[2384] in java-interest
Re: Thoughts on the multiplication of files
daemon@ATHENA.MIT.EDU (dufourd@enst.enst.fr)
Sat Sep 30 06:16:32 1995
From: dufourd@enst.enst.fr
To: java-interest@java@sun.com
Date: Thu, 28 Sep 95 19:47:56 +0100
>It is a Java specific problem because it has to work in stand alone
>applications. It is a lot less important for WWW use because lots of small
>classes is an efficient way of using the net bandwidth. Having said that I
>don't understand your (1).
No, it is not an efficient way. I have read that TCP/IP connection setup
time is often bigger than transmission time for small files.
As for the (1), instead of code=something.class, we would specify
code=something.packed_class, and *.packed_class would be a format
like Jarchive. Jarchive is a kind of tar implemented in java
with an API for getting individual files from inside programs.
The class loader would have to be aware that it should first
try to find the next requested class file in the packed_class before
trying on the usual path.
JC
-
Note to Sun employees: this is an EXTERNAL mailing list!
Info: send 'help' to java-interest-request@java.sun.com
