Hi,

Opt-in is the norm where there is a busines model around an upgrade path. You 
want to move from X to Y, it'll cost you $Z (obviously with the exception of 
FOSS). 

Modification to an existing service product is generally always done by 
default (which client side installed software isn't). 

But regardless is a side argument which is only a distraction to the question:

Is doing this really a bad thing? If so why.

Your argument is essentially that its promoting poor user behaviour that may 
result in security issues because people wander onto nasty sites.

Fair enough. That's an entirely valid point of view. Can I extrapolate that 
out then to your support of the Fed Govt's proposed internet filter, cause 
surly that'd fix all those sites once and for all, because users can't manage 
their own behaviours?

I don't know my own view is there is a lot of 'this might happen' or 'that 
might happen' in the long term. But generally I think life will go on, there 
will be a very minimal affect out in the real world, and users who do dumb 
things because of this, are probably equally likely to undertake a range of 
other unsafe behaviour online regardless of whether BigPond are doing this or 
not doing it.


> Hi Brett,
> 
> On Fri, Nov 20, 2009 at 12:40 AM, Brett Fenton
> 
> <brett.fenton&#167;netregistry.com.au> wrote:
> > Like most product changes you make, it's not opt-in, you make the change
> > and publicize, it's a far more normal way for things to occur. I can't
> > think of a product mod we've ever done at NR where we've asked for opt-in
> > acceptance, nor as a customer of tech services have I ever been asked to
> > opt-in to modification to a product.
> 
> Actually 'opt-in' is the norm for technical products.
> 
> What version of Word / Windows / Safari / Mac OS X / Linux (as
> appropriate) are you using?
> 
> If it is not the latest the greatest, then why didn't you 'opt-in' to that?
> 
> You may be thinking of websites, such as google / gmail / etc. But
> they are wholly different.
> 
> > As a direct comparison, did ISP's ask end users for opt-in when they
> > blocked Port 25 to remote mail servers? No, but from a technical point of
> > view I
> 
> That's a good point. And I'm sure some people were nodding along in
>  agreement.
> 
> But let's look deeper. What happens when a user find that they can no
> longer send email directly from their on port 25? They have to modify
> their behaviour to send out via their ISP or their own/companies MTA.
> 
> That is a net benefit to the Internet.
> 
> What happens when a Telstra user mis-types a domain? They get
> presented with a webpage offering them a correction.
> 
> They "learn" that even if they mistype, they can still get where they are
>  going.
> 
> Extrapolate that out and, they will also learn that behaviour when
> they are on another network, visiting potentially unsafe sites.
> 
> That is a serious drawback.
> 
> I do not believe that helping users to learn to blindly trust returned
> results is beneficial to the rest of the network long-term.
> 
> > I don't see it as a 'good thing'. I'm actually pretty neutral to it. A
> > journalist called me, asked questions for 2 minutes and I explained what
> > I
> 
> From what was written, it sounded like you were smacking yourself for
> not being thinking of/doing this before.
> 
> > because they are essentially monetizing failed searches, much like while
> > I've never used a sponsored link on a monetized domain I've ended up on,
> > I don't
> 
> Hello? We are both technical people -- we are both probably able to
> diagnose why a https:// connection is failing.
> 
> The vast majority of people who use the Internet never will. They see
> "broken padlock", but it looks like paypa1 and assume it is good.
> 
> Helping them learn behaviours that is ultimately to their detriment is
> not a useful goal.
> 
> Cheers,
> Anand
> ---------------------------------------------------------------------------
> List policy, unsubscribing and archives => http://dotau.org/
>