I observe that a mere transport HTTP proxy typically costs about two
thousand lines of code.
A filtering HTTP proxy costs about four thousand lines of code.
But the caching proxies that I can find cost about eighty thousand lines of
code.
Why the huge discrepancy? Where is this huge complexity coming from?
I am interested in writing a caching proxy that will run on localhost, and
implement special knowledge of user usage that the browser does not have.
It is planned to merely add intelligence to the local caching provided by
Netscape and Internet Explorer.
Is this proposed project likely to take eighty thousand lines of code, or
four thousand lines of code? If it will take about four thousand, my
initial estimate, it is feasible. If it will take about forty thousand, it
is not feasible.
Received on Wed Aug 11 1999 - 13:23:05 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:47:56 MST