Mary Hodder

June, 2004

Linking Practices and Expressions, A Socio-technical analysis.

 

Linking is the act of connecting two sites or locations together, with a hyperlink between them, constructed with hypertext markup language.  Links are implemented by the act of linking, usually a task performed by a person who creates a webpage or weblog.  Links link to something, they point to a website or online location, with a uniform resource locator (URL).  The purpose can vary greatly, in intention, endorsement or recommendation or referral, or as an expression of some sort that may be more literary or artistic.  Linking is by definition a one-way act.  Links can occur reciprocally, but this is not required for linking to work, and it is this design element, intentionally created by Tim Berners-Lee[1] when he made the first browser and set about the terms of the web, which is a key element in many ways to having an internet that functions well.  That is not to say that everyone is happy with one way links, and Ted Nelson has made that clear[2]. He believes that two-way links are imperative.  However, this paper is not concerned with what might have been if that were the case, but rather what happens with links as they occur now on the internet, and the information and actions they convey in the act of linking.

 

It is this quality of one-way linking which allows for the directing of attention or referring to occur, without agreement to the linking process from the receiver of the link, that causes the web to function well, unencumbered by any agreement between linkers, even if links break or were never properly set up.  While the unstated nature of the reason for the link, or the endorsement of what is on the other end of the link is interesting, as is the linking expression in the construction of the link itself, this one-way nature adds a great deal to what makes linking interesting as a social phenomena, because of the lack of agreement required.  Those interested in linking, or expressing some attention, to something can simply do it, without requiring notice or consent, in either social or legal or technical terms.  Therefore, a link can act as a pointer, without the burden of agreement, or the need to reciprocate the link. It is this freedom that allows links to embody intentions and actions that do not rely on anything but the linker’s own set of thoughts, without regard for others.  In this environment, one-way linking represents a purer and more natural set of intentions that can stand on their own, without being tainted by other’s expectations or intentions or agreement.  Often links are made without the knowledge of those being linked to, depending on the situation.  However, the ability to know easily and quickly that one has been linked to by others is changing as more tools and better search are developed, and this is changing the relationships between those linking and those linked to on the web.

 

Actor network theory, particularly as understood by Bruno Latour in his research about the process of documenting science investigation in Pandora’s Hope[3] sheds some light on understanding linking relationships.  This comparison of linking to traditional scientific investigation can be seen either in the linking process and interactions that occur, or in seeing how linking can be different that the citations and evidence processes utilized in the traditional science process is a useful comparison.  In addition, his “Visualization and Cognition”[4] essay is useful partly for seeing relationships between those who argue an idea, with support, and those who link to documents to support their arguments and assertions online.  Both these works will be reviewed against the behaviors of linking to find those similarities and differences to online linking behavior, as well as how these properties apply to the construction of links, what the person linking offloads within and across the link, conveying the person's intentions, as well as the mobility of the linking expression, the creation and destruction of links, how links bridge a gap between the information explicitly conveyed at the top level of a website or blog, and the linked information at some other location, collapsing the distance between the two, creating a relationship where none may have previously existed.   There is a transformation of the linked item, by the linker, and there is a realization across the two pieces of information in their home locations, as one pulls the other in, and creates a changed understanding of the previously understood state of the item that's been linked.

 

Terminology

 

Traversal, as a concept, in layman’s terms, is about crossing over, or making a path from one area to another, or going through something to another space, and this concept is helpful in understanding in a basic way that linking as a path, a direction, from one node to another.  Links do express a path to another site, and clicking a link can be an act of traversal.  But there is a more technical description that explains the computer science definition of traversal:

 

Traversal of a tree is recursively defined to mean visiting the root node and traversing its children. Visiting a node usually involves transforming it in
some way or collecting data from it.

In "pre-order traversal", a node is visited _before_ its children. In "post-order" traversal, a node is visited _after_ its children. The more rarely used "in-order"
traversal is generally applicable only to binary trees, and is where you visit first a node's left child, then the node itself, and then its right child.[5]

 

Understanding traversal as a functional, flexible act is helpful because of the interactive nature of linking, and that due to the functionality of the tools that support links, which include browsers, which have back buttons that allow the reverse of a traversal, means that the browser can be made to go back and forth between nodes. 

 

Blogs, or weblogs, are webpage logs used as personal publishing tools.  They have several kinds of linking, that express different sorts of intentions, partly because of where they are used in the blog, and partly because of nature of the linking expression.

