Fwd: BBC News - 'Fresh proposals' planned over cyber-monitoring

[k.brown at bbk.ac.uk]

On 13 May 2013 18:45, Florian Weimer <fw at deneb.enyo.de> wrote:


> If you look at typical IPv6 textbooks, they give you a long list of
> advantages:
>
> * larger address space
> * simplified address structure
> * universal reachability of all end devices
> * protocol header optimized for efficient forwarding
> * more flexibility due to scoped addresses
> * improved security through IPsec
> * smaller routing tables due to aggregation
> * stateless auto-configuration
> * automatic renumbering between different provider aggregates
> * no broadcasts
> * improved multicast
> * built-in mobility
> * better for QoS with flow labels
>
> A lot of that turned out to be totally undesirable...

And those that were desirable were really only problems for people who
write software for routers. Not for end users. Or even people who run
computers for end users. Or even people who configure networkds for
people who run computers for end users.  And they have pretty much
been solved in the last twenty years by those people who write
software for routers.

And "simplified address structure" is only true if you are writing
software for routers.  To everybody else IPv4 looks simpler because
its just about possible for the average person to remember four
decimal numbers in a row, most people can't hold eight 4-digit hex
numbers in their head, which means they can't *read* them, which
means they are basically machine-readable-only for the average punter.

Basic nerdview mistake. Describing things from the point of view of an
insider,  so making it harder for anyone without the rignt background
to follow. (Other classic bits of nerdview in this field might include eduroam,
Freeradius and Shibboleth installation documentation - you can only
understand them if you already know how to do it; and everybody who
does know how to do it can't see why its so hard for everyone else to
follow ;-)

OK. some things are fine described from an insider point of view. I
used to do IOgens for IBM mainframes (about 25 years ago). It didn't
matter that none of it made sense to anyone who didn't have at least
about three years of system programming experience, a good idea of how
channel io worked, knew their way round a control block or twenty, and
ideally was pretty good at JCL, JES2/3, dump reading, and 370
assembler as well. Because no-one who didn't fit that description was
likely to ever get near an IOgen.

But IP addresses have escaped. They are out of the box.  They aren't
quite general knowledge (though I'd guess that at least a quarter of
the people who drink in my local have at least some idea of what they
are) but they have certainly got out into the world of PC support and
help desks and cable TV.  And IPv6, unless it is *completely*
invisible, will make a lot of people's lives harder at that sort of
level.

So we hang on untill it is completely invisible. The time to decide to
convert end-user PC networks and domestic WiFi to IPv6 is the day you
find its already been done for you and you didn't notice.


-- 
Ken Brown