the architecture of the internet

& the world wide web

duluth quantum computing project

  1. reading list
  2. discussion
  3. dqcp topic list

reading list


deeper read (swimming ::: tl;dr)


The internet is often referred to as the cloud. It is a beautiful metaphor ::: we are skywalkers, astronauts, birds, trapeze artists. We are spiders weaving a silk web ::: listening to data stream past our taut silk chords.

In fact, the internet is a very physical thing ::: cables, switching stations, power supplies, antennas, satellites. No one person or organization or country owns the internet. It is a dispersed infrastructure : a network of arteries. Packets of data flow through this vast labyrinth swimming towards their far-away destinations at lightning speeds. Fiber optics ::: glass ::: electrical pulse. Satellites beam down information in every language ::: symbols and functions ::: characters and computations. Journalists submit news, citizen observers upload live video, activists transmit electronic calls to organize. Revolutions are ignited, surveillance tightens its iron grip, evidence is gathered. And markets thrive.

The internet is an infrastructure : a vast collection of interconnected networks. The world wide web (WWW) is the information space that rests on top of this infrastructure. The WWW is the blood ::: the pulse ::: the flow.

Hypertext existed long before the WWW. It was used in early single-machine, digital settings to create links between chunk of text within a document. It makes "clickable" what previously existed in print form as footnotes, citations, sidebars. Hypertext implements navigation within a document ::: within a single electronic reading machine.

The birth of the world wide web happened in large part through a simple abstraction. What if these hypertext links could communicate across machines? What if a link could access information from another document residing on another machine that itself resided in a remote location?

In 1989 Tim Berners-Lee was an independent contractor working for CERN (a particle physics laboratory in Geneva). He proposed creating a system to share documents between scientists at the lab. He envisioned using an early Internet communication protocol already in existence to support this new interlinked, inter-computer information layer. This led to the development of the Hypertext Transfer Protocol (HTTP).

HTTP is a simple exchange format for the transfer of hypertext documents between machines (and their human listeners). It is the backbone for the WWW. It is a language of call (request) & response. It has a simple vocabulary ::: a lean grammar. HTTP uses the verbs ::: GET, POST, PUT, PATCH and DELETE. These verbs allows an HTTP broker to request (read), create, replace or update a resource. For example, a GET request issued by a browser might fetch a web page, a POST submiited by a browser might send data from a user-filled form to a remote database.

Rapidly the small, internal network from CERN became a network of networks ::: an interconnected world. The web's rapid global adoption was in large part due to the creation of a hypertext reader ::: an HTTP portal ::: the browser. From Wikipedia ::: "A web browser is a software application for retrieving, presenting and traversing information resources on the World Wide Web." We suddenly had cheap portals for virtual travel. We could navigate the rising waters of an information sea. We surfed the tidal wave.

We navigate this space by means of a special type of address called a uniform resource locator (url) of the form ::: scheme:[//[user:password@]host[:port]][/]path[?query][#fragment]

For example, we can go to ::: to learn more about the cryptic url description in the last paragraph.

Today resources can be much more than plain text documents. Resources can be images, video, sound, code. This enhanced, extended hypertext form is often referred to as hypermedia.

We have an ever-expanding linked graph of resources. A future cosmologist would see something resembling the big bang ::: one particle ::: one machine ::: a resource ::: a single lonely island of information. And then a first link appears across what seemed like an infinite distance between two, separate machines ::: a link to another island ::: a co-authorship ::: a relationship ignited.

And then we ourselves became authors. Creating content became increasingly accessible to more people. It became easier to distributing content ::: to place our work & reflections into this space. Artists impacted the web and, like any new technology, the web impacted art.

So here we are in the midst of a sea change.

This nonlinear, expansive landscape gives rise to new visions of space & time. We are called to imagine new means of traversal. Reading becomes a flying trapeze performance. We become aerial artists leaping from point to point ::: utterance to utterance ::: url to url ::: resource to resource.

Creating new work requires explicit attention to defining tools for navigation. We must decide on ways that we direct the reader ::: the pathways ::: the flows. We create a landscape ::: draw the map ::: add roads & bridges, rivers, topography, flows, junctures, obstructions, ladders & rabbit holes.

