This is indeed a big allocation, but this is misleading in IPv6 terms. This is equivalent in ~2 billion (2^31) /48 allocations (which is the smallest you can route via BGP) or ~140 trillion (2^47) in /64 terms, which is the literal basic unit of an IPv6 network (equivalent to having a single NATted IPv4 address). It seems to be big only because some server providers are too stingy to their IPv6 allocations (some of them only allocating one IPv6 address per server) that it flies in the spirit of relevant RFCs (and sometimes just straight up unamiguosly violate multiple RFCs).
> Maybe the real story here is that the ccTLD registrars, who weren't mentioned, are disproportionately good at deterring cybercrime.
I think that some ccTLDs requiring positive identification, usually as a side effect of residency or nationality requirements, immensely help here (versus most gTLDs requiring f***-all identification).
I definitely don't want to move to a system where making a website needs both a government and a private corporation vouching for you. That's the worst case scenario.
Your comment unfortunately implies that Google still maintains them, which is farthest from the truth (as this document shows, it is only maintained for automotive use - it is not usable on a regular phone).
I have always maintained that VoLTE has been a deficiently-defined specification, with many, many reliability concerns whisked away in the name of VoIP (which is what LTE is). It is possible to design an IP-based system capable of actual tolerance, but with the VoLTE spec so underdefined for years, these issues crop up badly.
(RCS has essentially the same problems until Google essentially monopolized it... which creates a big single-point failure problem on its own.)
Ah, no, like most journals (although NAP is not stricly a journal, PNAS is), there isn't a direct PDF link (OP: try to click that on a clean browser and you'll be disappointed).
Actually, it's Sony's patent 8,246,454 which has that "interactive networked video game" feature (https://patents.google.com/patent/US8246454B2/en), and AFAIK there isn't a "drink verification can"-style patent yet.
Not sure if it's this one in question, or even from a real patent at all, but I think this is what GP was referencing; it made the rounds a few years back.
If Mauritius decided to used a variant of Chagos/Chagas (so probably CS, since that's the only available code that still somehow fits) then IO will probably be ejected from ISO 3166.
Additionally (and more importantly), SU is still reserved in the ISO 3166 list (https://www.iso.org/obp/ui/#iso:code:3166:SU), so if United Kingdom somehow convinced ISO 3166 MA to reserve IO then it will be messy.
reply