4 ms·

Aside from filtering based on moral/taste/etc grounds, I've often thought it would be interesting to filter for the sake of time. Could you conceivably trim a

by gdw2 11y ago

Aside from filtering based on moral/taste/etc grounds, I've often thought it would be interesting to filter for the sake of time. Could you conceivably trim a 3hr movie down to 90 minutes and have it still make sense? It would be an interesting experiment.

chroma 11y ago
Have you tried watching at a faster speed? I typically watch stuff at 1.2-1.3x, and 1.5x+ if it's animated (where the sped-up movements don't seem unnatural). It can be odd at first, but your mind quickly gets used to it. If you go back to normal speed, dialog seems to draaaaaag onnnnn foreverrrr.
Retr0spectrum 11y ago
There's a "Cinematic Recap" of all 5 seasons of Breaking Bad, which compresses the main story down to about 2 hours. It's quite hard to find though, due to the inherent copyright issues. Edit: Some links here: https://www.reddit.com/r/breakingbad/comments/3jt531/does_anybody_know_where_to_watch_the_2_hour/ https://www.reddit.com/r/breakingbad/comments/3jt531/does_an...
- nfoz 11y ago
  This is a great example of creative work that is not allowed to exist in a system with copyright.
  - derefr 11y ago
    I don't see why; the ideal format for such a thing isn't a derivative video, but a timecode script (like we're discussing here) that takes a legal copy of Breaking Bad as input. Similarly, I've been thinking about using a bunch of copyrighted music in a game I'm making... not by shipping the songs, but rather by just using the copies people may or may not have in their iTunes libraries, with fallbacks to non-copyright ambient music. Remix culture is too focused on embedding. Everybody forgets about the possibility of transclusion.
    - fl0wenol 11y ago
      It's difficult to transclude reliably and across platforms over which an experience is ported; not only does it require you to have the media/appropriate resource in question, but also support the embedding type (which may vary on platform, applications present, their security configuration, etc.)
    - marco1 11y ago
      Thanks for pointing this out. The whole point of this project using timecodes is to make everything copyright-friendly and safe to use.
burkemw3 11y ago
I have often wanted something like this for the 1+ hour historical shows on the likes of PBS and The History Channel. I've found that the shows repeat a lot of things before and after commercial breaks that I don't need to hear again. I enjoy the topics, but my impatience combined with the editing drives me away from watching the shows. I think the main hurdle would be actually doing the editing because of how time consuming it is. Perhaps the edit lists could be sold, and the after market editors could get paid. I wouldn't want to get into the mess of distributing copyrighted content, so I was thinking of using the seek mechanism built into video players. A user would load original content and the edit list, and the player would handle seeking correctly. In the naive implementation of this scenario, time would have to be very well synchronized between source materials. I'm not sure how much of a problem that would be. While writing this comment, the possibility of using image and sound recognition floated into my consciousness. Perhaps there would be a way to tag the images and sounds around cuts, to make them work even if time were off (e.g. commercial break ran longer in one time zone than another). If you build it, I don't want my cut; I just want to know about it!
- marco1 11y ago
  Thanks, this is definitely something that would be possible. > I think the main hurdle would be actually doing the editing because of how time consuming it is. Perhaps the edit lists could be sold, and the after market editors could get paid. That's exactly what was envisaged. This is just an open format. There could be (commercial) platforms or markets that offer complete filter lists. You don't have to do it yourself. Even more than that, you shouldn't do everything yourself. This can be a community effort, where you contribute only small parts. > I wouldn't want to get into the mess of distributing copyrighted content, so I was thinking of using the seek mechanism built into video players. This is what this project is about. You don't have to distribute source material. You have to distribute the filter lists only. The format can be "transpiled" to popular playlist formats that make use of the seek mechanisms. > In the naive implementation of this scenario, time would have to be very well synchronized between source materials. This is built-in. You may take a look at the converter/transpiler tool.
derefr 11y ago
You mean, like this, for video timecodes? http://www.telescopictext.com/ http://www.telescopictext.com/
marco1 11y ago
Thank you! This is definitely some possible use case of this project. A new category like "dispensable" would have to be added, of course.

by gdw2 11y ago

chroma 11y ago

Have you tried watching at a faster speed? I typically watch stuff at 1.2-1.3x, and 1.5x+ if it's animated (where the sped-up movements don't seem unnatural). It can be odd at first, but your mind quickly gets used to it. If you go back to normal speed, dialog seems to draaaaaag onnnnn foreverrrr.

