Lecture -10 Electronic Mail


Now, continuing with our discussions on internet
applications, today we would be talking about perhaps the most widely used application that
people use nowadays over the internet namely electronic mail. So our topic of discussion today is electronic
mail. Now, you know everybody knows electronic mail
is perhaps the most widely used application on the internet. Well anybody who has a connection or connectivity
to the internet invariably uses electronic mail or email it is more popularly called
in some form or the other. Now talking about email there are a number
of underline protocols which are used in practice. For instance for sending mails we use a protocol
called Simple Mail Transfer Protocol or SMTP. In conjunction with it we use another protocol
called Multi-purpose Internet Mail Extension MIME. On the other hand when you are receiving mails
on computers there are two other protocols which are used. In fact one of these two, one is Post Office
Protocol version 3 or is it is also known as POP3 in short or Internet mail access protocol
or IMAP. Now let us start our discussion on Simple
Mail Transfer Protocol. In fact this SMTP forms the basis for all
electronic mail transaction that takes on the net on the internet. Now this SMTP is the basic protocol used to
send mail from one machine to other. So let us try to understand what this SMTP
protocol is? What are its capabilities and limitations
and how it works? The first thing is that you can get all the
details about SMTP in RFC 821, this is the corresponding RFC. Well SMTP is a simple protocol in the sense
that, it has the capability of transmitting simple text messages, only it cannot transmit
other than simple text. Well when you say simple text it means a simple
seven bit ASCII file. You cannot have any international characters
which may require 8 bit ASCII encoding. It must strictly be 7 bit ASCII format. But the mostly the significant bit of the
bytes will be always 0. And in the SMTP protocol, the mails are transferred
based on information which is written on the envelope of the mail. Just like when we send a letter to a destination,
we put the letter inside an envelope and on top of the envelope. We write all the details namely the address
which will help the post office to deliver the letter at the correct destination. So in the same inside the envelope you have
the Message header. Message header can contain the address of
the recipient and some other information this we shall see later what other information
are required. This SMTP while it is transferring or transmitting
mails it does not look at the mail contents or the message body as long as it is in simple
text format it only look s at the message header. That is only important to send the mail to
the correct destination. So effectively a mail message look s like
this there are two parts. One is the Message Header, other is the Message
Body. Basically when you are trying send a mail,
the Mail is created by a so called user agent program which is also sometimes called as
a mail client. Some examples of usage and programs are Outlook
Express Outlook or you can say Pine Eudora this kind of mails you can say client or user
agents are available. Based on the platform you are using you can
use one of these platforms to compose your mail whatever you trying to send you can type
in the body of the mail. Now once I have created the mail and have
given a command to send it, the messages get queued up and these messages go the input
of an SMTP sender program. We shall be illustrating this by an example. With the help of a diagram now this SMTP sender
program is actually responsible for sending the mails. This SMTP sender program is run usually as
a server process which for example in the UNIX environment it is sometimes called a
Daemon process. Some examples of some SMTP process are sendmail
or qmail. Both of these are widely used qmail is slightly
better than sendmail because of some of its attractive features. Now let us try to illustrate the process with
help of this diagram. Here we have the User Agent where the mail
is getting composed. But however this User Agent may not be running
the SMTP server program. Suppose this is the place where the nearest
SMTP program is running out here. So this SMTP server will be acting as the
SMTP sender or the mail sender. The User Agent after the mail gets composed
will open a connection to the SMTP server on Port number 25. Port number 25 is the Port for SMTP server. And as you can see several mail messages might
get queued up in the User Agent processed by the SMTP sender program, so each of these
queued up mails will be containing Mail Header followed by a Mail Body. Now this SMTP server program when it receives
the mails, it will try to forward the mail to the destination. Now it is possible, but it does not have a
direct path to the destination it may have to send the mail through a number of intermediate
mail servers this may be necessary. So for that purpose mail server program out
here this may send the mail to some intermediate SMTP server. One or more of them with a connection on Port
25 established in each case and finally it will be delivered to an SMTP server which
has a connection with the destination User Mailbox directly. So this SMTP server will receive the mail
and will find that well this particular mail is destined to a mailbox on my machine. So the mail will get deposited in the appropriate
