Why do ADSL customers get a 255.255.255.255 subnet mask?

That's a /32 network which works slightly differently to the networks you learn about in CCNA, but it is something that was come up with to ease the problem of IP address space exhaustion. It is all documented in the PPP specification if you are really interested but in short it isn't something you need to consider at all when preparing for your CCNA.

Subnet Mask 255.255.255.255

admittedly I'm new at CCNA, and I only scanned the RFC 3021, but it talks at length about a 31-bit prefix ie 255.255.255.254 mask (giving 2 IP addresses in the point to point connection). The only place it mentions 255.255.255.255 is:

The following text is added to section 4.3.3.9:

The 255.255.255.255 IP broadcast address MUST be used for

broadcast Address Mask Replies in point-to-point links with 31-bit

subnet masks

I can't say I know what that means, but to my reading it is still referring to a .254 subnet mask

Due to the new push with IPv6 there are going to be a lot of changes in how IP's are addressed as well as defaults, Subnet Masks, etc. Many protocols are going to be slightly tweaked to adopt the new changes.

it's a sub C subnet and if the ISP wants every IP address accessed through the gateway then it will use a subnet of that type .. it's normal for dialup and DSL