"As of writing in 2018, the average website requires downloading around 3MB and making over 100 individual HTTP requests. Browsh will turn this into around 15k

"As of writing in 2018, the average website requires downloading around 3MB and making over 100 individual HTTP requests. Browsh will turn this into around 15kb and 2 HTTP requests - 1 for the HTML/text and the other for the favicon." Does "100 individual HTTP requests" mean 100 TCP connections? As far as I know, according to RFC 2616, connection keep-alive was intended to promote making numerous HTTP requests. In fact, IME, most web servers default to setting max-requests at 100. Some are higher. Following the guidance of the RFCs, for decades I have been using this HTTP feature to make 100 requests to a site in a single connection. That is 100 pages of HTML in one quick TCP connection. If I retrieve 3MB, it is 3MB of HTML from the website and zero from third parties. (Further, I have written filters to remove junk from the HTML and print only the content I want, e.g. for reading or to import into database. If I need to split into separate files, which is rare, csplit works nicely.) In order to achieve this efficiency I could not and do not use a popular browser authored by ad-supported entities. I am an avid text-based web user who has no need for ads, graphics, and other external resources, e.g. Javascript. As such, I can use clients that can support http pipelining according to the RFCs. It works very well; no complaints. Best of luck with this project.