mail box. This is how the overall thing works. So starting from the User Agent you go to
an SMTP server which acts as the SMTP sender. To start with the SMTP sender will be sending
the mail to another SMTP server which may simply forward it to another SMTP server and
this process will go on till the mail receives the final SMTP server which acts as the final
destination or receiver which accepts the mail and puts it in the User Mailbox. This is how the overall SMTP mail transmission
phase works. Now in terms of the mail message contents,
each message that gets queued up contains of course the text of the message. The message text in fact contains a header
and body. Now in the header there is an RFC 822 which
defined the fields of header. So if you look at 822 you can know all the
details of the header fields. So this header will contain message envelope
containing some information as we shall see in some examples and also a list of recipients. And it will also contain the message body
which of course will be composed by the user. And in addition to the message text each queued
message will also have a list of mail destinations. This list can be derived by the user agent
or the SMTP server from the header. The idea is that although the mail destinations
are anyway present in the header of the message. But in order to process it in order to forward
it that same information is extracted from the main header and also kept in a separate
place which can be accessed very fast by the mail sender program. So the mail as a whole is kept somewhere and
only the destination addresses are kept separately along with a pointer to the corresponding
mail. So that it can be processes relatively fast. And this list of mail destination may sometimes
require expansion of mailing list. Like you know sometimes we have the group
mail facility we give an email address which actually is a mail alias which actually refers
to a group of email addresses. So when a mail is being sent to that alias,
that alias expands to all the email addresses that belong to that group. So this kind of expansion can also take place. With respect to the SMTP Sender it takes messages
from the queue one at a time. It tries to transmit it to proper destination
via an SMTP transaction. It is a TCP connection over port number 25
and when all the destinations corresponding to a mail has been processed please note that
a particular mail message may have more than one destination. Unlike a postal an electronic mail can have
more than one destination. So when all the destinations have been processed,
there are by the copy of the mail has been sent to all the destinations it is destined
to the messages deleted from the queue. Well some optimizations are carried out by
the mail server like, if a message is being sent to multiple users on the same host, then
the mail message will be sent only one to the destination host. The delivery to the individual users will
be handled at the destination host. So if the mail is going to two different users
on the same machine Hotmail.com. Say only a single copy of the mail will go
to Hotmail.com with information in the header that this mail has to be delivered to two
of these addresses. So the final destination say Hotmail.com will
do the rest after receiving the body. And secondly if there are multiple messages
are ready for the given host may not be the same user, multiple users then instead of
sending. Well this is not the same message in the first
case it was the same message going to more than one user on the same machine but here
multiple messages going to the same machine. Then to save time a single TCP connection
can be opened and all the messages can be sent. This actually overheads this actually reduces
the overhead of a creating and terminating. The TCP connection over port number 25 you
have to do it every time. We are sending mail, so it reduces the number
of time you are establishing the connection. Now while sending a mail there are a number
of errors which are often encountered. Like due to some problem a host may be unreachable. It can be a problem with the host; it can
be a problem with the link, through which the host is connected. Due to some reason the host of maybe out of
operation. TCP connection may fail due to some error
in the network. The destination address may be faulty like
the name of the user or the user id may be wrong. The target user address may have changed. Sometimes some of the mail servers allow redirection. If the user address has changed automatically
the mail will be redirected to an alternate email address. Now if this redirection is not possible usually
the user is informed in the form of an error message. And if this redirection feature is there then
the sender can re-queue the mail well after the message comes back from. The final sender final mail server now which
of course it can keep in the queue for certain period. After that it will give up because if the
mail has failed often a mail bounces with an error message. That well I have kept it in my queue. I am unable to have a connection or some problem. Now the mail will remain on an outgoing queue
for certain period of time that particular period of time is considerable. It can be 1 hour, it can be 1 day, it can
be 7 days, and it can be even more. Depending on the time the SMTP server will