 

Permalinks, a term used mostly in blogs, are links that generally sit at the bottom of a post on a blog.  A permalink is meant to express the exact location of a particular item or blogpost on a webpage or in a database, and is considered to be a deeplink.  This contrasts to a more general link or base URL.  For example, a base link might be made to a site like http://www.google.com, as opposed to a deeplink to a specific page on their site, which might be http://www.google.com/help/index.html which has permanent information about their help section.  Using a link such as the first one is a more general act, like referring to Google as a company in a verbal conversation, verses to a specific thing they do or say on a particular page, where because the link is made to the deeper material, the referral is specific to what is on that page.  Links that are general are used the refer often to entities or people, where as deeplinks are often used to refer to specific information or events.

 

Trackbacks are a system that allows a ping, from page posting link, to the linked page.  Sometime, trackbacks are automatically displayed on the pinged page, explicitly showing the link to viewers of the linked-to page.  This is somewhat unusual, but becoming increasingly prevalent.  There are also trackback displays on websites and blogs, that viewers must click on to see the trackbacks via a popup. However, this sort of trackback is not visible on the front page, except as a link to the popup, and therefore less explicit.

 

Actor Network Theory

 

There are several ideas in ANT that are directly applicable to linking, including enrollment which is central to ANT and linking.  ANT understands that enrollment must be voluntary, that there must be freedom to enroll in whatever is being offered, or there will be resistance.  Latour’s use of ANT in seeing the construction of science is helpful for seeing how hypertext linking is used and understood.  In particular, in his Visualization and Cognition essay showed how enrollment was used to get participants in the scientific process, funders and other supporters to buy into their plans.  There is a direct similarity to linking, though webpage writers and bloggers and not necessarily undertaking so ambitious or needy a project.

 

Bruno Latour asserts in his essay on Visualization and Cognition that holding a focus relates the visual and cognitive together for conveying the information and arguments.  Mobilization involves going and returning with the "things," which require adequate presentation so that others will understand and be convinced, and this involves the mobile, immutable, recombinable results of the research.  Mobile is what it says, movable, and immutable is to know the source, that it is incorruptible, and recombineableness for reshuffling, remixing and reducing of the information in order to discover new information about the subject.  These qualities make it easier to transfer knowledge.  And it is these qualities I am interested in regarding hypertext linking, to see where the are related and where they are different, to understand more about linking and what the human interaction is with it.

 

The question writers often face is how to convince someone to take up the writer’s position, to pass it along, to send out an idea or meme for several reasons:  the writer’s own prestige, the benefit of the commons of ideas, because the writer is emotionally attached to an outcome.  There are many more reasons, but what is important is that what webpage writers and bloggers are doing is attempting to convince their audience of their assertions, opinions, and authority as a source for filtering of information and of more general interest.  Linking plays an important part in enrolling the audience to see that the writer is competent, has based opinions on something real, and often in the case of bloggers, a news story or other article that acts as the base material. 

 

By linking, a writer can underpin his credibility with his audience.  The audience can see the basis for the writers assertions, and the links become the enrollment tool.  And by linking, a writer can delegate the responsibility of bearing the entire set of assertions they are making onto those links and what is contained at the other end of the links.  By doing so, the writer in essence is not entirely responsible for what is asserted, because they can point to the many other sources, even if they are arguing against some of them, to show that others have come here before, that the factual information has arrived and been documented by others, much like the scientific citation system described by Latour, and the evidentiary system he describes in documenting the expedition in Pandora’s Hope.  Writers delegating in this way, can spread responsibility, and give the link some of the power to perform these distributions.

 

Linking over time is also a way for writers to have their audiences subscribe to linking and tools that understand linking.  By linking often, to documents, other writer’s work, webpages as representations of people and institutions, media and news articles, linking becomes an act that audiences look to, or subscribe to, as a form of directing attention, literary subtext, supporting evidence, description of others, associating, making fun of or any of the other many possibilities for intent that a link might embody. 

 

All of these forms of linking cause audiences to see there is something there, to visualize the linking expression, even if they don’t go to the link, by just mousing over it to read the linking expression.  But it is just as important that there is something on the other end of the link, for those who follow the link, to see both the linking expression, the base URL where the document or website or information resides, as much as to see the information itself. Further along in this paper, I discuss the case of Chilling Effects hosting documents that have received cease and desist orders.  While the original location of a document in say, a Google cache, with a link to the originator of the document means one thing, a URL pointing to a document that has been removed from Google’s cache, and placed at Chilling Effects with a cease and desist order means something else to readers.  The URL is enough to queue readers to see this difference, even if they do not necessarily follow the link, though the fact that a link has been redirected to Chilling Effects by Google might make readers more curious, just to see what happened to cause the switch. 

