Sam Jackson
The concept of a packet switching computer network was first conceived in 1962 and implemented in 1965 with the Q-32 network. However, the more significant program that began what we think of the internet today was started in 1969 at UCLA and was called ARPANET. It started expanding on the west coast and moved east. NCP protocol that allowed for applications was implemented in 1972. Also in 1972 was a push for an open architecture design where independent networks (radio, satellite) could also communicate with ARPANET. This required a new protocol and Kahn began to develop TCP/IP. In 1973 xerox invented Ethernet. As ARPANET grew, DNS was invented as there were too many addresses to remember. Routers begin to have problems because there are too many types of networks and routing tables get bigger. To fix this, IGP and EGP protocols? are invented (IGP for regional, EGP to connect). This helped because now each router did not have to be constantly updated to accommodate the expansion. ARPANET converts to TCP in 1983. Through the 80's the internet was all funded by the government and academia, and its use was limited to educational and military purposes (though email was an important application from the beginning). In the 80's federal policy changed to accommodate commercial uses of the internet. The NSF encourages commercial use but only on the regional level, forcing private companies to develop their own cross country infrastructure. This worked well and by 1995, the NSF backbone was defunded. There was great collaboration in late 80's and early 90's among computer scientists and vendors to make the internet a commercial success, and soon the vendors began to contribute to the technical expansion of the internet as well.
QOS, or quality of service, is an important issue regarding internet communication today. Telecommunications companies and others want to assign priority to packets under the pretense that it will reduce congestion and allow the more important packets to get through easier. The author clearly disagrees with this idea and gives the analogy of a crowded road to help understand the issue. Assigning priority to certain packets would be like having firetrucks or police on a road, ignoring traffic laws and taking precedence. The problem with this, as the author argues, is that 1) it is ambiguous as to which packets deserve priority and 2) there is only a narrow range of congestion that this system would actually improve performance. He suggests that resources would be better spent increasing system capacity rather than trying to create complicated algorithms which may or may not actually increase efficiency, and I agree.
Net neutrality is the idea that the owners of the physical network of the internet, the telecommunications companies, should not be permitted to charge sites for the quality of service that they receive, or on the basis of how much bandwidth they use. The arguments against this are numerous, but the overriding fear is that, without net neutrality legislation in place, the telecommunications companies could reshape the content of the internet into a corporate mess, where only sites that could pay and were approved by the isp's could be viewed. The telecommunications companies argue that with the amount of bandwidth being used increasing faster than ever before, they need some new way of paying for faster networks. They think that it is only fair that data heavy sites that are causing the increased congestion should pay more than sites without things like streaming media. For example, one executive said that he would consider asking itunes to pay 5-10 cents extra per song downloaded. Net neutrality is a hot issue today because the Telecommunications Act of 1996 is soon to be rewritten, and lobbyists from both sides are flooding Congress. The biggest worry for net neutrality advocates is that while big companies will be able to pay, little start ups will take the hit, taking their innovation with them.
As it stands now, the internet is an end-to-end system: all the control is held by the users and producers, not by the connecting middle-men. If this were to change, and net neutrality was not enforced, it would be like putting up tollbooths all over the internet, slowing down traffic for the smaller, shallow-pocketed sites and speeding traffic up for deep-pocketed corporate sites. Timothy Wu calls this concept of allowance the "Tony Soprano business model" where the telecommunications companies exhort protection from everyone, with the potential to block or slow websites that don't pay up. As 60% of web content is run by regular people rather than corporation this screws over fresh innovation and production that has in the past come from these regular people. Telecommunications companies are spending tens of millions on lobbying and creating fake grassroots movements because they are greedy.
According to economic theory, markets won't work efficiently unless prices fully reflect the costs of all products sold or services delivered. On that basis, if the public wants faster, better internet service, then the price of sending data-heavy content over the web should be accounted for. To build the next generation of networks, someone has to pay the tab, and it is hardly fair that sites like youtube, the source of the problem, should pay the same as some blogging site would. If net neutrality is enforced, then the charge could be so great that only corporations would be able to pay, and then there would not be enough revenue collect to build new networks. I am skeptical of this article.
