Re: Security through wide system use?

From: LEE Sau Dan (danlee_at_informatik.uni-freiburg.de)
Date: 01/05/04 

Date: 05 Jan 2004 22:29:13 +0100

>>>>> "ÂŽik" == ÂŽik <et57> writes:

    ÂŽik> In a population with little (or no) genetic diversity, a
    ÂŽik> bacterial or viral disease can wreak havoc: all individuals
    ÂŽik> will get ill. This is why viruses in computerworld can be so
    ÂŽik> devastating: Microsoft does not provide for genetic
    ÂŽik> diversity. Neither does Intel. They are inbreeding. Even
    ÂŽik> cloning ! How many species would there be in computerworld ?
    ÂŽik> a few hundred? How many are there in the biological world ?
    ÂŽik> a few million ?

    ÂŽik> Is standardization good ?

Yes, as long as the standards only specify the *interface*, and leave
room to the *implementation*. We as biological beings have similar
interfaces: we can eat the same food and have to avoid the same kinds
of poisonous foods. We brief the same air. We drink similar water.
We need the same vitamins, minerals, etc. These are the interface
parts, which is common for us. But the doesn't mean we eat the same
kind of cuisine (implementation). I eat rice every day, and you may
eat bread and noodle instead. Some people don't eat cows, some don't
eat pigs. Some don't eat meat at all. That doesn't matter. As long
as we eat enough and have a balanced diet we're healthy.

Look at the different unix implementations (Linux, Solaris, AIX, HPUX,
IRIX, AU/X, ...) The have similar interfaces. So, the same sh script
can work across these platforms. C programs can run across these
platforms by just recompiling. Why, because the interfaces is
similar. Yet, they are different *implementations*. One defect
(e.g. bug) in one implemenation doesn't propagate to another (assuming
they're _independent_ implemenations). This is a useful resource for
debugging programs: Try running a program in an environment different
from the development environment, and many trivial bugs reveal
themselves immediately! Thanks to different implementations. 

-- 
Lee Sau Dan                     李守敦(Big5)                    ~{@nJX6X~}(HZ) 
E-mail: danlee@informatik.uni-freiburg.de
Home page: http://www.informatik.uni-freiburg.de/~danlee

Relevant Pages

  • Re: Security through wide system use?
    ... How many species would there be in computerworld? ... But the doesn't mean we eat the same ... Look at the different unix implementations (Linux, Solaris, AIX, HPUX, ... The have similar interfaces. ...
    (comp.os.linux.security)
  • Re: Security through wide system use?
    ... How many species would there be in computerworld? ... But the doesn't mean we eat the same ... Look at the different unix implementations (Linux, Solaris, AIX, HPUX, ... The have similar interfaces. ...
    (comp.security.misc)
  • Re: Java Feature Proposal
    ... > conflict. ... No, there is no naming conflict with Java interfaces, since they contain ... class to inherit multiple conflicting implementations of a method. ... I don't see why multiple inheritance of fields is any more of a problem ...
    (comp.lang.java.programmer)
  • Re: Repository
    ... the Abstract folder while interface implementations go in the Concrete ... By default interfaces and implementations are in different ... The namespaces can be manually altered if desired. ...
    (microsoft.public.dotnet.languages.csharp)
  • Re: How to implement pppIp MIB on Linux
    ... considered ioctl() but it's not something I've used often enough to be ... Admin status is shakier. ... Some OSs implementations or some network programmers? ... If you're looking for IP interfaces, the standard way to get these (on ...
    (comp.protocols.ppp)