Adult content blocks on mobile ISPs

[Ian Batten]

> 
> But this is way beyond the sort of complication the authors were expecting to be legislating for [albeit that's for a court to decide, not the authors].

It strikes me that ISPs that offer progressively finer-toothed filters on content run another risk.   If an ISP offers a filtering service as a value-added proposition ("you can buy a phone for your children safe in the knowledge that we are looking after them and keeping them away from all that nasty porn") and their filtering is not (100-epsilon)% effective against said porn, wouldn't the contract holder, in this scenario a parent, have a straight-forward cause of action for non-performance, trade descriptions, etc?   

In the case of most  things you might sell with a child-protection purpose --- child car seats, or stairgates --- there are BS or DIN or EN standards that you can show compliance with, and if God Forbid a child is harmed through the failure in some way of your product, demonstrating that you have a correct test and quality regime so that both the design and each individual realisation of it meet the standard is probably a defence.   But if you are _selling_ filtering as a contractual part of an ISP offer, what effectiveness is being offered?  

Assume filtering would be aligned with BBFC criteria, so that an ISP would offer 12A, 15, 18 or R18 feeds.   What ISP is going to be able to perform filtering to a "nothing that would cause a video to get an R18 certificate" standard reliably?  Considering deep controversy about those standards (Dark Knight 12A, Made in Dagenham 15, discuss) just how are they going to do it?  Cleenfeed finesses this by punting the problem to the IWF who have a quasi-governmental role; I cannot for a second believe there is an appetite for the BBFC to "classify the Internet", a la the train-wreck of Australian government policy.  But without a legal source of ratings, just what is an ISP's position if a child's phone downloads something the parents --- perhaps rightly --- see as outside the rating that has been established.

And then, of course, we have the problem that the paying customer for the filtering (the parent) isn't the end user (the child) and the latter may have no incentive to comply with the former.  Ergo we hit a non-performance problem on the circumvention-resistance of the blocking technology.  Fun for all.

ian