A post for last week...

Last monday I did not understand why the internet is not rhizomatic. The internet connects many disparate people with much disparate information; it appears to be free of hierarchy, due to the fact that that any site may link to any other; it appears to allow free movement along hyperlinks, through searches, and with direct reference to URLs; it includes both “lines of segmentarity” in websites, hyperlinks, and databases and “lines of deterritorialization” in RSS feeds, tag clouds, and tumblelogs. It appears to consist in a set of inherently flexible and connective technologies, and to consist of disjointed but interrelated pieces of information.

I think I see the problems now, in large part due to the thorough treatment of the issue in chapter 1 of Chuen-Ferng Koh’s 1997 thesis, found at http://wwwmcc.murdoch.edu.au/ReadingRoom/VID/jfk/thesis/ch1.htm. The internet may be rhizomorphic, but to call it rhizomatic ignores such attributes as its trace-based mode of presentation and the difficulty of manipulating content free from structure. If the internet were truly rhizomatic, such notions as “a map of the internet” and “formless content” would be products of an inherent organization rather than things which could be embedded back into a flawed original. What I would like to consider now is if a truly rhizomatic information meta-network technology might be realized.

There are really two problems with the internet that keep it from being a rhizome. The first is a software problem, and the second is a hardware problem. The software issue is that the information on the internet is only navigable; that it must be accessed page by page (even by a search engine or other machine) along the edges of a graph; that two pieces of information do not relate in a bidirectional, mutually-inductive way but rather a unidirectional, trace-based way. This problem does affect the internet deeply, but it is a relatively tractable problem: one can picture an internet where no distinction between servers and clients existed. All parties would simply make available to their peers the information they wanted to display. To connect would be to have access to a set of data or locally-run programs. All nodes would be on equal footing. To access a service would be identical from a technological perspective to hosting a service. When two peers initiated contact information would flow between them; their informational spaces would merge. To host a website where users could store content, for example, you would make available to all connected parties both the database of their content and some helpful interfaces to it, rather than just an interface as sites like Flickr or Gmail do now. And in return, ‘clients’ would make available their verification data and perhaps some storage or processing on behalf of the connected party. A peering of computers would be a merging of computers; a human transaction mediated by machines would remain a human transaction.

But, because of the current hardware structure of the network, the notion of a “peer” or “shared presence” is meaningless. Everything is connected to everything through trees of ISPs and telecom firms… but at the same time, nothing is connected to anything other than them. ISPs impose rules on traffic, are necessarily subject to legal and financial oversight, and are really a third party interjected into all interpeer transactions.

For me, this is the single most problematic aspect of the internet. When viewed as a marketplace this doesn’t really make too much of a difference… services are services, whether provided by chains of companies or just one. But in terms of the interpersonal connectivity and self-publishing aspects of the internet, I think this structure is undesirable. At any rate, it is certainly not rhizomatic.

I think that mesh-networking technologies could bring about this shift. The internet is clearly very useful, but I think that a more rhizomatic technology could provide things the internet can’t, or do them better. When people communicate via the internet they are doing exactly what the always do, amplified. To be in the same space as someone else is to share communicational and informational resources. To be connected via the internet entails similar transmission of information without any sense of intimacy. A rhizomatic network technology would require the effortless direct and complete connection of any node, i.e. person, to any other, mediated by protocols which allow free bidirectional flow of everything being exchanged.

Connections through the internet, whether to a personal site, a company site, a repository like wikipedia, a service like gmail, are essentially remote linkages of functionality and existence. To read anything is to receive patterns which originated somewhere and were packaged as language; to read a webpage is to receive patterns generated by human-designed machines, based on human-designed rules, carrying human-designed (if not written) content. The connectedness of peered computers actually reaches above that of of peered people (people in the same physical space) in that the peered machines can work together.

When two people share the same space, the same ecosystem, they form a rhizome, just like the wasp and the orchid. Even if they are just sitting in the same cafe or walking past each other along the street, simply by virtue of being where they are they share a plane of consistency, they affect each other, and while there may be no explicit relationship or connection the link, nevertheless, exists. In contrast, when people walk into a storefront, the relationship they form with the company in question is not rhizomatic: information flows in only one direction and there is a definite hierarchy.

The internet, on the other hand, provides only relationships with a hierarchy. When you connect to a website the website has and you do not. The information only flows toward you. The two places where this seems to break down, hyperlinks and Web 2.0 style interactions, are illusions. That you can pick your superior is not really a claim at equality, and though you upload you photographs to Flickr you are unable to host your own photographs. You must depend of Flickr to provide the service. When you post to a forum you are subservient to the forum’s host, when you talk on AIM your messages are relayed through AOL’s servers.

But the difficulty here is that these services are useful. Though the internet stretches human interaction wildly out of its natural rhizome and enacts hierarchy where none is necessary, hosted mail is convenient and a central repository of photographs provides a sense of community as well as an easy way to share with people you know in real life. But the large-scale mode is is only one mode of interaction. As forums and IM exemplify, small-scale interactions are often desirable as well. The internet as it stands can only provide an illusion of this.

To tie this back together, the internet is a mediator of human-human interactions and human-corporate interactions (which generally include human-computer interactions). Human-human interactions outside of the internet and in which hierarchy is not involved constitute the formation of rhizomes. Human-human interaction via the internet can not. It would be good for this to change. On the other hand, human-corporate interactions, both inside and outside the internet, are never rhizomatic. There is always a limited flow between parties. This seems to work pretty well for many things. So, the existing technology should be augmented and not replaced.