I consider myself a wizard. Every day, I walk around with a magic wand in my pocket. This wand grants me many different powers. I’m never lost and can find delicious food anywhere I go. I can instantly contact people across the world. Amazingly, I can even pause time and replay it.

Even though I think I’m a wizard, I’m still an ordinary person. Of course, my magic wand is just my smartphone. I can get directions via Google maps and search for restaurants on Yelp.  By taking a photograph, I can pause time, and by watching a video, I can witness history. Finally, I can communicate in almost too many ways: phone calls, video chats, ordinary text messages, or one of a zillion text messaging apps. For more formal communications, I might send an email.

Although email is one of the oldest magic powers of my phone, I find it the most intriguing. You can almost imagine other communication technologies being implemented via email, even though they don’t work that way. When I request directions through Google maps, there could be a person on a computer waiting to email directions back. There might be two individuals working for Yelp: one receiving emails and compiling restaurant reviews, and another sending reviews and recommendations to those in need. When I want to watch a video, a Netflix or YouTube employee could be tasked with sending me thousands of emails, each containing an image or video frame (watching such a video might not be pleasant).

By contemporary standards, email seems so old and ubiquitous that many people no longer consider it magic. Rudimentary forms of email have existed since the 1960s, although they would be unrecognizable now⁸. For context, the Internet, which allows different computers to communicate, was created in 1983, and the World Wide Web (different from the Internet!), which allows you to visit websites, was invented in 1990^{1, 6}. I’ve had access to email since I was ten — nearly half my life. Yet the longer you think about it, the more magical email becomes. Right now, I could email my extended family living in India. I click “send” now, and poof, my relatives can read it seconds later.

How did my email get all the way to India? The Internet, you might reply. But really, how exactly did my email get to India? It’s not like my laptop is physically connected to my relative’s laptop. Even an explanation consisting of terms like electrical signals, Wi-fi, fiber optic cable, and TCP/IP leaves too much to magic. Let’s consider an email’s journey, from start to finish.

My email message originates as an idea within my head. Through my laptop and keyboard, I convert the idea in my mind into words on my laptop. This is done through an email client. In the recipient box, I enter shashvatscousin@gmail.com. I click “send”, and my email’s journey begins.

My email’s first stop, an Internet modem, is similar to a mailbox; any data sent or received stops at the mailbox before being picked up². The email can get to the modem in two ways. First, I could physically connect my laptop to the Internet modem through an Ethernet cable. My email would then travel through the cable to the modem. The more complex method is Wi-Fi. With Wi-Fi, the email is converted to electronic signals and sent to a Wi-Fi router through radio waves; your router is basically a small radio⁵. The Wi-Fi router is physically connected to the modem; usually, the router and the Internet modem are sold in one device⁴.

Naturally, since I put my email in my mailbox, my message needs to find its way to a post office. For email, this post office is called an SMTP (Simple Mail Transfer Protocol) server, which is basically another computer⁷. My email address is shashvat@mit.edu, so this computer would belong to MIT. If someone were using Gmail, the SMTP server would be owned by Google. Recall our modem, which connects to other modems and the SMTP server via cables, or roads in our analogy. These roads allow the email to get to the SMTP server.

The SMTP server is responsible for delivering my email to my cousin. First, the server needs to know where to send it⁷! Recall that my cousin’s email is shashvatscousin@gmail.com. The part after the @ symbol specifies who operates the receiver’s email service. Someone needs to own the computers that run my cousin’s email, and that owner is Google. The SMTP server then contacts a DNS (Domain Name Service) server, another computer⁷. Basically, when we send our email, we specify the recipient and their email provider, but we don’t specify where they live. The DNS server looks up the owner — gmail.com — converts that information to numbers called an IP address, and returns those numbers to the SMTP server. The SMTP server is now able to send the email.

We have encountered two servers already, so it helps to explain what a server is.  A server is just a computer running somewhere that provides services for people or other computers. For example, the DNS server serves computers’ IP addresses. The SMTP server serves people by sending their emails. A server is simply a computer that serves.

My email is then sent to an MTA (Mail Transfer Agent) server⁷, my cousin’s electronic post office. However, my email’s journey to the MTA server is quite perilous. Until now, we used “snail mail” as an analogy and assumed that my email corresponds to a letter or envelope. However, my email is more like a massive stack of papers. For transport, the email needs to be broken up into small packets of information, which correspond to individual pieces of paper. The method that splits the data into packets is called TCP, or Transfer Control Protocol³. The packets then travel from computer to computer until they reach the MTA server. IP, or the Internet Protocol, is responsible for giving the packets directions³. Otherwise, the packets would know where to go, but not how to get there. Finally, once packets arrive at the MTA server, TCP is used to reassemble the packets into my email³. TCP and IP, collectively called TCP/IP, form the backbone of the Internet and are thus called the Internet protocol suite³.

Now, sending packets through the Internet is a messy process. Like roads, cables can get congested, so different packets might take different paths to their destination. Sometimes, packets can get lost on the way. Luckily, TCP keeps track of this! The MTA server and the SMTP server send packets back and forth until they are sure that the MTA server has received all the messages³.

Finally, the email is delivered from the MTA server to my cousin. The MTA server is responsible for sending my cousin his emails when he asks for them. Modern email typically uses IMAP (Internet Message Access Protocol). Basically, all of my cousin’s emails are stored on the MTA server, and he can access them however he chooses. If my cousin wants to see all emails with the word “book,” the MTA server will send my cousin’s email client all such messages. However, my cousin simply wishes to check recently received emails, and the MTA server will comply.

At last, my message arrives onto my cousin’s email client. What should have taken days has occurred in mere seconds. Photons and physics happen, sending the text of my email to my cousin’s mind, and neurons are fired. My cousin finally reads my email.

However, the journey is not quite over; now, my cousin thinks of a response and begins to write. An email starts to make its way back towards me.

 

 

References

1. Andrews, Evan. Who Invented the Internet? 18 Dec. 2013, www.history.com/news/who-invented-the-internet.
2. Definition of DSL Modem. www.pcmag.com/encyclopedia/term/dsl-modem.
3. Definition of TCP/IP. www.pcmag.com/encyclopedia/term/tcpip.
4. Definition of Wireless Router. www.pcmag.com/encyclopedia/term/wireless-router.
5. Escobar, Eric. How Does Wi-Fi Work? 15 July 2015, www.scientificamerican.com/article/how-does-wi-fi-work/.
6. A Short History of the Web. home.cern/science/computing/birth-web/short-history-web.
7. Trivedi, Yatri. How Does Email Work? 22 Sept. 2016, www.howtogeek.com/56002/htg-explains-how-does-email-work/.
8. Vleck, T. Van. “Electronic Mail and Text Messaging in CTSS, 1965-1973.” IEEE Annals of the History of Computing, vol. 34, no. 1, Jan. 2012, pp. 4–6., doi:10.1109/mahc.2012.6.

 

Back to the Table of Contents