The most hierarchical (and thus controllable) structure on the net is the domain naming system, the DNS, a networked database which your computer uses to translate hostnames like “google.com” into numeric addresses like 184.108.40.206, which can be used by internet routing hardware to direct packets to their destination. The routing hardware is a form of distributed autonomous intelligence (designed by darpa to survive nuclear war) but the DNS is under highly centralized control. My suggestion is that you figure out what your favorite sites are and generate a list of their numeric adddresses. There are web sites that do dns lookups and print out the address but the most sure-fire way of avoiding possibly manipulated data is to use your own computer to do the lookup. In windows you can do this by opening a command window and typing “nslookup google.com” for instance. If you want to save the output, the following appends it to the file “addresses.txt”:
nslookup google.com >>addresses.txt
On linux or mac the “host” command works better.
When you have a list you want to use, append it to your “hosts” file. On windows this is in C:\Windows\System32\Drivers\etc. You’ll need to open it as administrator in something like notepad. On linux and mac, it’s in /etc, you’ll need to be root.
If and when “they” shut down the net because of some kind of virtual false flag attack http://thoughtcrimeradio.net/2017/03/wikileaks-cias-marble-tool-framework-for-cyber-false-flags/ , or just because el presidente feels like it, the first thing they’ll do is take over the DNS. But as long as they don’t issue a router virus that propagates throughout the net (a distinct possibility) your packets will still get through to the sites you’ve listed. Any links or references from your listed sites to other sites won’t work unless they use numeric addresses (unlikely) so some things might be broken, but at least you can get to the specific pages you’re after.