A hot topic in recent years has been governance and control of the internet. Though some founders of the internet would argue that the web cannot be controlled by a single centralized source due to its inherently decentralized nature, the debate still rages on, carried out mostly by uninformed politicians and bureaucratic types. At the center of the tug of war is ICANN (Internet Corporation for Assigned Names and Numbers), an group founded in 1998 that has responsibility of four technical domains: 1) Operating and making policy choices regarding group top level domain names .com, .uk, .edu and so on 2) Managing the assignment of IP addresses (making sure no two computers have the same one) 3) Managing the root name servers 4) Establishing tech standards that maintain the internets interoperability ICANN was created to take over the responsibilities of one man, John Postel, one of the founders of the internet, who had been essentially managing this entire system himself. It is composed of a 15 member board of directors, 6 of whom represent subgroups dealing with specific policy areas, the other 8 being independent public advisors. There are also a few other groups included, significantly the Governmental Advisory Committee (GAC) which is composed of representatives of world governments. ICANN is private sector non-profit business, though it is technically responsible to the attorney general of California and Bush said in a statement in 2005 that the US government plans maintain its official oversight of ICANN indefinitely. For many, this is the heart of the dispute. Many governments around the world are unhappy that the United States maintains control over what has truly become a world wide web. These governments are pushing for the abolishment of ICANN, to be replaced by an international, UN affiliated group management. They often cite the International Telecommunications Union, which handles telephone communications worldwide, as setting precedent for such an organization and many think the ITU should absorb the duties of ICANN. There are numerous problems with this argument. First of all, ICANN is involved primarily in the technical details of internet management, with the policy decisions that it makes limited to essentially one function- determining what new top level domains to introduce. In one of the readings, the example of the .tw (Taiwan) domain level was discussed. ICANN created this domain, and soon after the FCC received angry complaints from Chinese ambassadors demanding to know why Taiwan got a domain (as China doesn't recognize Taiwan as a country). Luckily, as ICANN is a private company, the FCC and US government could skirt the issue and ICANN got away with it because it is a privately run company. Imagine this same issue if it came up in a UN controlled bureaucratic forum. It would take weeks of arguing over political boundaries and policy choices that would slow down the growth of the internet and complicate it with unnecessary bickering. Furthermore, the domain name choices that it makes are strictly governed according to a policy code called UDRP which was written in conjunction with the WIPO (world intellectual property organization), a branch of the UN concerned with the protection of intellectual property. Secondly, the argument that ICANN is a US controlled institution that should be internationalized are far off base. ICANN is composed of citizens of more than 80 different countries and is advised by representatives of nearly every world government. Of the board of directors, only 4 of the 15 members are American. ICANN is also only very loosely tied to the US government. Under a 2006 agreement signed by president Bush, ICANN's responsibilities were updated as follows: "- ICANN will no longer have its work prescribed for it. How it works and what it works on is up to ICANN and its community to devise; - There is no requirement to report regularly to the DOC. The DOC will simply meet with senior ICANN staff from time to time." http://www.icann.org/announcements/announcement-29sep06.htm Indeed, even before this agreement, ICANN was not unduly influenced by the US government. Back in 2005, there was a big push in Congress for all porn sites to be relabeled under the top level domain .xxx, and there was a direct request made to ICANN to institute this new domain. ICANN rejected the proposal. Finally, many countries want more control over the their personal top level domains and are wary of so many of the root servers (10 out of 13) being in the US. Both of these concerns have been addressed. ICANN is taking steps to give contries more control over their domain names. It is also not quite true that the 10 