3 ms·

>I'm old fashioned and read much faster than a human can talk, so I don't see the reason to spend 45 min watching a talking head. I agree that people can read

by jasode 8y ago

>I'm old fashioned and read much faster than a human can talk, so I don't see the reason to spend 45 min watching a talking head. I agree that people can read at 200+ wpm while most people only talk at ~100 wpm. However, some programming presentations are visually intensive with demonstrations of interactive features. That would be videos such as showing how an IDE works (debugging tips and tricks), CSS transforms, GUI workflow (e.g. navigate AWS admin screens to set up security, auto-scaling, etc). For example, I see that the landing page has severgal Google I/O keynotes and those are always heavy with demos. But yes, for presentations of static text such as explaining the new syntax for C++11, a "talking head" reading the slides may not add much value and just slow you down. As for transcripts... Since many videos seem to come from YouTube, it may be possible to download the auto-generated captions.[1] (I don't have a YouTube account so I have no idea how well this works.) I noticed one presentation where a speaker was talking about "Qt" framework but the Youtube's speech recognition auto-captioned it as "cute". I don't know if mistranslation is a rare and minor annoyance or if it makes the transcripts unusable. To downvoters: please let me know what I wrote that didn't add positive contribution to this discussion. [1] https://www.google.com/search?q=download+youtube+captions&oq=download+youtube+captions https://www.google.com/search?q=download+youtube+captions&oq...

Bromskloss 8y ago
You don't need a YouTube account to download things. Here's how to download captions using youtube-dl: https://github.com/rg3/youtube-dl/#subtitle-options https://github.com/rg3/youtube-dl/#subtitle-options I wouldn't bother with the automatically generated ones, though. They are not good enough. (Whenever there are real captions, I turn them on to be able to speed up the video beyond the point where I can no longer hear what people say. That's great!)
- jasode 8y ago
  >Whenever there are real captions, How can you tell if the captions are real vs auto-generated? Do you mean "real" as in subtitles that are burned into the video source instead of being dynamically overlaid?
  - Bromskloss 8y ago
    I don't mean burnt into the video. The YouTube inteface says "auto-generated" when it is the case (example [0]). Sometimes, though, it seems that uploaders have taken auto-generated captions and added them as if they were manually made, so then YouTube doesn't label it correctly. [0] https://www.youtube.com/watch?v=om6HcUUa8DI https://www.youtube.com/watch?v=om6HcUUa8DI
    - jasode 8y ago
      >The YouTube inteface says "auto-generated" Ok, I never noticed that. I just read the captions and assumed obvious misspellings were auto-generated. For example in this video[1], the caption text is "L1D cache misses" but he's actually saying "L1-dcache misses". (The Linux terminal screen he's showing does display "L1-dcache".) Even though that video is not labeled as "auto-generated", I assumed it was because of the bad caption. Based on your info, I guess CppCon uses humans like Mechanical Turk or other non-domain typists to manually add the captions. [1] https://youtu.be/2EWejmkKlxs?t=21m16s https://youtu.be/2EWejmkKlxs?t=21m16s
      - yorwba 8y ago
        Manual captioning is almost always not done by domain experts, but by people who have some training with a captioning system and work as professional captionists. Their main advantage is that they'll caption much faster and much cheaper than having domain experts do it, but the quality tends to suffer. In college, I met a deaf guy who always had two women accompany him to lectures; one of them would repeat everything into a mouth-covering microphone to generate an automatic transcription and the other went over it to correct obvious errors. They generated a lot of nonsense, especially when the German professor was using some English loanwords for CS concepts. I was always amazed that the deaf guy still somehow managed to learn something from these garbled transcriptions.
        Bromskloss 8y ago
        Why not sign language?
nottorp 8y ago
(Re: AWS config) Oh yes, the MCSE books with screenshots of wizards. Those can be done as a video, or condensed 100:1 if you give up on the screenshots and just write the option names. All admin interfaces that I've been in had text descriptions for every element, you just need to read them.
- jasode 8y ago
  >Oh yes, the MCSE books with screenshots of wizards. Yes, I get the point that many books have gratuitous screen shots to pad the page count. But that's a separate issue from the idea that some videos don't translate well to transcripts. >had text descriptions for every element, you just need to read them. Sure, I understand the value of reading. However, sometimes a presenter showing a live demo will wiggle the mouse cursor around certain options on the screen to emphasize particular things or clarify what may be confusing with a list of check boxes. ("You want to click this; you don't want to click that.") Or maybe demonstrate clicking on a setting and then immediately show the system dashboard statistics changing in realtime as a response. That combination of dynamic "spatial"+"time" to convey technical information is the value of live demos over reading man pages. The videos are not replacements for concise reference texts but a supplementary teaching tool. Here's an example of a video demonstrating high interaction with tools that would be very cumbersome to put into text: https://youtu.be/Wuy_Pm3KaV8?t=20m00s https://youtu.be/Wuy_Pm3KaV8?t=20m00s Lots of of mouse of movement to guide where the eye should look. Lots of scrolling and zooming. Voice annotation that's synchronized with the screen data that's constantly changing. It wouldn't translate well to text. Even if one manually created an accurate text transcript of that talk, all the missing visual cues would inhibit learning how to "drive" the tool like a car. Transcripts are only a reasonable substitute for some types of tech talks.

