I think he raises a good point, inadvertently. Sometimes you put more on slides not because you want your audience to read them, but because your slides are going to be an artifact of the talk for reference later. I think this is a tough tradeoff sometimes. Of course, if you think your slides are really going to undermine your talk by having too much information, then cut it down. But it's not a free lunch: you're removing information from those who may download them as a reference later.
I realize slides are not really meant to be referenced, and that that type of thing belongs in supplementary documentation, but you know that's not how it works. People download the slides and fill in the gaps in their own minds (often to the detriment of the author, since they are misinterpreted,) so you might as well try to help them fill them in with as much information as possible without sacrificing your ability to give a good talk.
If you really need to include your "paper" as part of your presentation and not as a separate document (due to constraints of SlideShare or whatever), you can attach your paper as an appendix at the end of your presentation, so it doesn't interfere with your presentation proper.
I realize slides are not really meant to be referenced, and that that type of thing belongs in supplementary documentation, but you know that's not how it works. People download the slides and fill in the gaps in their own minds (often to the detriment of the author, since they are misinterpreted,) so you might as well try to help them fill them in with as much information as possible without sacrificing your ability to give a good talk.