servers are located in the US. While there are 10 IP addresses of DNS servers in the US, there are actually far more physical servers in Europe. This whole argument boils down to a few countries wanting more control over the internet, and ironically the few countries that seem to push for it the most (ie China) and argue for a democratic system are the ones who restrict it the most in their own countries. ICANN is a privately run organization and that allows it to escape the bureaucracy of government run institutions and continue to keep up with the innovation and fast growth of the internet today. Extra Sources: http://www.icann.org/announcements/announcement-29sep06.htm http://en.wikipedia.org/wiki/ICANN http://en.wikipedia.org/wiki/Root_nameserver
The Digital Millenium Copyright Act (DMCA) forms the basis for protecting and restricting media and technology on the web today. There are many controversial points to this act, one of which being that it gives more copyright protection to technology designed to protect copyright than it gives to the technology itself. As this author puts it, it protects code that protects copyright. The problem with this is that it interferes with legitimate manipulations of the code and also presents security issues, where any potentially constructive criticism of the weaknesses of a security mechanism could be taken as a federal offense. The law should help the owners of creative content wherever possible, but to push beyond the protection of copyright is going too far. Youtube is at the heart of a firestorm of litigation coming from copyrighted content owners demanding that their videos be taken off of the site. It defends itself based on a clause in the DMCA that says users that did not have "knowledge of infringement and did not expediously remove" copyrighted content are not liable. This is not available to anyone who derives financial benefit from the hosted content. The case against youtube is that its only function is to host video, and that all of its revenue comes from advertisers. Without the videos that may or may not be copyrighted, there would be no traffic and hence no financial benefit. To be fair, youtube does monitor its content and remove things that are objectionable or copyrighted. The problem is that owners of creative works shouldn't be responsible for monitoring the entire web for infringements on their intellectual property. DRM is code that protects downloadable media content from illegal sharing or other copyright abuses. It also places heavy restrictions on what people can do with that media content. In one camp, business people argue that the DRM is key to consumer choice and allows content owners to sell their products in a variety of different ways- be it one time use or for permanent use. The only problem with it now is that the technology is not perfect. On the other hand, pundits argue that DRM restricted media makes culture and technology poorer. It doesn't let any transformational changes to be made to the content, for example tivo would not be allowed in a digital setting. Also, it discourages everyone but big businesses to make technical innovation because all of the smaller guys fear lawsuits over their new products. The example of DVD players was made- they haven't changed in nearly a decade. The other side argues that DMCA is actually a stimulus to technology because it lets content owners put out material without fear of massive pirating. The retort is that commercial pirates have the technology to do it anyway and it really only limits the consumers use of something they rightfully bought and own. According to this article, the DMCA violates the first amendment, the right to free speech, by outlawing discussion or criticism of security software, in this case Microsoft's Kerberos program. On a online forum, posts were made criticizing the program and also offering info about how to circumvent some aspects of the software. Microsoft asked that the website, Slashdot, remove all posts on this topic and is suing. This is a freedom of speech violation because it allows Microsoft to squash any criticism of its products or the way it conducts business, no matter how valid that criticism might be. Not only that, but it can ask any party to remove content just based on the threat of a lawsuit, even if it has no legal basis, scaring people into removing any contrarian commentary.
According to this article, bittorrent takes up to one third of the internets bandwith. Whether or not this is true and where this is true is questionable, but the point remains that it is a very relevant topic. Within the bittorrent software there are a number of terms that need defining.
