[rc5] Security, Java, and Source

Eric Gindrup gindrup at okway.okstate.edu
Fri Oct 10 14:49:37 EDT 1997


        I'm not sure that the Java-client camp is (entirely) suggesting 
     that the Bovine developers NOT be the single point of release.  I 
     would still prefer to see the Bovine "seal of approval" on any client 
     or core that I might download in the future, Java or not.
     
        You say that Java is neither secure nor mature.  While I will 
     conceed that the hype on these points is overoptimistic, it might be a 
     bit much to claim either.  The security model is considerably better 
     than the security model of any other Algol-derived language we might 
     actually write clients in, even C/C++.  And, since people have been 
     developing in Algol-style languages for a while, there is a lot of 
     early learning that (could have been/) was skipped.  It's certainly 
     more mature than if it had leapt out of a vacuum.
     
        I also like the usable and (somewhat) friendly code that Bovine 
     releases now and I have no intention of destroying the central 
     authority for client approval.  I'd prefer to lighten their burden by 
     simplifying their code base and "auto-porting".
     
        Sure, Java isn't as secure as it's proponents would like you to 
     believe, but it's considerably more secure than C/C++.  Further, I'd 
     like to point out an abvious insecurity in using C/C++ interchangable 
     core modules:
     Client is running using "known" core.  Client reports done work and 
     requests new work.  Client discovers that it will have to download a 
     new core module to do the work.  Client attaches to the Bovine site to 
     download the new module.  Client runs using new native code module.  
     Module formats HD.
     What?  Yes, those pesky crackers cracked a DNS server near you and 
     redirected your module retrieval to be from their own machine.  You 
     got their code.  Looked like it was from Bovine, though...  And 
     there's no "perfect" signature scheme to avoid this.  Consider the 
     current effort as sufficient refutation.
            -- Eric Gindrup ! gindrup at okway.okstate.edu


______________________________ Reply Separator _________________________________
Subject: Re: Re: [rc5] Security, Java, and Source 
Author:  <rc5 at llamas.net > at SMTP
Date:    1997/10/10 17:26


On Fri, 10 Oct 1997 18:48:32 +0200 (MET DST), Ivo wrote:
     
>Two disadvantages: code size and easy_to_install.
>How many of you of there actually have a java-[interpreter|whatever] 
>running on your machine? I only have java within Netscape, but not
Nope.  Someday, perhaps, but not now.  Java is not a secure language.  Java 
is not a mature language.  Secure is a _single_ point of release client, like 
we have now.  I trust the Bovine team to release usable, friendly code.  
     
     
-- 
the Unsig
----
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the
body.
     


----
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.



More information about the rc5 mailing list