WebAIM - Web Accessibility In Mind

E-mail List Archives

Thread: Caption: presentation rate, line length


Number of posts in this thread: 1 (In chronological order)

From: Björn Fisseler
Date: Wed, Sep 14 2022 3:49AM
Subject: Caption: presentation rate, line length
No previous message | No next message

Hello everyone,

I have a few questions regarding the line length of captions for videos
accessed over the internet. Doing some research on kind of "caption
guidelines", one quickly stumbles across various recommendations about
how long one line of caption text should be. So far I've read about
anything ranging from 37 characters per line to 42 or even 47 characters
per line. What I have not read is a rationale of these recommendations.
I can only imagine that the rationale behind the limit of 37 characters
originates from the limitations of teletext on TV, which is (was?)
limited to 40 characters. So if you would like to have some whitespace
surrounding the text you have to limit the number of characters per
line. But videos accessed over the internet, e.g. educational videos in
higher education don't have these limitations. Or am I mistaken?

Next one is about the presentation rate, as to how many words to present
per time unit. This is when people write about "160 words per minute" or
"15 characters per second". The problem I have with these
recommendations is that (a) they differ between languages, e.g. words in
English are generally shorter that words in German
("Donaudampfschifffahrtsgesellschaft"). And they don't align. A word in
German has an average length of 15 characters which sums up to 900
characters per minute or (an average of) 60 words per minute. Quite slow
for an average reader. Again, I'm missing the rationale of such
recommendations. The only summary of studies I could find is from the

And then again the DCMP recommends to edit the spoken word
(https://dcmp.org/learn/601-captioning-key---presentation-rate), which
is a no-go in academics.

The reason why this bugs me is that our captioning team refused to
provide captions for the videos that one of our professors recorded.
Their argument is that he would be speaking to fast, which would not fit
their internal guideline. Not to be said that they adopted their
guidelines from recommendations for television, not for internet videos.

My impression is that the priority should be to transcribe what a
lecturer is saying word-by-word, syncing the captions to the spoken
audio. Everything else is second-order. I would really like to know if
I'm mistaken and there is a reason why a line of caption text should be
limited to 37 characters. Videos on the internet can be stopped any
time, they can be slowed down or speed up. So for me their is no obvious
reason why to adhere to limitations from the TV era when you can watch
the video and read the captions at your own speed. Or am I wrong?

Thank you for reading this long post. I'm really interested on your
opinions on this.