Peer Set: the peers that a user is connected to
Swarm: the number of people simultaneously downloading the same file
Leechers: those downloading the file (note that they simultaneously upload)
Seeders: those that have already completed the download
The downloading of files usually takes place in 256 kB chunks. This article mainly deals with international law regarding reproduction rights, a bunch of legal jargon, and tries to determine if current laws can be applied to bittorrent. Conclusions reached are as follows: reproduction rights to be determined by member states can be applied to bittorrent. It is arguable, however, that bittorrent entails acts of communication to the public. It is also clear though that BitTorrent infringes upon the copyrights and related rights of the right holders. The problem brought up at the end of the article is an interesting one, stating "[D]espite the aim of the WIPO Treaties to assure effective and uniform protection of the rights of authors, technology may continue to outstrip the ingenuity of the drafters…"
The second article deals at a national level rather than an international one. The most interesting part to me at least were all of the current adjustments to copyright law that reach beyond blatant, or direct, infringement. The first is Inducement and is split into 2 parts. "1)Affirmative Act: The accused inducer has made statements or taken other active steps directed at encouraging infringing uses. 2) Intent: The accused inducer intended to promote copyright infringement." The second is contributory infringement, again divided into 2 parts. " 1) Knowledge: The accused contributory infringer knew of the underlying direct infringement. 2) Material Contribution: The accused contributory infringer caused or materially contributed to the underlying direct infringement." The third is vicarious liability, 2 parts again. "1) Right and Ability to Control: The accused vicarious infringer had the right and ability to control or supervise the underlying direct infringement. 2) Direct Financial Benefit: The accused vicarious infringer derived a "direct financial benefit" from the underlying direct infringement." In addition, 4 major defense strategies were discussed. 1) The no direct infringer strategy- basically claiming no one directly infringed and thus no one can be held liable for any of the above accessory charges. 2) Betamax defense- the Betamax defense references a precedent set in 1984 in a Sony case about vcr's where they got off of charges because the product was "capable of substantial noninfringing uses." 3) DMCA safe harbors: this sounds like a poor defense. The safe harbors are so narrow that it doesn't seem like anyone could get away with it.
Between facebook, myspace, and the plethora of other social networking sites used by college students today, a lot of personal information can be found online. After a postegame riot at an OSU game, college security used facebook to identify offenders and enact disciplinary measures. Taylor Behl was a freshman at Virginia Commonwealth university who was murdered; her social networking sites were combed for clues on the case and contained a surpising amount of personal information. This makes cyberstalking a real danger. Some colleges even block facebook from their servers, for example University of New Mexico, which cites security conerns. Prospective employers and campus security could potentially use it to gather information. I think that all of this is important for people using these sites to understand, but I also think that any great fear of them being manipulated like this is overblown because the fact remains that you control what you post as well as who can see it through your privacy settings.
Facebook news feed was a controversial new addition to facebook that alerted you of any activity that any of your friends made with their facebook accounts that still exists today. This author is adamantly opposed to it. She argues that it makes information too readily available to too many people who might do better not seeing it. She gives the analogy of being in a room at a party and having a conversation, when all of the sudden the music stops and everyone hears what you just said rather than whoever you were talking to and one or two people who happened to overhear. It also allows people to pick up on thigns they might not have before and changes the relationships that you have with people, often over misperceived information. Furthermore, this just adds to what is becoming a bigger problem: the fact that people are making cyberfriends rather than real friends. Gossip is like grooming, and its not the same when you are behind a computer screen rather than sitting face to face with someone, especially because it can be completely one sided. To me, I have begun to lose interest in facebook so I don't spend much time looking through it. The news feed makes it easy to see if anything interesting has happened recently so I usually just glance at that before going to a different site. Therefore, I think this article was an overreaction written by someone who probably spends far too much time within these social networks themselves.
According to this article, 90 percent of college students have a facebook profile. This makes it fairly easy for colleges to monitor students and get a good grasp on their social and personal lives. The question arises, should colleges be using facebook as a tool to crack down on illicit activity? If the illegal activity involves someone under 18, and the college knows about it, they are obligated to report it. However, it is almost impossible to monitor facebook when so many people are posting so much information. Also, courts have held that students are personally responsible for what they post and not the college network that they operate on. Therefore, I think it is none of the college's business to go combing through people's personal lives and that they have no good argument for doing so.