be keeping on trying to send the mail during that period. And if it fails then that period will simply
discard the mail after that. Talking about the reliability of SMTP protocol. Well we have mentioned that it uses TCP connection
over port number 25. Well we know that TCP is a connection oriented
protocol it tries to recover from errors. Supposedly TCP is reliable but in an application
like electronic mail which runs on top of TCP there can be so many different kinds of
errors. Like I just mentioned some host may be unreachable,
some user email address may have been redirected, some other address and so on. So under these conditions, so all though TCP
attempts to provide reliable service, there is no guarantee that the mail will be delivered
directly. Some messages may be lost. There is no guarantee to recover lost messages
see although tries to ensure that end to end transactions successful. For example from your machine to the nearest
SMTP server that transaction is successful. But somewhere beyond that something might
have happened which has called the mail message to get lost. So there is no effort made by the SMTP protocol
to recover from these lost messages. Similarly, there is no end-to-end acknowledgement
from the final destination back to the sender that is used in SMTP. Similarly error indication report is not guaranteed
sometimes you get back a message that your mail has bounced. But even if you do not get back a message
there is no guarantee that the mail has actually reached the destination. We trust SMTP well if you do not get an error
message we assume that the mail has reached the destination. But in practice it may not be so. Well, now at the side of the receiver well
when you are talking of SMTP receiver we were talking of the SMTP server at the final destination
which is finally connected to the user mailboxes. These are the user mail boxes. Now these SMTP receivers will be accepting
the arriving messages and they will be placing the messages in the appropriate user mail
boxes. If the mails are destined to a mail box connected
to the particular server. But if it finds that the destination address
does not belong to a particular user in the mailbox of its own, machine own server. Then it copies the mail message to an outgoing
queue to some other SMTP server to a process called mail forwarding. Suppose I am an SMTP server I receive a message,
first I see if the destination corresponds to a user on my machine or not. If so I put in the appropriate mailbox. If not, I try to find out which SMTP server
I have forward it next I simply forward it to the next SMTP server. This is how it works. So in doing so the receiver must verify the
local mes mail destination of course. Because it will have to verify before actually
putting them in the mailbox and it deals with some kind of headers like. It can check whether the packet that has arrived,
has any error or not. There is a checksum computation or it can
also check whether the user mailbox where you are trying to put the mail whether it
has some free space or not. Many mail servers allow something called a
disk quota where the maximum amount of space you can use for storage of your mails is limited. And if that quota gets exceeded or filled
up, no more new messages can be put in there so that is also an error condition. And SMTP forwarding means you have an SMTP
server out here. It gets a mail as input it finds out. It has to forward the mail to some other SMTP
server. This I a process called forwarding. Now this forwarding may need to go through
a number of intermediate mail servers in general. Now it is possible in electronic mail in email
that the sender can specify the exact route. Well if it does not then like routing each
SMTP server will try to find out the best route. Well the sender can also specify a route through
the mail messages will go the sequence of SMTP servers the mail message will traverse
before reaching the final destination mail server. So again looking back at the same diagram
I want to just make a one comment here. If you look at the first SMTP server in this
chain. The SMTP server which receives mails directly
from the User Agent. This works as a SMTP receiver and sender both
it receives mails from the User Agent over Port number 25. It forwards some of the mails again over Port
number 25 to some other SMTP server. So what I mean to say is that the intermediate
SMTP server they have dual functionality. They act as server when they are receiving
mails from somebody else. They also act as SMTP client when they are
forwarding mails to some other SMTP server. So you see these. These SMTP nodes intermediate nodes they have
a dual role of both client and server. They are client when they are trying to send
the message to another server. They are server when they receiving a message
from another node which is acting as the client. So this is what we remember. Now talking of the basic commands that SMTP
uses well there are commands, there are responses. One node sends a command to the server; a
client sends a command to the server. The server sends back a response to the client. Now let us see what this command and response
messages look like for they actually mean and represent. The first thing is that this command and response
the initiative lies with the sender. Because if the sender is not responsive this
dialogue cannot go on the process starts by the client establishing a TCP connection with
