CrabbyT, that may be, but it doesn't mean the underlying principal doesn't apply to LVLs. This is a case for engineering judgement, not blind code following. If the reason to avoid it is tension perpendicular to the grain, and an LVL has nearly perfectly aligned grain in each ply, then wouldn't it follow that you're just as likely (if not more) to develop the same tension perpendicular to the grain that you would in a piece of sawn lumber? And I don't think the glue helps that much, since it's a glue joint failure we're trying to avoid with glulams.
LVLs are proprietary products without overarching grade standards, so the NDS mostly leaves them alone and lets the manufacturer make a lot of the rules. Just because they don't specifically call this out as a problem doesn't mean it isn't. Sort of like some of the cross grain tension that develops in Simpson connectors that they don't really tell you about - not a big deal in most situations, but it can become a big deal if you push some of the larger connectors to their limits.
XR - no, I wouldn't be concerned. Most face mount connections, for instance, spread the fasteners over the whole face. The amount of tension that might develop is negated by the connector in those cases, as I believe a case could be made that the connector is acting as the necessary reinforcement of the beam there.
EDIT: CrabbyT - I just realized you crossed out the whole thing. I absolutely DISagree with that. Footnote 2 is specifically for sawn lumber and glulams (though through engineering judgement should be applied to engineered lumber as well unless the manufacturer provides evidence to the contrary). The edge distance stuff there is applicable to LVLs. Chapter 12 is not specifically for sawn lumber, it's for dowel fasteners. Chapter 8 has nothing to do with connections, just engineered lumber member design.