by jasode 8y ago

Bromskloss 8y ago

You don't need a YouTube account to download things. Here's how to download captions using youtube-dl: https://github.com/rg3/youtube-dl/#subtitle-options https://github.com/rg3/youtube-dl/#subtitle-options I wouldn't bother with the automatically generated ones, though. They are not good enough. (Whenever there are real captions, I turn them on to be able to speed up the video beyond the point where I can no longer hear what people say. That's great!)

jasode 8y ago
>Whenever there are real captions, How can you tell if the captions are real vs auto-generated? Do you mean "real" as in subtitles that are burned into the video source instead of being dynamically overlaid?
- Bromskloss 8y ago
  I don't mean burnt into the video. The YouTube inteface says "auto-generated" when it is the case (example [0]). Sometimes, though, it seems that uploaders have taken auto-generated captions and added them as if they were manually made, so then YouTube doesn't label it correctly. [0] https://www.youtube.com/watch?v=om6HcUUa8DI https://www.youtube.com/watch?v=om6HcUUa8DI
  - jasode 8y ago
    >The YouTube inteface says "auto-generated" Ok, I never noticed that. I just read the captions and assumed obvious misspellings were auto-generated. For example in this video[1], the caption text is "L1D cache misses" but he's actually saying "L1-dcache misses". (The Linux terminal screen he's showing does display "L1-dcache".) Even though that video is not labeled as "auto-generated", I assumed it was because of the bad caption. Based on your info, I guess CppCon uses humans like Mechanical Turk or other non-domain typists to manually add the captions. [1] https://youtu.be/2EWejmkKlxs?t=21m16s https://youtu.be/2EWejmkKlxs?t=21m16s
    - yorwba 8y ago
      Manual captioning is almost always not done by domain experts, but by people who have some training with a captioning system and work as professional captionists. Their main advantage is that they'll caption much faster and much cheaper than having domain experts do it, but the quality tends to suffer. In college, I met a deaf guy who always had two women accompany him to lectures; one of them would repeat everything into a mouth-covering microphone to generate an automatic transcription and the other went over it to correct obvious errors. They generated a lot of nonsense, especially when the German professor was using some English loanwords for CS concepts. I was always amazed that the deaf guy still somehow managed to learn something from these garbled transcriptions.
      - Bromskloss 8y ago
        Why not sign language?

nottorp 8y ago

(Re: AWS config) Oh yes, the MCSE books with screenshots of wizards. Those can be done as a video, or condensed 100:1 if you give up on the screenshots and just write the option names. All admin interfaces that I've been in had text descriptions for every element, you just need to read them.

jasode 8y ago
>Oh yes, the MCSE books with screenshots of wizards. Yes, I get the point that many books have gratuitous screen shots to pad the page count. But that's a separate issue from the idea that some videos don't translate well to transcripts. >had text descriptions for every element, you just need to read them. Sure, I understand the value of reading. However, sometimes a presenter showing a live demo will wiggle the mouse cursor around certain options on the screen to emphasize particular things or clarify what may be confusing with a list of check boxes. ("You want to click this; you don't want to click that.") Or maybe demonstrate clicking on a setting and then immediately show the system dashboard statistics changing in realtime as a response. That combination of dynamic "spatial"+"time" to convey technical information is the value of live demos over reading man pages. The videos are not replacements for concise reference texts but a supplementary teaching tool. Here's an example of a video demonstrating high interaction with tools that would be very cumbersome to put into text: https://youtu.be/Wuy_Pm3KaV8?t=20m00s https://youtu.be/Wuy_Pm3KaV8?t=20m00s Lots of of mouse of movement to guide where the eye should look. Lots of scrolling and zooming. Voice annotation that's synchronized with the screen data that's constantly changing. It wouldn't translate well to text. Even if one manually created an accurate text transcript of that talk, all the missing visual cues would inhibit learning how to "drive" the tool like a car. Transcripts are only a reasonable substitute for some types of tech talks.