I always love to read this argument because it comes up so frequently and some people will always say the same things...
"I'm being nice with my OGC, but if anyone ever decides to actually USE any of it, I'm going to cut down to the bare minimum, because god forbid my open content actually becomes open."
If Phil is only talking about closing up flavor text and fluff, then I see that as something that should be done in the first place. Crunch sells, and there's no need to reuse fluff. I hope that's the case in phil's argument. If it's the "screw you all, I'm not putting out any more feats or classes because then someone else can use them" then I'm all for someone going out and creating a site like this.
There are already a ton of feats, classes, spell, AND monsters out there. I think we've seen that you can only do so much of something before it gets worn out (How many books on dwarves are there?). But, if I want to create me own SCI-FI product, I shouldn't HAVE to reinvent 10 new classes because it's a giant waste of time...there's already 50 decent ones out there, and I should be able to take the 10 best and use them in my product. Let's say for example that all of the 10 best are Phil's...and I use all 10 and base my product around them...suddenly everyone gets up in arms because I'm using someone else's open content, which is (IMHO) why the OGL was designed.
Personally, I don't think it should be free either. The person/people who would do such an endeavour deserve to be paid the for hard work it is to compile that much OGC and format it correctly, AND have a rock solid section 15. Publishers don't want others to profit on their work, so that's why they don't like this idea much.