I just watched the video linked to in the OP and I watched the other link to Thomas Lunds presentation and I read the webpage at TC Electronics.
It seems like a lot of talk and very little info.
The webpage at TC Electronics mentions "K-Weighting" (hint: not K-system) and I think that should be emphasized more as it is the basis for the use of the term "loudness" in these loudness meters.
If these new meters did not measure "perceived loudness" than they would just be arbitrary re-designations of scale nomenclature like the so called "K-system".
The discussions mentioned the idea of averaged loudness and hinted that there are many proposals for how long the averaging period should be. The webpage at TCE mentioned the idea of gating out quiet periods so they did not effect the statistics, and it mentioned throwing out the top and bottom % of the measurements for a similar reason.
There were some hints that loudness standards may be described for different genres, but there was no suggestion of how to implement the use of genre based standards to multi genre distribution.
Thomas Lund's presentation seemed unorganized. He seemed to suggest that producers should anticipate where their work product will end up on the LUFS scale, and that they should try to mix to the broadcast level if they want to know what end listeners will be hearing, but then he went off and warned us that the wide variety of playback systems make predicting what something will sound like at the end of the stream seem impossible. He pointed out that in Europe the playback capability of portable devices is arbitrarily constrained and that you can not turn the volume up enough on wide dynamic range content to overcome environmental noise. He also pointed out that many of the examples of the "bad sound" that he played back couldn't be appreciated as bad by most people because they don't have good enough playback systems. (I thought that was funny... even if I also think it's true) I thought the idea of listening to the Side split of a Left/Right mix seemed interesting but the presentation failed to acknowledge that all the really matters is how the whole sounds when the mid and side are mixed together and interpreted through the psychology of the listener.
He mentioned that broadcasts are, or will be, audited for conformation to loudness standards, and that while there are still issues with embedding loudness info in meta data, producers will need to learn to embed accurate loudness data in their media so that broadcasters and playback devices can make adjustments to the loudness. He hinted at the idea that there is or will be an audit process that will confirm that the media had correct level descriptions embedded.
It seemed like there were some tidbits of info here and there.
It seems like promoting the idea that "loudness" isn't simply an RMS measurement on a freshly labeled scale but rather an estimation of pyscho acoustic perception would have been the most interesting part to discuss.
Instead the presenters seemed to want to talk about compression and aesthetics as if there is some authoritative way to define taste.
It seems like missing the forest for the trees.
best regards,
mike