Retr0spectrum 11y ago

There's a "Cinematic Recap" of all 5 seasons of Breaking Bad, which compresses the main story down to about 2 hours. It's quite hard to find though, due to the inherent copyright issues. Edit: Some links here: https://www.reddit.com/r/breakingbad/comments/3jt531/does_anybody_know_where_to_watch_the_2_hour/ https://www.reddit.com/r/breakingbad/comments/3jt531/does_an...

nfoz 11y ago
This is a great example of creative work that is not allowed to exist in a system with copyright.
- derefr 11y ago
  I don't see why; the ideal format for such a thing isn't a derivative video, but a timecode script (like we're discussing here) that takes a legal copy of Breaking Bad as input. Similarly, I've been thinking about using a bunch of copyrighted music in a game I'm making... not by shipping the songs, but rather by just using the copies people may or may not have in their iTunes libraries, with fallbacks to non-copyright ambient music. Remix culture is too focused on embedding. Everybody forgets about the possibility of transclusion.
  - fl0wenol 11y ago
    It's difficult to transclude reliably and across platforms over which an experience is ported; not only does it require you to have the media/appropriate resource in question, but also support the embedding type (which may vary on platform, applications present, their security configuration, etc.)
  - marco1 11y ago
    Thanks for pointing this out. The whole point of this project using timecodes is to make everything copyright-friendly and safe to use.

burkemw3 11y ago

I have often wanted something like this for the 1+ hour historical shows on the likes of PBS and The History Channel. I've found that the shows repeat a lot of things before and after commercial breaks that I don't need to hear again. I enjoy the topics, but my impatience combined with the editing drives me away from watching the shows. I think the main hurdle would be actually doing the editing because of how time consuming it is. Perhaps the edit lists could be sold, and the after market editors could get paid. I wouldn't want to get into the mess of distributing copyrighted content, so I was thinking of using the seek mechanism built into video players. A user would load original content and the edit list, and the player would handle seeking correctly. In the naive implementation of this scenario, time would have to be very well synchronized between source materials. I'm not sure how much of a problem that would be. While writing this comment, the possibility of using image and sound recognition floated into my consciousness. Perhaps there would be a way to tag the images and sounds around cuts, to make them work even if time were off (e.g. commercial break ran longer in one time zone than another). If you build it, I don't want my cut; I just want to know about it!

marco1 11y ago
Thanks, this is definitely something that would be possible. > I think the main hurdle would be actually doing the editing because of how time consuming it is. Perhaps the edit lists could be sold, and the after market editors could get paid. That's exactly what was envisaged. This is just an open format. There could be (commercial) platforms or markets that offer complete filter lists. You don't have to do it yourself. Even more than that, you shouldn't do everything yourself. This can be a community effort, where you contribute only small parts. > I wouldn't want to get into the mess of distributing copyrighted content, so I was thinking of using the seek mechanism built into video players. This is what this project is about. You don't have to distribute source material. You have to distribute the filter lists only. The format can be "transpiled" to popular playlist formats that make use of the seek mechanisms. > In the naive implementation of this scenario, time would have to be very well synchronized between source materials. This is built-in. You may take a look at the converter/transpiler tool.

derefr 11y ago

You mean, like this, for video timecodes? http://www.telescopictext.com/ http://www.telescopictext.com/

marco1 11y ago

Thank you! This is definitely some possible use case of this project. A new category like "dispensable" would have to be added, of course.