7Network Working Group M. Crispin
8Request for Comments: 1733 University of Washington
9Category: Informational December 1994
12 DISTRIBUTED ELECTRONIC MAIL MODELS IN IMAP4
17 This memo provides information for the Internet community. This memo
18 does not specify an Internet standard of any kind. Distribution of
19 this memo is unlimited.
22Distributed Electronic Mail Models
24 There are three fundamental models of client/server email: offline,
25 online, and disconnected use. IMAP4 can be used in any one of these
28 The offline model is the most familiar form of client/server email
29 today, and is used by protocols such as POP-3 (RFC 1225) and UUCP.
30 In this model, a client application periodically connects to a
31 server. It downloads all the pending messages to the client machine
32 and deletes these from the server. Thereafter, all mail processing
33 is local to the client. This model is store-and-forward; it moves
34 mail on demand from an intermediate server (maildrop) to a single
37 The online model is most commonly used with remote filesystem
38 protocols such as NFS. In this model, a client application
39 manipulates mailbox data on a server machine. A connection to the
40 server is maintained throughout the session. No mailbox data are
41 kept on the client; the client retrieves data from the server as is
42 needed. IMAP4 introduces a form of the online model that requires
43 considerably less network bandwidth than a remote filesystem
44 protocol, and provides the opportunity for using the server for CPU
45 or I/O intensive functions such as parsing and searching.
47 The disconnected use model is a hybrid of the offline and online
48 models, and is used by protocols such as PCMAIL (RFC 1056). In this
49 model, a client user downloads some set of messages from the server,
50 manipulates them offline, then at some later time uploads the
51 changes. The server remains the authoritative repository of the
52 messages. The problems of synchronization (particularly when
53 multiple clients are involved) are handled through the means of
54 unique identifiers for each message.
60RFC 1733 IMAP4 - Model December 1994
63 Each of these models have their own strengths and weaknesses:
65 Feature Offline Online Disc
66 ------- ------- ------ ----
67 Can use multiple clients NO YES YES
68 Minimum use of server connect time YES NO YES
69 Minimum use of server resources YES NO NO
70 Minimum use of client disk resources NO YES NO
71 Multiple remote mailboxes NO YES YES
72 Fast startup NO YES NO
73 Mail processing when not online YES NO YES
75 Although IMAP4 has its origins as a protocol designed to accommodate
76 the online model, it can support the other two models as well. This
77 makes possible the creation of clients that can be used in any of the
78 three models. For example, a user may wish to switch between the
79 online and disconnected models on a regular basis (e.g. owing to
82 IMAP4 is designed to transmit message data on demand, and to provide
83 the facilities necessary for a client to decide what data it needs at
84 any particular time. There is generally no need to do a wholesale
85 transfer of an entire mailbox or even of the complete text of a
86 message. This makes a difference in situations where the mailbox is
87 large, or when the link to the server is slow.
89 More specifically, IMAP4 supports server-based RFC 822 and MIME
90 processing. With this information, it is possible for a client to
91 determine in advance whether it wishes to retrieve a particular
92 message or part of a message. For example, a user connected to an
93 IMAP4 server via a dialup link can determine that a message has a
94 2000 byte text segment and a 40 megabyte video segment, and elect to
95 fetch only the text segment.
97 In IMAP4, the client/server relationship lasts only for the duration
98 of the TCP connection. There is no registration of clients. Except
99 for any unique identifiers used in disconnected use operation, the
100 client initially has no knowledge of mailbox state and learns it from
101 the IMAP4 server when a mailbox is selected. This initial transfer
102 is minimal; the client requests additional state data as it needs.
104 As noted above, the choice for the location of mailbox data depends
105 upon the model chosen. The location of message state (e.g. whether
106 or not a message has been read or answered) is also determined by the
107 model, and is not necessarily the same as the location of the mailbox
108 data. For example, in the online model message state can be co-
109 located with mailbox data; it can also be located elsewhere (on the
110 client or on a third agent) using unique identifiers to achieve
116RFC 1733 IMAP4 - Model December 1994
119 common reference across sessions. The latter is particularly useful
120 with a server that exports public data such as netnews and does not
121 maintain per-user state.
123 The IMAP4 protocol provides the generality to implement these
124 different models. This is done by means of server and (especially)
125 client configuration, and not by requiring changes to the protocol or
126 the implementation of the protocol.
129Security Considerations
131 Security issues are not discussed in this memo.
137 Networks and Distributed Computing, JE-30
138 University of Washington
141 Phone: (206) 543-5762
143 EMail: MRC@CAC.Washington.EDU