Originally Posted by dazza9075
Where there's a will there's a way, in my opinion I couldn't care less, I'll be getting an x86 tablet, I don't see the point in getting a tablet with with its balls chopped off so lock down isn't a big issue
I've heard that argument before, as well as "if man can build it, man can unbuild it" but I've been in many hacking/development scenes where this has fallen apart. Ultimately yes, exploits always exist, because in theory you could just rebuild the whole thing from scratch without the forced signature checks. However the lengths you must go to achieve these exploits can become impractical to the point that people just give up and the development just quietly dies off.
The tivo scene is a big example of this. I used to be a very active member of the modding scene where we would run our own scripts on tivos to add features (for example, I was the very first person to transfer a recording from one tivo to another tivo, long before tivo themselves later added that feature.) However by the third generation it became necessary to do SMC rework just to run custom code. This slowed development on the S3 platform to a crawl, but it still existed to a degree.
In the S4 generation, the CPU itself actually checks for signatures, which then goes along down the chain of trust (bios > kernel > initrd etc.) I figured out a way around this, but the method requires building a daughterboard, and it's just not worth doing. There is no development going on at all for the S4 platform.
The same thing can happen with any platform.