First we take up the structure of an email address.

   Net nodes corresponding to internet machines have IP addresses.
   whose Syntax is:

           [d1.d2.d3.d4]

   where d's are decimal numbers.

   Human internet addresses have the Syntax:
   
           Fully Qualified Domain Name
                |
           usr@FQDN

   where FDQN has the form:

           machine.subsubdomain.[...].subdomain.domain
   
   A Domain is the name of the broadest internet subdivision.
   Standard existing top level domains:
   
           .com        Commercial Oroganizations

           .edu        Educational and Academic organizations

           .int    Special international organizations

           .mil    Networks & orgs under Milnet run by US dept. of defense

           .gov    US government: Whitehouse, Senate, 
                   Treasury dept., IRS, NASA, Library of Congress,
                   Smithstonian ...

           .net    Organizational networks of Internet, info & serv centers
                   [Some Inet services are under .com]
                   Backbone, regional and commercial networks:
                   AlterNet, ANSnet, BARRnet, CERFnet, NEARnet, PSInet,
                   Gov type nets: ESnet, NSFnet, NSFnet
                   National Services Center

           .org    Reasearch and no for profit organizations
   
   NB local nonprofit networks in the us and other countries use a
      different terminating field in the FQDN:
   
           .us        United States
           .ca        Canada
           .il        Isreal
           .uk        United Kingdom
           .us        United States
           .au        Australia
           .pt        Portugal
           .de        Germany

            Full Country Code List
   
      Outside us, the penultimate field "ac" is equivalent to the "edu"
      terminator in .us, while a penultimate field "co" is equivalent to
      the "com" terminator in us.

      Human Internet Addresses are looked up and converted to the
      internally used IP addresses at selected computers that maintain
      conversion tables.  These are often referred to as "name servers".
      Your Internet server (the machine) keeps a list of several of
      these name servers that it can use, in case any one of these is
      in heavy use and times out or happens to be down.  Your server
      may also maintain a relatively small cache of most often needed
      conversions so that it does not have to interrogate a name server
      as often.

      While within an email address, every thing to the left of the
      separator symbol '@' is case insensitive, what appears to the
      right, may or may not be be case sensitive.  Write it as you
      see it in a From field to be safe.

   Now, secondly consider the URL syntax:

      this has the general form

           [method]://[IP address/[Full path name]
      or
           [method]://[FQDN]/[Full path name]

      where
           [method] is a variable that can be filled with names
                    of specific methods of access, exaples of which
                    are

          http
          ftp
          telnet
          gopher
          veronica

 
      These method desigbations are case insensitive.

      The [IP address] is as explained in the case of email addresses
      above; the same can be said for [FQDN].  Remember the FQDN is
      case insensitive.  There is a quirk in many FQDNs: it is often
      said that the prefix "www" or "WWW" is unnecessary.  In some
      cases it is and in others not.  If it is specified, use it.
      I can be a signal to a remote site regarding the selection of
      a particular access method, namely the http method which has a
      physical entrance or port.

      The Surprisre comes, especial for the mind dulled MSrobots
      and unix impaired internet users, when the full pathname part
      of the URL contains UC and LC  letters.  Unlike most of the crap
      that comes from MS, unix systems are and always have been
      case sensitive.  A file or directory "FOO" is absolutely
      distinct from a similarly named "foo" or "Foo".  If you
      see case distinctions in the pathname  part of a URL, use them.
      All the real internet servers run some brand of unix; increasingly
      the brand is Linux.

      Prefixes and Suffixes in full path names.

      1) DOS and its MS spawn like WINCRAP only allow one
         suffix (one "." in a name ane 3 characters in the
         extension (that which lies to the right of the dot).

      2) Unix systems have no such restrictions.

      3) The proper extension for an HTML file is ".html"
         a locution that is impossible on DOS and WINCRAP
         systems.  If you see the extension ".htm", it is
         probably not a mistake, merely an indication that
         the file resides on an inferior and less than unix
         system.

      4) Other possibile extensions for WWW accessible files
         are ".txt" (no hypertext), ".shtml", "gif", ".aps",
         ".jpeg" and others, or even no extention which is
         actually an undeclared ".txt" file.

       5) On unix systems, the "~" (tilde) symbol has a special
          significance that comes from a shorthand of the C shell
          (csh).  "~jerk", in the full pathname refers in csh
          to the home directory of the user "jerk", where ever it
          may be in the hierarchical dile system.

          On the net, the expression "~jerk" refers to the
          subdirectory "public_html" of the users home directory.
      
          If a net surfer specifies a directory such as

               http://making.cake.com/~bread/

          One of two things can happen depending on whether user
          "bread" has put a file called "index.html" or "index.htm"
          in the directory "~bread/public_html/".  If the index
          file is there, that is what will be accessed, if it is
          not there, the contents of the directory will be displayed
          as a menu.

   Other than these few rules and their exceptions, the URL structure
   and syntax is remarkably simple and clear.  As an old math professor
   once put it: why it's so simple, even a kindergarten child could do it.

   There are more strange things that can follow the full path name
   stucture which involve machine transfers, and database queries.

   Check back for when I feel like writing about and parsing that
   mumbo jumbo into understandable English.

   Have fun -