the SMTP server of course over port number 25. There are a number of commands we shall see
through a complete example. Sender can send some command to the receiver. An example command is HELO. In one such command HELO followed by a domain
name followed by carriage return and line field this is a command. And the server after receiving the command
the reacts by responding with a reply. A typical reply look s like this it starts
with a 3 digit response code followed by the actual string indicating the meaning of this
code. For example 250 requested mail action completed. Talking of the 3 digit codes. These 3 digits code can start either with
2, 3, 4 or 5. Well any numbers starting with 2 means Positive
completion. That means it is a positive response it is
successful and the step that it is referring to has been completed successfully. Well anything staring with 3, this also means
positive. But this is not the end of the story. We are only half way through the process this
is sometimes like a positive intermediate reply. You may have to go through one or more such
positive intermediate reply. So as to reach or arrive at a positive completion
reply. Similarly anything staring with 4 is a Transient
negative completion reply negative completion means some failure. But Transient means well we are not very sure
about what it is about yet maybe this error will be recovered. But 5 starting with 5 means this error is
of a permanent nature and the mail transaction has to be stopped or aborted these are meaning
of the codes. Now in terms of the dialogue that goes on
between the mail client and the mail server broadly. There are three steps. First is the connection setup. Second is a step where a number of command
and responses are exchanged between the client and the server. And finally the connection gets terminated. Now let us see how these three steps look
like. Well in the connection set up step the sender
opens a TCP connection with the receiver. This is very easy this can be done by the
sender giving an explicit telnet command. Telnet followed by the server name whatever
server you are using, followed by the port number of
SMTP. Telnet server name followed by 25, this will
allow the client to open an SMTP connection with the SMTP server. Now once connected the receiver will identify
itself by message like 220, its domain name server ready. Sender will then identify itself by HELO message. You will see a complete transaction very shortly
receiver will accept the sender identification by sending back a 250 positive reply. Now if the mail service is not available at
that particular point. Then instead of 250, a message like this will
come 421 indicating this is a negative completion reply, 421 services not available. Now in the second step when we are talking
about the actual mail transfer commands, there are some commands which identify the email
address of the source and the destination because email of email address of the destination
is required for obvious reason the mail has to be delivered sometimes email address of
the source is also required particularly when error messages have to be sent back in case
of some errors the receiver might like to send back an email message notifying the error
back to the sender. So for that reason the email address of the
sender is also required to be present as parts of the email message the header. So in the mail transfer command these kinds
of messages are used. There is a command called MAIL FROM command
which identifies the sender of the email or the originator. This as I had said this is required to provide
the reverse path for error reporting. Suppose you have to send back an error message
you can use this email address for doing that and if the email address, if it has the proper
syntax this the receiver will return with a 250 message or if there is something wrong
then there will be an appropriate failure message. Similarly there will be a receipt TO command
where you can identify the email address of the recipients. There can be more than one recipient which
is identified by more than receipt TO commands. There will be reply for each recipient. Finally there will there is a command called
DATA which indicates that now the actual body of the email starts. So after the DATA command the client can start
sending the messages which will form the body of the email to the mail. You can say to the SMTP server for processing
and this message can go on continuously. The end of the message will be indicated by
a line containing just a full stop or a period. A full stop appearing in the first column
of a line will indicate the end of the message. Before that you can go on typing whatever
you want. And finally when you are trying to close the
connection, there are two steps. Basically the sender or the client will be
sending a quick command to the server and will wait for the reply. After the reply comes back from the server
it initiates TCP close operation. The receiver similarly after sending the response
to quit. It also initiate TCP close. This is how the connection is closed. Well, now we try to demonstrate an example
SMTP session. Well normally the way we send and receives
mails is that we use some mail software look Outlook, many of us also use a web based like
Hotmail, Rediff Mail, and Yahoo Mail. They have a suitable web based user interface