Economics and security have more in common than one might think. According to this author, systems often fail because of misplaced incentives: the people who could protect the system are not the ones that suffer the consequences if it does. For example, hospital medical records have in depth and detailed billing management systems, but bad privacy for patients. In England, ATM security was not up to par because poor regulation merely allowed banks to pass on the cost to their customers. Many intrusions go unreported because it is against the economic interests of a company to do so. Much like defense spending though, it is hard to find a balance of spending too much or too little on security. And it also brings up other issues of the motivations of the companies behind the security systems. For example, Microsoft's Trusted Computing- is it providing more and better security or is it merely another way of enforcing their monopoly of the PC? It is tough to say.
Apparently, not everyone thinks spyware is a problem. Spyware is defined as software that ads itself without user permission and does not alert users the information it is gathering or its intentions. Adware is similar except that it tells the user these things. Marketscore is a piece of adware that this article discusses that comes with a bit of freeware known as iMesh. Marketscore essentially has access to every single bit of information that you send out on the internet, not excluding bank passwords, emails, and even down to if you use shortcut keys. Yet somehow people rationalize that without Marketscore they wouldn't be able to get their iMesh software, and that it is an inevitable tradeoff. These people are foolish.
Spyware is not only annoying and slows your computer down, but it also poses grave security threats to businesses and government. Spyware can infect a computer through piggybacking on another program or just download itself as one surfs the web. It can be used to: track information, open a computer to remote access, log keystrokes, and steal files. Furthermore, adware slows boot times by up to 3.5 minutes and web performance by nearly 5 times, transferring lots of unknown data in the process. It could easily be used as an espionage tactic in the business or political world. To confront this problem, the author suggests a combination of 3 tactics: education and protection, disclosure through legislation, and active prosecution. The first and third mean exactly what they sound, the second suggests that the current legislation be changed so that all applications should be readily identifiable prior to installation and made easy to remove.
With the internet evolving into something far different than its creators ever envisioned, it is up to the current technologists to guide the direction of the internet in such a way that maintains all that has made the internet what it is today. Unlike in the early days of the internet when it was only used by a few academics and for military purposes, today there is no sense of common purpose among those involved. There are ISP's, and customers, music makers and music aficionados, hackers and grandmothers. The internet has become a mirror image of the society in which it operates, constituting constant struggles, or tussles between adverse groups of people. The challenge, then, is to update the technical architecture of the internet to make it faster, easier to use, and more open to innovation while at the same time preserving the goals of scalability, reliability, and evolvibility and also not favoring one side of the tussle over the other. The design must accommodate for a variation in outcome, and allow the tussle to take place within the framework of the design. Furthermore, the design must be modularized, so that each individual tussle does not spill over into other domains. In other words, the technical architecture should dictate the playing field, not the outcome of the tussle.
Also, the future internet must preserve choice for its users. For example, end users should have their pick of ISP's rather than a monopolized system. With the current advances in technology, it seems that the ISP's will be whittled down to only a few large competitors, the ones who can afford to lay and upkeep the new fiberoptic network. This obviously is cause for concern.
Another major change from the early days of the internet is the current lack of trust between users. Firewalls right now slow traffic down and lose packets, they should be designed and operated on the basis of who the user is communicating with. To make this a reality, users must know with whom they are talking, which means instituting a framework to certify identity. The biggest challenge to developing this framework is the current status quo where people consider it permissible to be anonymous on the internet. People feel that by giving up their anonymity, the internet will lose its openness. And it is true that oppennes has been the biggest component of the success of the internet as openness often equates to competition.
In addition, the loss of trust calls for less transparency, which violates another founding principle of the internet, end to end arguments. This principle operates on the basis that the core of the network is transparent, that is packets go in, packets go out, but are not interfered with in the core process. If the core were to integrate specific applications, it would inhibit the development of new applications and could also hurt the reliability and robustness of the network as a whole.
In conclusion, the evolution and enhancement of existing applications is inevitable, and in the spirit of openness, the best thing to do for up-and-coming applications is to bias the tussle in their favor. Keeping the net open and transparent for new applications is the most important goal. To deal with the issue of trust then, encryption schemes need to be made easy to use and accessible to consumers for implementation at their own discretion.