Yes, it absolutely makes sense to not release details of the actual lab topology. I suppose in all reality, it would be better to ask somebody like Doug Marshke what he intends to use in the new -ENT track book, since that's what we'd want to emulate to practice.
It's a shame that there are no training vendors (official, or otherwise) that offer self-study workbooks, such as InternetworkExpert or IPExpert has for the CCIE tracks. I think this is a huge niche and could make a whole lot of money if somebody with the right skillset was inclined to do so.
As far as lab hardware goes, I know what will be in the lab (thanks to an NDA I signed while out for dinner with some Juniper guys), I just don't know how many of each device. Now, without breaking the NDA, I think its fair to say that the assumptions made so far have been good ones - however keep in mind that the -ER track also had some Cisco routers in the mix, and my guess is that the -ENT will be no different. In fact (and this is all of course 100% speculation) I would almost expect there to be a couple of Cisco switches in there too, for working with spanning-tree interoperability.
Hope that helps!