through which we type mails we shall see later. How these are implemented in practice. See at some point in the implementation, these
machines where you are sending your mails the body of the mail addresses subject whatever
to they must respond to the SMTP request and response. There must be an SMTP server situated somewhere
which must be contacted to send. The messages we have typed as an email message. So this simple example that we will show,
will demonstrate how a low level SMTP session is carried out. Well although we may be using a high level
tool or a high level package to send the mail. But actually in internally these are the kind
of commands that get generated automatically and your mail gets sent. So let us see how the process starts. Here I am assuming that the client is a simple
window a command window from where I am establishing a connection with my server. Well this can be any mail server I am giving
a hypothetical name, server name. If you give telnet over port number 25, this
will open a TCP connection over port number 25. So a TCP connection gets established over
port number 25. Now after this connection is established the
dialogue can start. Let us see the slide shows you how the dialogue
goes on. Here S indicates the mail server client indicates
the mail client. So after the connection is established the
server responds by a welcome message 22. Suppose this server is Hotmail.com server
I am connecting it to Hotmail.com, it will send 220 hotmail.com Simple Mail Transfer
Service Ready, something like this. Now after that the client will give a HELO
message and also it will identify the domain name. The client belongs to suppose the client belongs
to the domain Yahoo.com. It will respond sending HELO Yahoo.com. The server will again be sending back a positive
response that well HELO Yahoo dot com I acknowledge you. This means that the server has identified
or has registered the domain name from where the client is sending. And again the client uses the MAIL FROM command
to identify the sender of the mail. Here you can specify any add email address
you want. Well when I say any email address it is actually
any email address you can also write in place of the email address that we will specify. You can even write a say for example Bill
Gates at Microsoft dot com you can write anything out there. So if the syntax is correct the server will
respond by 250. Similarly after that there can one or more
receipt TO command. In fact in this example there are two where
you can specify the email addresses of the recipient. Again the server will be responding by 250
message. So now the sender and receiver of the mail
has been identified. So now let us (doubt) so now the client sends
command DATA. Server sends back an intermediate positive
response 350 saying that start mail input end with dot. Depending on the server this mail message
may differ slightly but actually the meaning is the same. This is intermediate because the mail transfer
is still in progress; the body of the mail is still being typed. So it is just an intermediate message asking
the user to type in the message and ending it with a full stop or a dot in the first
column. So after this the actual body of the mail
can be typed and finally a single dot in the first column will indicate the end of the
mail message. So at this point the server will respond by
a positive response 250. Then the client can send QUIT and the server
will finally send this message and close down the connection. Now the point to note here is that this is
a simple mail dialogue I have shown in fact anybody can use this command to have a connection
with the mail server which is available to you. For example in your organization if you have
a mail server, you can directly do a telnet to your mail server over port number 25 and
you can send a mail message by fabricating any arbitrary source address. Well if somebody is not that much familiar
with how this email headers and other stuff look like, they might get confused by receiving
a mail. For example if I get a mail message from a
say for example Bill Gates. But without knowing that it is really from
Bill Gates, but some of friend has played a prank on me. It is quite possible but if I look at the
email header carefully I will know the mail although it is saying it is coming from somebody
at Microsoft dot com. But in fact the mail originated from some
other place. In fact today when you have so many email
spam filters which are supposed to block and identify spam mails. They will in fact detect these kind of you
can say mischievous mails. Let me call it, but otherwise the SMTP does
not check for these kind of what should I say misuse, misuse of mail of transfer facility. So far we have talked about how mails can
be sent from one machine to other using SMPT. Now we are talking about mail access protocol. See here the idea is like this you have an
SMTP protocol running on a machine out here. This is your SMTP server, but I am sitting
on a computer out here. And here I may be having a software like Outlook
Express. I can establish a connection with the SMTP
server. And through this connection I can access,
I can read and I can send mails sitting on my computer. Only I need not have to login into this SMTP
go to the SMTP server and check my mails. There are in fact two protocols which allow
you to access your mails from a remote mailbox. The idea is that you have a mail server located
