Why would anyone want to host in a data center with such high latency?
Why would anyone want to host in a data center with such high latency?
Otherwise, NZ is fine for the AU/NZ market.
Iceland would be fairly ideal as a "datahaven" with 1) better laws (easy) and 2) a few more redundant fiber connections, ideally in diverse routes.
HK or somewhere nearby could work for Asia. China could make an argument that HK has its own legal system, but would be providing enough overhead cover that the US couldn't just blacklist them.
A Caribbean territory might work, too, if it's an EU country affiliated (e.g. St. Martin) with connectivity back to the EU and to the US. Someone other than France would work better, since overseas French departments are the same as Hawaii is for the US. Maybe NA or a British Overseas Territory.
Of all of the datacenters in Iceland, the majority of it is all local business, and although there always seems to be chat of how perfect it would be (cheap power, better laws, strong fiber infrastructure, etc.), the reality is much different.
So, where to start? More specific specialized services? Advertising?
The other thing would be having very clear and pro user terms of service, and whatever changes in national law are needed. Maybe some tech changes (DDoS protection?) + specific marketing to vertical markets (casinos, vpn providers, usenet, torrent sites, ...), but I'm not sure what the current state is in Iceland.
If you're doing colo, it's more knowing people and maybe having a few case studies. For some kind of virtualized/managed/etc. offering, then I'd personally judge based on terms of service, measured service quality, and how secure the administrative procedures are (i.e. no rackspace/linode style administrative override...)
Second, There are millions of people living relatively close to New Zealand, so not everyone will have high latency.
It seems like the idea here is to prevent Mega's "legally challenged" (as another commenter put it) servers from being disconnected by going all the way up the 'chain of command' and creating his own backbone. He intends to give the backbone legitimacy (to forestall any funny business at the PoP) by providing free internet to New Zealanders, thus generating large amounts of 'legitimate' traffic on the link.
Edit: Just to be clear, like rdl, I don't actually think this is a worthwhile thing to do.
You could also build a cable system with huge bandwidth, 99% legit NZ origin traffic (or onward transport from AU), encrypt everything, run the cable to Los Angeles, and then buy a bunch of transport from 1 Wilshire to various locations outside the US. Encrypt your "special" traffic all the way to the various locations, announce your routes there.
None of it is airtight or bulletproof, but you can add layers of obfuscation, some of which also slow legal proceedings (some of which are still civil, some are criminal).
Still probably not worthwhile. Ultimately the best strategy for "challenging hosting" is to replicate the content itself as widely as possible, then build some kind of blacknet to send transactions and command and control. This is capable of resisting arbitrarily strong efforts at censorship.
(A great way is to distribute encrypted data widely, then drop the key at some point in the future. That's the "wikileaks insurance file" approach.)
Just translate some of his usenet postings and you will see it yourself. http://arnold.babsi.de/KIMBLE.txt