Most printed books are at least 1-2 years out of date.
See:
Mesh and ad-hoc network topologies are essentially similar and appear to be somewhat arbitrary and random. Well, actually, they are arbitrary and random. Kinda difficult to draw a sane topology diagram. See:
IEEE standards:
There are many glossaries that include wireless terms, acronyms, and buzzwords on the web. However, most offer only one line definitions, with little detail. If you run into a non-obvious term, I suggest just using Google to search for an explanation.
Wireless Security:
Some manufacturers databases, marketting hype, and buzzword manufacturing is intended to confuse rather than clarify. Do you get driving lessons from your autombile manufacturer or dealer? Your best sources are the various tutorials, white papers, product data sheets, and hated power point presentations by the various chip manufacturers (Broadcom, Atheros, etc) which are intended to educate the panic stricken neophyte engineer stuck with delivering a working product before the traditional unrealistic deadline.
My domain is LearnByDestroying.com for obvious reasons. Some things are best learned by experience.
What is "it"? What locks up? Can you ping the various boxes involved?
Simple economy. Tech support has become an entry level position years ago. If they knew what they were doing, they wouldn't be doing tech support. It's also difficult for anyone on the phone to answer questions about systems they've never seen or worked with. How much hands-on hardware time do you think they get in support. In my never humble opinion, experience is more important than access to the internal knowledge base. However, experience is expensive.
The reason you're getting different answers is not because of different individuals. It's because operation of the knowledgebase is dependent upon the selection of the appropriate buzzwords. Each person on the phone will tend to select a different assortment of buzzwords to pound into the search engine. Of course, results will vary. In addition, they are usually not allowed to guess, or offer advice which is not in the database.
It would also be nice if the useless and excessively over-simplified data sheets and product specifications detail such things. Yeah, you're correct. Now, imagine the multitude of configurations that one can assemble a wireless network using repeaters, WDS, bridging, and mixed mode radios? All the combinations and permuations should be tested, but often one is missed. More likely, the engineering specification indicated that it should be possible to do a mixed mode repeater (which is why it's not grayed out), but someone either forgot to test it or some manager decided that it wasn't important, or could be fixed in the next inevitable firmware release. Meanwhile, you fall into the trap.
DLink, Linksys, and Netgear are all bottom of the line vendors. One's expectations should generally follow the price tag.
Well, you'll find quite a bit on Point to Point and Point to Multipoint protocols in the various IEEE 802.11 specifications. It's rather rough reading. Pay attention to the contents of the management frames, which run the show. However, the IEEE 802.11 specs idea of point to point is NOT a transparent (workgroup) bridge, but a single MAC address wireless bridge. Point to mutlipoint is called infrastructure mode.
However, you'll find very little on proprietary implimentations such as Super-G, Turbo-G, and such. You will also have difficulty finding specifics on WDS (wireless distribution system), and repeater mode, which are similar, but proprietary to the chip manufacturers. I've found that reading the WDS code for the various open source WRT54G firmware to be the best source of detail. Also lacking in the 802.11 specifications is any protocol for transparent bridging, which also tends to be proprietary by chip manufactory. The thing to remember when reading all this stuff is that everything in 802.11 happens on the MAC layer (Layer 2) and the IP address features are just along for the ride.
Good luck...