somewhere in the network. Your mailbox is present on that mail server,
you have a mail client, that mail client should have a capability of pulling your mails from
the mail server into your machine. So that sitting on your machine you can view
the contents of your mails. In fact there are 2 protocols which are used
for this purpose. One is called the Post Office Protocol version
3 or POP3 in short; the other one is Internet Mail Access Protocol version 4 or IMAP4. These are the two alternate protocols available. Let us look briefly at the capabilities of
these. The way they work as I had mentioned is like
this. This refers to an SMTP server mailbox. This is your Mail Client. So whenever the Mail Client wants to read
mails, it sends a request to the SMTP server to in fact not the SMTP server, to the machine
which is running the SMTP server and will be sending back the response in the form of
the mails and other information you want to have a look. Now in fact the SMTP server is one program
which is running. Suppose when you are trying to access a mail
through POP3, you must in addition have another server the POP3 server running on the same
machine as the SMTP server. So first let us talk about POP3. So for POP3 to work the client, POP3 software
must be installed on the recipient machine. Well in fact Outlook Express and the other
mail client they have an in-built feature of specifying which mail client you want to
have. And the server POP3 software is typically
installed on your mail server, the same machine which runs the SMTP server. So the client which is user agent here opens
a connection with the server on TCP port number 110. This 110 is the port number of POP3, then
the client authenticates server by sending a valid user name and password. And once the verification of the user name
password has been done the client can directly have an access to the mails sequentially one
by one. So the mails can be read by the client by
transferring them from the mail server one at a time one by one sequentially. Now when the mails have been read sequentially
one after the other, there are two different modes in which you can do so. One is the Delete mode. Delete mode says that as you are reading the
mails they are getting deleted from a mail server so that you are not unnecessarily occupying
the disk space of a mail server. You are freeing the disk space as soon as
you have brought your mails to the machines for reading. But the other option is that if you do not
want to do so, there is Keep mode where you can leave the mails intact in your mailbox. So POP3 has basically commands to Log in,
Log out, Fetch messages without deleting and Delete messages. So in the Delete mode, Fetch and Delete work
together in the Keep mode. Only Fetch works not Delete. So the POP3 protocol provides you with a basic
facility to access mail from a remote mail server to read mails into a machine so that
you can see them. But if you look at the mail clients that are
available today it can be Outlook explicit, it can Hotmail, and it can be Rediff Mail. All of them are based on the same technology. You will find that they provide. Many more facilities than just reading the
mails one at a time. So there has to be something more supported
by the mail access protocol than what POP3 provides and in fact IMAP provides you with
the answer. So in addition to the capabilities provided
by POP3, it has the following extra features. First an user can simply check the email headers. See in POP3, you did not have an option, you
have fetch the entire mail to your machine and then view them. But using IMAP you can simply fetch the headers
of the emails. You just view the headers which one you like
you just click on the header. Now the mail will get transferred and you
can view it. This is one option; the second is that the
user can also search the contents of the email for a specific string. Suppose I want to look only at those mails
which have a specific string as part of its body. So I can even do that in IMAP and you can
manage mailboxes on the mail server which was not possible in POP3. So a user can create multiple mail boxes delete
a mailbox rename a mailbox or also can create a hierarchy of mailboxes. Like here what I mean is that if you just
recall how the typical mail you can say mail utility systems work it can be Outlook, it
can be Yahoo, Rediff. See each of these mail servers or let us call
it mail systems. They have a facility for the user to specify
folders and subfolders. You can manage your mails in terms of multiple
folders you can move a mail into a folder; you can delete mails from a folder and so
on. So these kind of folder management is directly
supported by IMAP because IMAP supports it. Modern day email systems also have a facility
to support in the indeed support this kind of you can say management of folders through
the IMAP capability. Now we talk about Multipurpose Internet Mail
Extension or MIME. Let us try to understand why we have to go
for another protocol? Because using SMTP you can send mails using
POP3 or IMAP you can read mails on to a machine. The main problem we have till now is that
the SMTP protocol has a facility of transmitting only pure text messages, simple text messages
which are encoded as 7-bit ASCII. But anything beyond that SMTP is incapable
of handling. So you must have a mechanism through which
these kinds of non-text documents can be handled. So the first thing is that this SMTP cannot
