RFC 1925 - The Twelve Networking Truths

RFC 1925 - The Twelve Networking Truths

原贴地址:
http://www.faqs.org/rfcs/rfc1925.html
原贴时间:
1996-01-01
原贴作者:
RFC

--------------------------------------------------------------------------------

Network Working Group                                  R. Callon, Editor
Request for Comments: 1925                                          IOOF
Category: Informational                                     1 April 1996

                      The Twelve Networking Truths

Status of this Memo

   This memo provides information for the Internet community.  This memo
   does not specify an Internet standard of any kind.  Distribution of
   this memo is unlimited.

Abstract

   This memo documents the fundamental truths of networking for the
   Internet community. This memo does not specify a standard, except in
   the sense that all standards must implicitly follow the fundamental
   truths.

Acknowledgements

   The truths described in this memo result from extensive study over an
   extended period of time by many people, some of whom did not intend
   to contribute to this work. The editor merely has collected these
   truths, and would like to thank the networking community for
   originally illuminating these truths.

1. Introduction

   This Request for Comments (RFC) provides information about the
   fundamental truths underlying all networking. These truths apply to
   networking in general, and are not limited to TCP/IP, the Internet,
   or any other subset of the networking community.

2. The Fundamental Truths

   (1)  It Has To Work.

   (2)  No matter how hard you push and no matter what the priority,
        you can't increase the speed of light.

        (2a) (corollary). No matter how hard you try, you can't make a
             baby in much less than 9 months. Trying to speed this up
             *might* make it slower, but it won't make it happen any
             quicker.

   (3)  With sufficient thrust, pigs fly just fine. However, this is
        not necessarily a good idea. It is hard to be sure where they
        are going to land, and it could be dangerous sitting under them
        as they fly overhead.

   (4)  Some things in life can never be fully appreciated nor
        understood unless experienced firsthand. Some things in
        networking can never be fully understood by someone who neither
        builds commercial networking equipment nor runs an operational
        network.

   (5)  It is always possible to aglutenate multiple separate problems
        into a single complex interdependent solution. In most cases
        this is a bad idea.

   (6)  It is easier to move a problem around (for example, by moving
        the problem to a different part of the overall network
        architecture) than it is to solve it.

        (6a) (corollary). It is always possible to add another level of
             indirection.

   (7)  It is always something

        (7a) (corollary). Good, Fast, Cheap: Pick any two (you can't
            have all three).

   (8)  It is more complicated than you think.

   (9)  For all resources, whatever it is, you need more.

       (9a) (corollary) Every networking problem always takes longer to
            solve than it seems like it should.

   (10) One size never fits all.

   (11) Every old idea will be proposed again with a different name and
        a different presentation, regardless of whether it works.

        (11a) (corollary). See rule 6a.

   (12) In protocol design, perfection has been reached not when there
        is nothing left to add, but when there is nothing left to take
        away.

Security Considerations

   This RFC raises no security issues. However, security protocols are
   subject to the fundamental networking truths.

References

   The references have been deleted in order to protect the guilty and
   avoid enriching the lawyers.

Author's Address

   Ross Callon
   Internet Order of Old Farts
   c/o Bay Networks
   3 Federal Street
   Billerica, MA  01821

   Phone: 508-436-3936
   EMail: rcallon@baynetworks.com