At the very end of 2010’s The Social Network (spoiler alert), Mark Zuckerberg (played by Jesse Eisenberg) sits alone at his computer, repeatedly refreshing his Facebook page after sending a friend request to an old girlfriend. In the official screenplay, he refreshes his browser twice, but in the film he refreshes seven times.

A moment after Zuckerberg sends his friend request, a snappy tune begins to play in the background. The beat of the music — a light clapping sound — is remarkably similar to the sound of the refresh keystroke. The keystroke sound is not very loud to begin with and when the timing of the music matches the timing of the keystroke, the keystroke is nearly inaudible.

A sound’s function is always more important than its volume level or clarity. Quiet, masked, buried, and canceled out sounds need to be captioned if they are significant. The seven browser refresh keystrokes are significant, despite being difficult at times for hearing viewers to hear. The question is whether every single browser refresh sound needs to be captioned. When repetitive sounds are signaled by visual cues such as hands descending onto a keyboard or a computer screen flashing its refresh state, every instance of a repetitive sound may not require a caption.

Only three captions are used in the DVD clip below: one music song title (BABY, YOU’RE A RICH MAN PLAYING) and two browser refresh captions: (REFRESHING PAGE) and (CONTINUES REFRESHING). Is this sufficient?

This brief clip raises a number of questions about captioning repetitive sounds/actions: Why weren’t the music lyrics or the initial “friend request” captioned? Do all seven “browser refresh” sounds/acts need to be captioned? Why were the second and seventh refresh acts captioned and not the others? In other words, when are visual cues insufficient and captions therefore needed? (This clip suggests, insofar as the second refresh is captioned [see 0:29], that Zuck’s face may not have provided sufficient visual evidence of a browser refresh.) Should captioning be driven by the assumption that viewers will recognize a repeating pattern of visual action in the absence of repeating captions? In other words, do we need a caption each time Zuck refreshes his browser if we can see his arm come down on the keyboard repeatedly and know what that action means?

The most important questions in this context: Does every repetitive sound need to be captioned? What visual cues are sufficient to indicate a repeating sound in the absence of a caption? Do repeating captions for highly visual actions ever become wasteful of caption space and insulting to caption viewers?

A scatter plot chart showing the seven browser refresh actions by time. Full details are included in the table below.

Refresh Number Clip Time (seconds) Caption View/Frame
1 21.0 Computer screen: Arm descends to the keyboard
2 29.8 (REFRESHING PAGE) Face/Zuck: Arm descends to the keyboard
3 41.0 Face/Zuck: Arm descends to the keyboard
4 47.4 Computer screen: Arm descends to the keyboard
5 55.2 Face/Zuck: Arm descends (keystroke sound occurs on the beat of the music)
6 65.4 Face/Zuck: Arm descends (Keystroke sound is inaudible or very faint)
7 74.1 (CONTINUES REFRESHING) Computer screen: Can’t see arm but screen flashes to indicate the refresh state

[Fair use notice: The videos on this site are transformative works used in good faith, in keeping with Section 107 of U.S. copyright law, and as such constitute fair use of copyrighted material. Read this site’s full fair use notice.]