I work at TV broadcast company, you can trust me on this:
It's "recomandation" by EBU (European Broadcasting Union) called R128. It's targeted especially at commercials being much louder than the actual content at tv/radio.
That paper dercsibes three main things:
1) It defines LUFS. They are essentially full scale decibels, but with a twist. For meaturing LUFS manufacturers need to build precisely described high-shelf filter into their meters, that compensates ear's sensitivity around 3-5k. So if you measure 1k sine with LUFS, it will return same value as dBFS meter would. If you measure 5k sine of the same volume, it fill go up by I believe 4LUFS.
2) It defines integration times and measuring gates. That means it tells you (and meter's manufacturers) how to measure average Loudness across whole "program" you're making. ...leaving silent parts out of the measurement and so on...
3) It says that the recommandation is, for all TV's and radios, to normalize all their content to -24 LUFS. That way, in theory, you shouldn't have those huge volume jumps when commercial spot goes on.
...problem that I have with R128 is, that it's far from perfect, but engineers tend to praise it. It's totally useless for certain types of content (classical music, some football matches etc...). Lot of engineers (I guess including the one you spoke to.) think it'll save the world and they require everyone and everything to target at -24LUFS. But in music bussines, as far as I know, nobody gives a damn about R128, so you would end up with intentionally quiter mix than everyone else. ...I'd say, just leave this to radio station engineers. It's essentially their problem.

EDIT: If you're interested, here's the R128:
https://tech.ebu.ch/docs/r/r128.pdf...and here's the ITU paper it refers to for definition of LUFS measurement:
https://www.itu.int/dms_pubrec/itu-r/rec/bs/R-REC-BS.1770-4-201510-I!!PDF-E.pdf