[ Quick link: http://www.sixxs.net/misc/coolstuff/ ;) ] Robin Whittle wrote:
In an off-list message, Fred Templin suggested something which should have been obvious to me and which prompted this train of thought - about a way IPv6 could be of immediate benefit to end-users in a way that IPv4 can't: Home and SOHO end-users typically have their PCs etc. behind NAT and can only accept incoming communications directly, or run a server, by messing with the NAT in their modems and doing other tricky things.
Which is the sole reason I got into IPv6: I had only one public address and that was not enough, thus I then joined the then tunnelbroker IPng.nl as a user and started using it in various places, later joining the team at which point we made the move into making it into SixXS as we have today. Per default you get a /64 tunnel, and a /48 if you have earned enough credits (providing a Linked/XING link gets you there).
But this has been true for years. If it is attractive, then where is the evidence that end-users, beyond those with IETF technical interests, are actually keen to pay for a real IPv6 service?
The problem with IPv6 is of course chicken and egg: the local and remote side AND the path in between need to support it.
"Normal" users don't know about e2e though and they just know how to prod their NAT box to do what they want, also most applications circumvent those things with ease using uPNP and other such protocols.
AICCU (http://www.sixxs.net/tools/aiccu/) exists to fill the niche where people who don't want to know how to setup a tunnel can still use these kind of services. This helped in usage rate quite a bit over the years:
http://www.sixxs.net/misc/usage/ http://www.sixxs.net/misc/requests/The numbers (which are actual users, not people who ever signed up and are not using the service anymore) are still going up but not very fast, most likely because we very strict policies on who we accept and require valid data to be provided, if that doesn't happen people get rejected; I can assume that quite a number don't even signup because they don't want to provide this data, they just forget that they do provide those details to their ISPs and we are just that: an IPv6 ISP.
Also it is not an anonymous service and we loath abusers, this signup policy weeds out quite a number of those already. If we would have an anonymous signup I am pretty sure that the numbers of users would explode and so would the abuse, thus that is not going to happen, we really love being a stable users for the people who actually use it in a decent way. (Of course an ISP can have an anonymous mode for their own PoP etc, it is their PoP thus their policies are implemented, but we are not going to handle abuse for that mode of setup)
If there are attractions to at least some ordinary end-users, then is this happening with free IPv6 tunneled services?
Yes, from what I have seen in the request reasons quite a few people signup because of NNTP access to a couple of servers in .nl, see http://www.sixxs.net/misc/coolstuff/ for a list of other "Cool Stuff".
Multicast is also something that I sometimes work on and try to get up and running fully, though demand is not really big unfortunately thus people don't report it when it breaks. Because it does break it is marked as an experimental service.
on my CentOS 5.1 machine which directly handles the fixed IP address of my DSL service. It went smoothly and I got a semi-permanent /56 . Firefox on that machine could access http://ipv6.google.com .
Depends on where you are though, as from what I understand the go6 service is heavily overloaded and up/download speeds are thus very low as such hurting deployment as people then state "IPv6 is slow", ignoring the point that they are running into a bottleneck on the other side of the planet. This might quite well shy away people who try to go down that avenue too.
Check http://www.sixxs.net/misc/traffic/ for the traffic stats per PoP. Not a lot, but it is going somewhere compared to a few years ago.
Greets. Jeroen
Attachment:
signature.asc
Description: OpenPGP digital signature