a public-sector health body, teaching a skill whose whole value is at the point of need, delivered on a fixed 1024-pixel canvas that does not reflow.

It also contains one graded question, an 80% pass mark that does not determine success, and 48 developer to-do markers left in the published package.

46s

Total runtime, all slides

Graded question in the course

1024px

Fixed width, non-responsive

TODO:: markers shipped

What was tested, and how


Course	Case C: a tutorial on finding articles on a topic in a bibliographic database · 2,345,924 bytes · SCORM 1.2
Publisher	A public-sector health body. Identity, filename and SHA-256 are held on file and will be provided to that publisher on request.
Built with	Adobe Captivate 11.5.5
Shape	Six slides · 1,380 frames at 30fps, 46 seconds · 1024×768 fixed · `responsive: false` · 151 screenshot images · four audio files
Manifest	No `masteryscore` declared
Relationship	None. Actyra did not author this course and was not engaged by its publisher.

This is the smallest course we have published a run against. That matters for what follows: a short tutorial is a legitimate format, and several findings here are about how it was packaged rather than how much of it there is.

The same packaging failure, again

From the authoring project's reporting block:

"quiz": { "minScore":0, "maxScore":10, "passScore":8 }

"contentReporting": {
 "CCCriteria":"view",  "CCData": { "viewDataIsPercent":true, "viewData":75 },
 "SCCriteria":"view",  "SCData": { "viewDataIsPercent":true, "viewData":75 }
}

CC is the completion criterion. SC is the success criterion. Both are set to view, at 75% of slides.

The author weighted the question at 10 and set a pass mark of 8. Then success was defined as having looked at three-quarters of six slides. The graded question must be answered before the course advances. There is a "You must answer the question before continuing." prompt, but with three options, unlimited retries and a "Try again." on failure, answering is a formality, and whether it was answered correctly does not reach the record either way.

Why this matters more than the arithmetic suggests

This is the same defect as Case B, produced by a different authoring tool, five years apart, at two unrelated organizations. In both cases someone set a pass mark deliberately and the export configuration quietly overrode it with a view-based criterion. Neither would show up in an authoring-tool preview. Both are invisible until someone reads the packaged artifact.

The container, not the content

The project declares its geometry in the shipped metadata, and the declaration is unambiguous:

"width": 1024, "height": 768,
"responsive": false,
"scalable": true

A fixed 1024×768 canvas that does not reflow. It will scale, which means the whole thing shrinks together, text included.

Our touch-only mobile profile ran this course on a 375-pixel viewport:

Course canvas1024px

Phone viewport375px

2.7 times wider than the screen it is being read on.

Because the course scales rather than reflows, the screenshots, controls and text all shrink together. What survives is legible only at a size no one reading a screenshot of a search-results page can use, and the course is 151 screenshots of search-results pages.

A literature-search skill is inherently useful at the point of need.

The course teaches a point-of-need task in a format that does not work at the point of need.

What we are and are not claiming

A fixed, non-reflowing 1024px canvas is the kind of issue a WCAG 2.2 audit would investigate under SC 1.4.10 Reflow. Preflight is not a conformance audit, and we do not issue accessibility findings.

What we can say: our 375px mobile profile could not use the delivered course effectively.

The one graded question, and it is a good one

Which is the best [database] search?

"antibiotic resistance" soap
antibiotic resistan* soap
antibiotic resistance soap, keyed correct

Correct-answer feedback: "Use no punctuation, tags or operators for best results in a simple topic search of [database]."

The item discriminates properly. The distractors are the two things a competent searcher from another database would actually try, phrase quoting and truncation, and the feedback explains the underlying rule rather than announcing a verdict. Whoever wrote this knew what they were doing.

It is also the only one. A 46-second course teaching a procedural skill offers a single opportunity to attempt that skill, and the incorrect feedback is "Try again." with unlimited attempts.

One more sign nobody inspected the package

48 TODO:: markers shipped in the published artifact, one on every content element, where a semantic class name should be.

They do not affect a learner and we do not score them as though they do. They are another indication that nothing inspected the final package between authoring and publication, which is the same category of signal as the editor backup files in Case A.

Verdict

Not Ready for Deployment

The hard gate fires. Success is recorded for viewing 75% of six slides. A learner can be marked successful without answering the one question correctly.

And a substantial part of its audience cannot operate it. A fixed 1024-pixel canvas, teaching a task people perform on a phone, is a defect of the delivered artifact rather than of the teaching inside it.

The question is well built and the content is accurate as far as a 46-second tutorial can be.

The content isn't the problem.

The container is.

The question is well designed and the instruction is sound for what it covers. The packaged experience undermines both the assessment and the access. This is a rebuild-the-container problem, not a throw-the-course-away problem.

What this run cannot tell you

We tested the package, not the deployment. We also did not listen to the narration, four audio files are present and we have established that no caption content ships with them, but we have not established whether the narration duplicates the on-screen text. If it does, the caption gap is less severe than it looks. That requires listening, which this run did not do.

Wonder what the package your learners actually receive contains?

Preflight checks the delivered artifact, not the course as it previews.

Preflight my course, $1,500 See what Preflight checks

If you think this might be your course

Email hello@actyra.com with the subject “Is this our course?” and we will tell you, privately, within two business days. If it is yours you get the identity, the package hash and the complete working notes at no charge, and no follow-up unless you ask for one.

That offer stands for any public-sector publisher whether or not one of these runs is theirs. If you want your own course checked against the same three critical-issue conditions, ask and we will do one free, for a course of up to 3 hours.

Read the full Learning Readiness Report for this run → · All published runs