There is an art to constructing the index for a book. It is difficult to decide which elements / topics to list. Similarly in a hypermedia world, it is a challenge to define the core chunks ::: the most elemental utterances ::: the stand-alone nuggets of meaning. For a poet, is it the poem itself? a stanza? a line? We define what we would want to link to ::: what we would choose to reference. These become core pathpoints that are then laced into one or many paths. These core nodes become the juncture points ::: landing places ::: reflection pools ::: cartoon panels ::: blog entries ::: stanzas ::: acts ::: movements ::: soliloquies. These are our beads. We can make a single, linear chain with them (like a printed book) or we may place them in a nonlinear context ::: a lattice ::: a graph.

Pathpoints can be nested. A stanza can be linked to other stanzas in a poem. The poem can be linked to other poems in a book. The book can be linked to other books in a library. They can be ordered according to a library of congress classification scheme or sorted and linked through keyword searches ::: a more ephemeral way of defining relationships.

In a hypermedia world an author's pathpoints can be referenced by others ::: the constellation becomes fluid, infinite, communal. The relationships between nodes might shift, the pathways might be temporary, an author might transport the reader to Brigadoon. The unreliable narrator becomes the unreliable landscape. Back, next, home buttons can work as expected or they can shift ::: become random ::: become unstable. The reader can fall through the looking glass.

Hypermedia has been used effectively to mark up content with background material. The form is often used to further educational goals ::: to enhance content. In a linear presentation links can provide the viewer with access to referenced materials ::: to side journeys ::: to drill-down, in-depth exploration.

In creative work hypermedia forms can be used quite freely. We don't need to mimic didactic forms. We don't need to embrace the marketer's intent. We aren't constrained by what has become codified ::: the website. This cultural artifact shapes readers' expectations but just as nonlinear, experimental work sought to break the rigid world of linear print book fiction, we can loosen the grip of expectation. The form can be used in poetic, associational, unexpected ways to express the strange ecstasy of traversal ::: flight.

In the early days of film ::: moving pictures ::: it wasn't at all clear how creative work would map to this new technology. Vertov's 1929 silent film, "Man with a Movie Camera" was a hallucinatory collage of image and effects. It reveled in this new time-based, visual riverway. Vertov was unbound by a traditional way of using the form. There was no tradition. Eventually though, film evolved into patterns ::: the medium became codified. Filmmakers found it natural to map the linear narrative structures of printed stories or plays to this linear, time-based form. It then took a conscious act to break linear expectations ::: to create "experimental" work that deviated from this norm.

Humans didn't always adhere to a linear, storytelling practice. Linearity perhaps mirrored the physical form of the bound book. However even the bound book has always held nonlinear deviations (sidebars, footnotes, indices). Authors often employ narrative strategies such as flashbacks, non-sequential action, associative breaks and suggestions for the reader to deviate from a linear reading ::: alternate paths, optional endings, substitute chapters, sudden shifts in perspectives. Storytellers break out of the confines of time-based, linear plotlines. Oral traditions were more fluid ::: didn't adhere as closely to fixed story arcs. Often a chunk / episode would be paired / juxtaposed with another ::: a subset of a larger storyworld would be selected to stress a particular meaning or to honor a particular place or season.

In hypermedia forms nonlinearity can be physically manifest in the structure itself. The explicit nature of stand-alone, url-referenced units (resources), relationships (links) and the timing of link activations require explicit decisions. The author isn't creating content alone. There is a complex orchestration at play ::: a choreography to define ::: a set of theatrical parts and interactions to create.

Let's look closer at the concept of the link in the context of this web-based world.

Most web pages articulate a hyperlink with the following markup ::: <a href="">the datapoets website</a>. Note that the href attribute on the anchor ("a") tag specifies the url for the resource.

There is a more complex specification for hyperlinks called XLink that was agreed upon by the standards body (the world wide consortium or W3C). This specification, however, is not implemented by most browsers. Despite its lack of use in the real world, it is interesting to take a look at this deeper form of a hyperlink to gain insight into the act of traversal. XLink articulates the reference url and traversal behavior for a link with the following attributes :::

(the same meaning as the traditional anchor tag)
Defines when the linked resource is read and shown. Its value can be "onRequest", "onLoad", "other" or "none"
Specifies where to open the link. Its value can be "new", "replace", "embed", other" or "none". The default is "replace"

There are other possible attributes but I wanted to talk specifically about these.

. . .

topic list