transmit non-text messages. There are some solutions which were developed
earlier as adhoc solutions. For example in some UNIX system there was
a command called uuencode. This converted a non-text messages into a
text equivalent. But it was not accepted as standard. This existed only between UNIX systems. When you want to transmit something between
UNIX systems you could do this moreover if you have text that includes international
characters like this special symbols which are typically encoded in 8-bit ASCII. Here also you are your simple SMTP will not
be allowed; will not be able to handle this kind of text. So for these kind of problems, the protocol
MIME was developed in fact MIME can handle something more like. Many servers may reject mail over certain
size. So the MIME protocol also has a facility of
breaking mails into smaller pieces. Moreover some SMTP implementation do not adhere
to standards like, even if it is a text, some SMTP implementations will remove actual word
spaces like tabs and other things, replace them with single space. So your mail formatting will get lost when
it reaches its destination. So MIME tries to keep this mail formatting
intact in addition. Now let us try to understand what is MIME
is it a new protocol in the same sense that SMTP is? See SMTP is a mail transfer protocol it is
a program which is running as a server the client can contact the server and can start
a message exchange date. But you remember MIME is a little different
from SMTP in this regard. MIME is not a program which is running as
a server and someone can contact with MIME and can send and receive something. MIME is not running as a server as a Daemon,
rather MIME acts as translator. MIME sits on top of SMTP, if it is non-text
mail MIME will try to translate it in some way into an equivalent text version along
with relevant information. So that the final receiver can decode it back
and there it can forward the message, the translated message to SMTP which can transfer
it. This is the basic idea behind MIME. Now in order to have we must have some additional
information. In the translated version I told you, so these
additional information are present in the form of 5 new message header fields in the
mail messages. There is something called MIME-version, Content-type,
Content-transfer-encoding, Content-id, and Content-description. They have their own medium; without going
into details of these I shall some examples little later. See these content type actually means what
type of content it is? Whatever is following? Whatever you are transmitting is it an image,
is it a word document, it is a PDF file? What it is so it specifies the type of the
content so you can have types answer subtypes? For example it is an image subtype is it a
jpeg image. Sib type can also be a gif image so on. This transfer encoding tells you that what
kind of translation you have done. There can be more than one choice available. So the receiver must know about the translation
you have done. So that it can be basically translate it back
to the original form. So some of the context types some of content
type examples are given here. It can be a Text body; it can be a Multipart
in indicating that the mail has multiple parts with the subtypes being Mixed Parallel Alternative
Message. Several subtypes Image, Video, Audio, Application
and there are several others this is an example the way it is written. For example Image, you give this the type
a slash followed by the subtype. This is how you specify the type and subtype. MIME Transfer Encoding specifies how the translation
has taken place which in technical term is called how the mail body is wrapped for transmission. Because mail body has to be translated or
transformed somehow into a text form. So that SMTP can transmit it this is mandatory. Content transfer encoding can have 6 possible
values. Out of them 7 bit, 8 and binary these are
special cases. There are no separate encoding required for
these because these header, these content transfer encoding type directly tells you
what kind of information it is. If you want to send them in plain binary you
simply send them 7 bit, 8 bit similar. There is one method called Quoted-printable
or quote-printable. Here you use this for documents which contain
mostly simple ASCII characters with a few non-printable characters. These non-printable characters can be represented
by a quote followed by their hexadecimal equivalent. That is why these are called quote-printable
if you have a special character give a quote followed by 2 digit hexadecimal equivalent. Base 64 is perhaps the most widely used technique
for transforming arbitrary non-text information into a form which can be handled by SMTP. So it maps arbitrary binary input into printable
output we shall see how this works and X-token is a method where the user can specify its
own nonstandard encoding method. Now to talk about the Base 64 encoding method,
it works like this. Base 64 encoding method can transform any
non-text mes method. Message into its equivalent text message. Now the way it works is like this, it takes
3 bytes of the input message at a time. It breaks up into 4 chucks of 6 bits. So the 3 bytes of message is broken up into
6 bits each. Each of these 6 bits are expanded into 8 bits
through a simple table look up process. The table look up process is called Are 64
