More on FAQ #204

204. Which medium makes it harder to deceive viewers, photography or video?

It is harder to deceive viewers with photography (and easier to deceive viewers with video).

For example, it would be very difficult to come up with a worldwide “trust” standard for video that is akin to TCQ’s role with photographs, because video has so many more ways of potentially deceiving viewers.

To list only three of those ways:

1. Video’s cascade of images

Most people know that a video or movie is just a fast succession of still photographs (they’re referred to as “frames”).

At the common rate of 30 frames per second, the viewer of a standard video sees 9,000 “photographs” every 5 minutes!

It would not be difficult to get away with doctoring even a small portion of those thousands of photographs — removing the depiction something in the background of a scene that lasts only a few seconds, for example — and viewers would be unlikely to notice as they immediately move on to view thousands more photographs.

2. Video’s matching of sound with visual images

It has taken four decades for the public to learn that in the digital age, “seeing” is not always “believing.”

It will not be easy for people to learn that “seeing and hearing together may be even less believable.”

(Tens of thousands of years of evolution — seeing people’s faces while hearing them speak — will do that to a person.)

But as the world moves permanently and irreversibly into the age of “deepfake” videos — when almost any words can be convincingly put with almost any faces — it will become more difficult for publishers of videos to win their audience’s trust.

3. Videos often are not seen whole

Anyone who has read around on this site might think, “Well, sure, but then the solution to #1 and #2 above (that is, avoiding doctoring and deepfakes) is just to record a Verifiably Unaltered Original of the video, right? Then anyone could compare the original video to the published video and see if the latter was doctored.”

Unfortunately, that solution is unrealistic, because videos longer than a minute or two typically are not published “intact.” They are edited with chunks taken out, added, and moved at various places in the middle of them before they are put before the public.

(Cropping the edges of photographs — which is allowed by TCQ — is akin to only cutting the ends off a video recording, not to removing large “chunks” from the middle of it. See also here.)

#3 is the single biggest impediment to establishing a universal TCQ-like standard for videos.

• A video standard that allows for editing videos would be toothless against abuse.

• But a video standard that does not allow for editing videos would be inapplicable to the millions of edited videos that are produced every day.