DNSSEC and ISPs faking DNS responses

Sat Nov 14 23:36:25 UTC 2015

On 14 Nov 2015, at 23:39, Royce Williams wrote:

> Downloading is now much more common 2than during the age of the 
> browser wars.

Sure, I understand that.

>
> As of October 2014, 64% of American adults owned a smartphone [1].  
> Phones
> don't usually come with Candy Crush, but somehow, 93 *million* people
> played it daily at one point.  They many not understand that when they
> installed the app, they were "downloading" it.  But the end result is 
> the
> same.

Yes, because that leads to them doing something they want to be able to 
do, that is very tangible.  The same motivations spur VPN use (e.g., 
watching Netflix out-of-region, your example of the Olympics, and so 
forth).

To put that 93 million in context, the most recent estimates I can find 
of Internet users put their number at about 3.2 billion:

<http://time.com/money/3896219/internet-users-worldwide/>

> It sounds like we're arguing about the definition of the word "most".  
> Your
> thesis appears to be that most people won't use a VPN -- and you're
> probably right.

Yes, we're in agreement.

> But what everyone else is saying is that the value of
> "most" is likely to shrink rapidly.

I don't know about that.  It seems to me that most people who're 
inclined to use a VPN are already using one.  Unless one believes that a 
relatively high percentage of people who don't yet have Internet access 
will become VPN users once they gain Internet access.

> But now, even if Facebook's estimate [2] of 450 million WhatsApp users 
> is 90% inflated,
> there are 45 million people using encrypted texting, which I would not 
> have
> predicted.

Sure, and Apple iMessage is somewhat similar in that regard, though it's 
more susceptible to MITM.

Again, as compared to 3.2 billion.

> Most of those users probably don't know what "encryption" is.  But 
> they're
> using it.

Sure, via http/s.  But VPNs used in the sense of this discussion tend to 
imply topological masking, as well.

-----------------------------------
Roland Dobbins <rdobbins at arbor.net>