it works like this. See in 6 bits you can have 2 to the power
6 or 64 combinations. So from 0 0 0 0 0 0 to 1 1 1 1 1 1, there
are 64 combinations. You have a table which says that for the first
26 of these you replace them by the by the ASCII code of the letters capital A to capital
Z. For the next 26 we replace them by their code
of small a to small z for the next 10. The digits 0 to 9 and for the last 2, plus
and slash, these are the 64 characters. So each of these 64 combinations you encode
them as one of 64 ASCII characters. This is how you encode it and replace each
ASCII character by its 8 bit ASCII equivalent. So 24 bit gets translated into 32 bits. So this means an expansion in message size
by 33 percent. So some examples of MIME header. So just you can look there are some header
information. There is a boundary which contains a string. The string wherever it occurs it indicates
some boundary in the mail message. There can several such parts in the MIME header. This is just another example. Here are also there are some header, here
the boundary name is boun 42. Here there is a boundary, here there is boundary,
here there is boundary. So the Content-type for example, multipart
alternative means there are multiple parts in the content and alternative means the receiver
can choose any one of these two alternatives. Depending on the capability the first one
says Content-type text plain. The second one says text enriched so if your
receiver mail client can display rich text there is a second alternative can be chosen. Otherwise the first alternative now I suggest
you can look at the headers of some of the MIME encoded messages to understand more about
how the encoding and the attachments works? The attachments when you send with a mail
they get encoded in the same way. Now with this we come to the end of today’s
lecture. Let us now quickly look at the answers to
the quiz questions which were posed as part of the last lecture. Solutions to quiz questions on lecture 9. The questions were: What is an iterative server? It is one in which the client requests are
processed one by one sequentially. What is a concurrent server? Here multiple client requests can be handled
at the same time. Multiple copies of the server are created
using process system call like or you can also use a thread for this purpose. Which of client or server must start first? Obviously the server. What are the components of a socket? It will be Protocol local IP local port or
Protocol remote IP remote port it is also called half-association. What is a domain? It is a set of computers which are related
either by their geographic location or by the way they function. What are named and nslookup?
named refers to the domain name server which runs on the UNIX system and nslookup is a
command which invoke the DNS name resolver through nslookup you can actually translate
a name into an address. How does recursive name resolution works? Host sends a request to a DNS server. This DNS server recursively forwards the request
to other DNS server the responses are sent back to the initial host along the same path. How can you connect to a server xyz.com over
port number one twenty 1234? By a command like this. Telnet server name followed by port number. Why does ftp use more than one port numbers? The first port number 21 is used for control
connection and the other one is used for actual data transfer. What are the functions of the hash and bin
commands in ftp? This hash indicates the status of the file
transfer. It displays some hash symbols. Continuously each hash indicates a chunk of
data transfer either 2 kilobyte or 4 kilobyte or 8 kilobyte depending on the system and
bin selects that you are trying to transmit a non-text file a binary file. Now some questions on today’s lecture. What are the basic drawbacks of SMTP? Which port number do SMTP servers use for
accepting client requests? Why does MIME does not have any port number
associated with it? Under what conditions can a SMTP server also
act as a mail client? What are the purposes of the MAIL FROM and
receipt TO commands in SMTP? What is difference between Cc and Bcc in the
SMTP header? This is something I suggest you can look up
and find out the solution. Why is IMAP preferred over POP3? A message of size 3000 bytes is encoded using
Base64 scheme. What will be the size of the encoded message? Is it mandatory for the DNS server to run
on the same machine that runs the SMTP server? How are mail attachments handled in MIME? Now with this come to the end of today’s lecture. In our next lecture we shall be starting our
discussion on the World Wide Web which is yet another very popular application that
we have on the internet. Thank you. Preview of next lecture. In the last class we are talking about electronic
mail which is one of the most widely used applications on the internet. Today we will be starting our discussion on
another very popular application which i am sure all of you are familiar with. You must have you must have used it in practice
and it is the World Wide Web. Now this World Wide Web has become so popular
that today World Wide Web and internet, they are used synonymously. Sometimes this World Wide Web or www is meant
to be referring to the internet or vice versa so this is an impact of the popularity of
this www.

5 thoughts on “Lecture -10 Electronic Mail

Leave a Reply

Your email address will not be published. Required fields are marked *