Re: [MBONED] FW: adoption call for draft-rekhter-geo-distribution-control

[Jeffrey Haas <[e-mail]>]

[As a note to the WG, I'll be spinning new text on this draft in the next day
or so in order to address the comments received during Vancouver.
Primarily, too much of the use case was placed in the MDCS/MDRS documents
leaving this document a bit hard to follow.]

Bert,

On Thu, Jan 16, 2014 at 10:12:48PM +0000, Manfredi, Albert E wrote:
> I also support this, although I feel obliged to point to this short sentence, concerning subscribers who are or are not permitted to receive a given content.
> 
> "However, this document assumes that this grouping is done consistently by both the content provider and the ISP(s) that the content provider uses for delivering its content."
> 
> While this one sentence opens up all sorts of nefarious non-neutral Internet possibilities, I'm going to propose that there may be legitimate reasons for the content owner and the ISP to NOT agree. In particular, "parental controls." Seems to me that the ISP would possibly be a better repository of parental control filtering instructions than each individual content owner.

Just to briefly comment on the "nefariousness" of the feature, the
mechanisms involved don't make it any more or less possible for a service
provider to do filtering of what is delivered to an end-user.  It simply
makes it easier than having a mechanism that hits the CLI either of provider
managed CPE or the PE delivery device.

Arguably, like flowspec in general, it at least pushes visibility of such
filtering into the routing system.

-- Jeff