Why presentation slides should not be treated as continuous video

cRUD Research TeamThe cRUD Company

6 min read

A common design captures a presentation as a continuous video stream, just like a camera feed. It works, but it treats a document as a moving picture. cRUD Morph takes a different view. A slide is an asset and a position, and it should travel that way whenever fidelity permits.

TL;DR

  • Video, audio, slides, overlays, text, polls, chat, Q&A, reactions and captions are treated as separate media or state elements.
  • Only content that genuinely requires video is encoded as video.
  • A slide travels as a protected asset version plus an authoritative position, advanced under a controller lease.
  • Late joiners recover from a snapshot plus ordered changes, and a still slide is never mistaken for a frozen live stream.

A slide is a document shown over time

A presentation is a sequence of mostly still images, each one shown for as long as the speaker needs it. Encoding it as continuous video asks the system to keep sending pictures of something that is not changing, and it ties the readability of every slide to the same network conditions that govern a camera feed.

Treating the deck as video also hides what the system already knows. Which deck is being shown, which version of it and which slide is current are facts, not pixels. Once they are flattened into frames, they have to be guessed back out.

Separate elements, separate treatment

cRUD Morph treats video, audio, slides, overlays, text, polls, chat, Q&A, reactions, captions and other elements as separate media or state elements. Only content that genuinely requires video is encoded as video.

For presentations, cRUD Morph places the relevant facts on a plane that keeps presentation state synchronised. Those facts include the identity of the deck and asset version, the authoritative slide position, sequence numbers, controller leases and the recovery data needed to rebuild the current view. Static presentation assets are delivered as protected assets plus state whenever fidelity permits.

Who moves the slide

If more than one person can advance a deck, the system needs a clear rule about who is in control right now. In cRUD Morph, a controller acts only while it holds a fenced lease. Each change is applied as an ordered delta with a sequence number, so every viewer moves to the same slide in the same order.

Viewers fetch the protected asset version they need using signed access that expires. The position and the asset are separate, so a slide change updates the state. Viewers fetch the protected asset for the new slide rather than receiving it as video frames. Presentation state is authoritative in cRUD Morph only when the customer product delegates it.

Joining late and reconnecting

A viewer who arrives late, or whose connection drops, does not need to replay everything that happened. cRUD Morph recovers presentation state from a snapshot plus the ordered changes that followed it. A disconnection in this plane recovers from that state, and it never infers the current slide from the media stream.

A still slide is not a frozen stream

cRUD Morph applies a strict rule to video of people. It must never silently freeze to imitate a live stream. When bandwidth is constrained, quality degrades first, and if usable video cannot be maintained the experience moves to an explicit audio or avatar state.

Slides are different. A static slide may remain static naturally, because a static image is not a frozen live signal. Separating the two is what makes the rule workable. The system can be honest about a struggling camera feed without treating an unchanging slide as a fault.

When video is still the right answer

The principle is not that presentations never become video. Screen sharing is carried on the media path, and content whose fidelity genuinely depends on motion belongs there too. The point is to choose deliberately. Use the asset and its state when that preserves fidelity, and use video when the content genuinely requires it. The cRUD Morph workflow shows how each element, from speech to screen content, follows its own path.

Questions to ask

  • Does the system know which deck, version and slide is current, or only what the last frame looked like?
  • Who is allowed to advance the presentation, and how are conflicting controllers prevented?
  • How does a late joiner arrive at the current slide?
  • Can the system tell a still slide from a frozen camera feed?

A reconnect example and a practical choice

A viewer joins while slide twelve is active, loses the connection for twenty seconds and returns after the presenter has moved to slide fourteen. A continuous video path can resume from the latest frame, but it cannot independently explain which presentation asset is current. A state-led presentation path sends the protected slide asset and the current position, so the viewer returns to slide fourteen without waiting for motion that has already happened.

Continuous video

Useful when the content itself moves continuously, such as a live demonstration, camera feed or animation that depends on every frame.

Synchronised assets and state

Useful when the content is a stable slide, document page or visual that should remain sharp while only its position changes.

Combined experience

Speaker video can remain live while the presentation travels as a protected asset with synchronised state.

The distinction also improves comprehension. Text and diagrams remain sharp at the viewer's display size, late joiners can receive the current slide directly and reconnecting viewers do not need to reconstruct presentation state from a sequence of frames. Captions and other structured information can remain available to the product instead of being permanently embedded in one video image.

Video is still the right representation when motion carries meaning. The decision should follow the content. A product should not convert every element into video merely because video is available, and it should not use static assets when movement is essential to understanding.

Questions for the representation decision

  • Does meaning depend on continuous motion or on the current visual state?
  • Will the viewer need to join late or recover after a connection loss?
  • Should text and diagrams remain sharp on different displays?
  • Does the product need presentation position, captions or interaction state separately?
  • Can the chosen representation be explained honestly when quality degrades?

Summary

Presentations deserve their own treatment. In cRUD Morph, a slide travels as a protected asset plus an authoritative position, advanced under a lease and recovered from a snapshot plus ordered changes. Only content that genuinely requires video is encoded as video. This keeps slides precise and keeps the rule for video of people honest.

About this article. It explains how cRUD Morph works. It makes no performance, quality, availability or cost claim. Those results are measured for each engagement against a matched baseline.

Frequently asked questions

Why should slides not always be sent as video?
Most slides are stable visual assets. Sending them as continuous video repeats nearly identical frames, can soften text and makes the current presentation state harder to recover independently.
What happens when a viewer reconnects?
The viewer can receive the current slide asset and presentation position directly instead of waiting for a video sequence to reveal where the presenter has reached.
When is video still appropriate?
Video remains appropriate when motion carries meaning, including camera feeds, demonstrations, animation and other continuously changing material.
Can speaker video and slide assets appear together?
Yes. Speaker video can remain a live media path while the presentation is delivered as synchronised assets and both are composed into the audience experience.

Discuss how your product presents.

Tell us how presentations, screens and speakers come together in your product today.