Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
We've started adding timestamps to transcripts, to facilitate doing this:
So everyone doing a transcript, lets start figuring out how to do timestamps. @jennschlick has been leading the way, so maybe she can explain. The software she's using makes it much easier.
Hey @cherylcolan! To my knowledge, the jump-to-timestamp JS isn't on the site yet.
I'm adding timestamps to transcripts in
Although I read
I noticed TED Talks used the
Either way, we'll be changing everything again if the transcript element becomes part of the HTML5 spec: (see link Jenn added below)
I'll go ahead with
Good point... the whole approach will need to change at some point. So cool about the transcript element! I didn't know about that. Found a bit more about it here if anyone's curious: http://www.w3.org/html/wg/wiki/ISSUE-194/TranscriptElement
Hi :) The
For example: http://codepen.io/joe-watkins/pen/yNGvye
Here is an example of a screen reader landing on that link: http://www.screencast.com/t/pRhUOr2r
You will notice that the link without it doesn't give the user much context.
Thank you @joe-watkins. We aren't tagging a certain time of day, but we DO need a way to display the approximate time stamp of the audio file that corresponds to the transcript dialogue.
So I'm still going to stick with
I just spent about 30 mins looking for The Definitive Best Way to markup a transcript. Yep, nope. There doesn't seem to be one.
As I'm sure you know, The Web Ahead currently uses:
I haven't seen anyone else use
Looking around, I see web standards folks using
Or, er, not? Shouldn't? http://html5doctor.com/the-dl-element/ and http://www.w3.org/TR/html5/common-idioms.html#conversations suggest using definition lists like this is wrong. Hm.
The debate at hand I suppose is whether to use
There's a note in the spec about using
We are not referring to a specific moment in time in relationship to a calendar, but we are handling timecode, and referring to a relative time, in relation to the total length of the audio recording. It's a thing to be debated for sure. Perhaps
This is interesting: http://www.brucelawson.co.uk/2012/best-of-time/
Of course, any impact on accessibility tools, screenreaders, etc that a choice in element has is super important. So let me know if I'm ignorant about how
We would likely need to switch to data if we add an attribute to the element.
This comment probably doesn't make sense anymore, even. I've rewritten it a few times to add more and more info, even though I'm contradicting myself. Ah, nerd debates about semantic markup....
Ha! I love it @jensimmons
Another thing we could entertain is adding