I think there has been some misinformation posted here.
First, the World Wide Web and the Internet are not synonymous. The Internet (begun in 1969 but not called 'Internet' until 1984) predates the World Wide Web (created by Sir Timothy Berners-Lee in 1989) by quite a few years and has been publicly accessible from the start. The HTTP/HTTPS protocols used by the World Wide Web are but two of many protocols used by the Internet (others include FTP [File Transfer Protocol], POP [Post Office Protocol], SMTP [Simple Mail Transfer Protocol], NTP [Network Time Protocol], etc.).
Second, HTTP/HTTPS headers do not route the packets of information being sent across the Internet — they define how the connection between a website and a web browser will function. (For more information, see List of HTTP header fields
.) The Internet uses TCP/IP (Transmission Control Protocol / Internet Protocol): a standard for sending 'packets' of information across the Internet. TCP/IP packets contain headers which route the packets from sender to destination. In each packet, information follows the header, including a numbering of each packet. If the packet is from a website the HTTP/HTTPS headers will be in the information which follows the TCP/IP header. If a website needs to send 500 packets to the requesting web browser to send an entire page, the packets will be numbered 1 to 500. All the packets in a transmission will not necessarily take the same path. One of the great things about TCP/IP is that a server receiving a packet which it will relay to another server (it is quite common for a transmission to go through 20 or more servers before it reaches its destination) is free to 'choose' which server is to receive the packet for relay, based on how busy other servers are at the moment. The recipient uses the numbering of the packets to reassemble the transmission in the proper order. If there are missing packets from the transmission, the recipient sends back a message to the transmitter effectively saying, 'please retransmit packets numbered [whatever] instead of requiring the entire transmission to be resent.¹
Third, there are no 'backbone servers' routing requests to the proper destination. There is a 'backbone' to the Internet: Tier One providers who have huge amounts of bandwidth (Tier Two providers have less bandwidth; Tier Three providers even less). Those huge amounts of bandwidth are the 'backbone'. Sometimes (erroneously), the root name servers which are authoritative servers for the Domain Name System (DNS) are called 'backbone servers'.
Because computers have to use zeroes and ones, but humans cannot deal in binary, DNS exists. Every ISP has at least one DNS server and there are a number of free DNS servers available. So, for example, if you want to use your web browser to go to www (dot) cnn (dot) com, you can just type the characters into the web browser's address field (no, you don't have to enter it in Google's search field and then click on Google's link to CNN!). Your computer will contact a DNS server and ask it for the IP address for www (dot) cnn (dot) com. If that DNS server has the address, it gives it to your computer and then your computer can then reach out to that IP address. If it doesn't have the address, it will contact another DNS server to see if it has the address and that will continue until the IP address for CNN is located. If necessary, the request might go all the way to one of the root name servers (there are thirteen in the world as of now), but the IP address will be sent back to your computer so your computer can reach out to CNN.
As of this writing, the IP address for www (dot) cnn (dot) com is (in decimal format) 220.127.116.11, but of course, a computer uses binary which is 100101111100101100000001001001 — I doubt anyone would prefer to enter that or even the decimal format instead of www (dot) cnn (dot) com!!
Finally, the push for HTTPS instead of HTTP is primarily because of the snooping by the 'Five Eyes'. That it makes 'injection attacks' (intercepting communication between a web browser and website and adding malicious content for nefarious purposes) more difficult is a secondary issue. A web server which has been improperly configured or running software for which a security flaw has been discovered is still vulnerable to injection attacks even if using HTTPS. Injection attacks have been going on for years, but the push for HTTPS did not start until the Snowden revelations.
¹ Because the Internet was designed to not care about the order packets are received, information being 'streamed' uses a 'buffer' which is designed to allow time for out-of-order packets to be received and resequenced — receiving a voice message out-of-order wouldn't work very well!