Improve the Internet? Me?

Thinking back to the first time I used the internet makes me feel kind of old, I remember chatting over ICQ (I seek you) and downloading music from Napster (before everyone found out it was illegal). Now, I love streaming Hulu (the free version), learning fun facts from Wikipedia (even if the validity of the information is questionable), and socializing over Facebook (since my parents haven't figure out how to join it yet). Reading through the article "The Design Philosophy of the DARPA Internet Protocols" forced me to reflect on the birth of the internet, and it's rise over the past few decades and the impact it has had on the world.

DARPA's main goal was to "develop an effective technique for multiplexed utilization of existing interconnected networks." Essentially, DARPA had several networks they wanted to combine to make an "interconnected network" or "internet." DARPA was not thinking on a worldwide scale at the time, but this vision of connecting networks over large distances was an important step towards the establishment of the World Wide Web we see today. The main goal in the creation of the internet was to improve communication channels between different military groups throughout the country, with such a strong focus on sending messages to one another it is no surprise that email and social networking have become so popular today.

The question of why this communication was so important, more important than the survivability or security of the internet, could have many answers and truly depends on the issues facing those in charge of developing the goals, which I unfortunately have no real insight into. Perhaps a more poignant question then is why study the history of the internet at all? Is it to understand the motivations of those who created it? Or is there a deeper purpose?

My networking professor would have me believe that we study the history of the internet in order to improve the future of the internet. Should we not learn from the mistakes of others, so we do not follow in their footsteps? As an undergraduate I may have eaten up my professor's words and then spat them back out during some mid-term or final, never again remembering their importance once the course was over. As a graduate student, however, I am more inclined to dig a little deeper and try and determine if that truly is a role I need to play: an internet improver. I'm not sure, my knowledge of networks is minimal, but hopefully the more I study them, the better equipped I will be to step into those shoes.

If I were to become an internet improver, I believe that the best place to start, of course, is how we can improve upon what has already been done. Where did DARPA go right and in which areas did they fail? While a full analysis is not possible, we can look at a few of the goals outlined in the aforementioned paper:

1) Connecting existing networks, which I believe was mainly in order to improve communication. With sites and applications like Google Voice and Facebook, can we be any more connected? I postulate that we can, especially due to the emergence of smartphones in recent years. Since video and voice transmission through the internet is already possible, why am I paying for minutes on my cell phone? Why can't I just have an unlimited data plan and then make calls and video chats via the internet on cell phones or computers? I assume it is possible, but would changes need to be made to the internet architecture and protocols to make it feasible? Are there other obstacles stopping this from being possible? An important question whenever suggesting changes would also be how would these changes affect us in the future? DARPA made the mistake of not thinking on a global scale, forsaking such measures as security for connectivity. Should I, then, be thinking about networks on an interplanetary scale?

2) Survivability of the internet, a topic which at first seem pretty much taken care of. Going back to our interplanetary scale, what would happen if the connections between two planets were lost? The internet would survive, but if there were a substantial number of connections on other planets, then the internet could essentially be split. What kind of satellite technology is needed to ensure that doesn't happen? Of course, just as there is a problem with thinking on too small of a scale, I also believe that thinking on too large of a scale could be detrimental, or at least a waste of time. Should I be worrying so much about the problem of interplanetary connections when its inception could be many lifetimes away?

3) The final goal I will discuss (although there are more goals outlined in the paper) is the ability to support multiple types of communication services. Since the internet is so large, it isn't easy to simply introduce a new protocol for streaming video or voice or other necessary services. Can we improve on TCP/IP or UDP? Surely we can, but, from the little that I've learned, apparently it is not feasible. How then can we improve these services? Is it possible to improve the overall internet architecture by introducing new protocols in the application layer? Or will we still be bottle-necked by the limitations of the existing protocols that have cemented their position in internet usage?

Hopefully as I continue to study the architecture of the internet and learn about networking I can begin to answer these questions and better understand these concepts. Who knows, maybe I can even make some small contributions that will help improve the internet...