by Michael S. Kaplan, published on 2011/07/09 07:01 -04:00, original URI: http://blogs.msdn.com/b/michkap/archive/2011/07/09/10184809.aspx
So if you go back a bit (to June 28th), Louis C.K. was on The Daily Show with John Stewart. It played again on Thursday night as a rerun (for now you cam see the episode here), and allowed me to do the one thing I really wanted to do, and knew IS'd have the opportunity if only it came on again
You see, as I have mentioned before, I always have closed captioning on.
And the "live closed captions that never get fixed even if it turns out they were wrong" that The Daily Show uses, when Louis C.K. was saying (at around 18:30 or so):
There is a lot of people that are grosser than I am, it's true.
and the following caption that showed up just after:
Wow, GROCER instead of GROSSER?
It reminds me a little of the whole Līvər vs. Lĭvər issue I was talking about in Pronunciational ambiguity.
Only less defensible, since no one is ever going to fix this....
I'm not sure why this bothers me so much, but it does. Are we really this okay with things that are just wrong, forever?
mpz on 9 Jul 2011 9:31 AM:
I hate to say this, but pot, kettle..
It's a question of if it's worth the while. I know OCD types like me (and you?) feel the urge to correct every single mistake we find on the internet (there was an xkcd about this), but in the real world we have to balance cost, time and benefit and often times it's just not worth it.
For an interesting tidbit, the live captions are usually not typed with a regular keyboard but a chording keyboard that predictively comes up with words based on the rough sounds the live typist chords in. So it indeed is related to the pronounciational ambiguity as you correctly guessed.
Michael S. Kaplan on 9 Jul 2011 12:28 PM:
If it were me, I'd have an eye on some of these more glaring cases. I wouldn't fix the whole Internet (I usually won't fix my own Blog!), but stuff like this....
s on 11 Jul 2011 8:29 AM:
It is also increasingly common to caption this stuff using speech recognition - especially in europe where there aren't as many court reporters trained to use the chording keyboards - which would give rise to much the same mistake. The problem is that with a live or "as live" show like this they lay down the captions at the same time as they record the program. In the old analog days it was recorded as an invisible part of the picture, with two 7-bit characters per frame making a continuous stream of data. Editing this stream of data is enourmously complex, and therefore expensive. Particularly as it may endanger the master copy of the program. For workflow compatibility reasons even modern non-linear all digital equipment still presents this data as a single stream of characters, even though it probably isn't stored that way, and it is still really complex to edit. The only way to realistically fix it would be to recaption the whole show, and offline captioning is expensive to fix the kind of issue that unfortunately most of the audience are used to seeing.
So it probably doesn't justify it, but it may explain why they don't come back to this. And next time you're watching the news and you see a politician promising "attacks on bankers" in the captions, maybe you'll just shrug it off and smile :)