 

It is both upon seeing the location of the document, and seeing the document itself, that the “visualization and cognition” Latour discusses, and the holding of the focus, is comparable and useful to understanding linking, and people’s understanding of it as they use it, both as writers and readers.  Links become queues expressing that there is more information available, that something else is happening, under the text, even if there is little understanding by a reader of what this means.  And in this way, the queue to the link becomes a form of subscription to the linking process. 

 

Subscription also happens when writers, by performing lots of linking, work with systems like Google and Technorati.  By constructing links in particular ways (using href and rel html tags wrapped around the link, as opposed to say, putting the link into the top level text, with no hyperlink beneath) writers are allowing these links to be cataloged and understood in deeper ways that the pure, flat text would otherwise be by the Technorati, Blogpulse or Google spiders (other services also spider, but do not understand the multidimensional layers of linking below the text, and so flatten everything into one layer, without the network of outbound links).  However, there is a form of resistance that has developed around this kinds of linking, with bloggers, because they understand how links are used by services and tool makers to create authority for those being linked to and that is that writers will describe but specifically not link to site they find abhorrent.  There have been proposals put forth to understand links (using rel=+, rel=-, rel=0 kinds of positive, negative or neutral characterizations) that would allow writers to explicitly state their recommendation of a link, their neutral stance, or their opposition to a link.  However, this raises many questions, as one writers recommendation or positive support may be taken as something else than intended by the readership, or the systems than collect these tags and redisplay them, and because like social networks, explicit links lead to the linkers feeling compelled to characterize their links differently than they might actually understand the information due to other reasons or pressures. 

 

Part of being explicit about the tags means that everyone who wants to look can see how a writer characterized a link, and if they link is back to another writer, in essence, the characterization becomes one person to another.  People are often reluctant to pass along these sorts of characterizations so easily and publicly.  Therefore, a writer might either hold back the characterization of a negative review, either as neutral, or simply revert to the act of describing but not linking, because linking will lead to a trackback or some other way of finding the reference by those being linked to, instead of putting up a negative tag.

 

Another issue with explicit tags has to do with the conveyance of what the writer meant when assigning a positive, negative or neutral tag.  The writer may assume that by putting a positive link up, that readers will believe it means the writer thinks readers should look at the linked item, when in fact readers think it is an endorsement.  If this system were carried out, either through publishing tools, or mass voluntary adoption of manually input tags, it is likely that the system would be corrupted as every other rating system like this has been corrupted before it, where people rated or endorsed for reasons other than those they intended as they were simply, in this case, linking to information.  By making the system explicit, many problems are introduced, not the least of which might be the gaming of the system by a few, who might want to manipulate the ratings of others they either did not like, or did like or otherwise wanted to change.  If tool makers then relied upon those ratings, they might report search results based on flawed or gamed rankings, thereby degrading the usefulness of their tools and users impressions of the accuracy of the tools.

 

Knowledge Offloaded

 

The person linking offloads within and across the link, conveying the person's intentions, as well as the mobility of the linking expression.  The distribution of understanding occurs, to use an example of a blog post, by a blogger who links to direct sources, rather than having to restate or quote information.  The link stands as a reduction of the information behind the link, as an expression of what the link may contain if that link is expressive in communicating what is there at the source, and as the agent for traversal if the reader wants the underlying source data.  The link serves as a placeholder where the blogger can offload their own thinking, no matter what that is, and it can be extremely varied in intention and meaning.  And instead of the sort of referencing that occurs with paper-based articles and endnotes referencing other articles and books located often far away, across time and space constraints for accessing it, a hyperlink can collapse this, making the offloading of cognitive understanding by the hyperlinker to a far more efficient sort of offloading. 

 

In addition, the sourced material being linked to, might have some mechanism, either automated, or manual, for returning the link.  An automated system might display a trackback on the front page of a blog, creating a link back to the original post and link.  In this way, a sort of circulating reference is explicitly generated.  Those who are linked to might also find out about who is linking, by using services like the Google link lookup, though there is often a lengthy delay, and the links are arranged by their pagerank algorithm.  For those wishing to find others linking more immediately, services such as Technorati and Feedster find those inbound links and display them by the most recent links.  In this way, readers of materials with a shorter term of relevance, or in the midst of a conversation occurring over a short period, can complete a link circle, which in effect manually completes what would otherwise occur with trackback, if it worked for users all the time.

 

Circular reference as Bruno Latour discusses it is a kind of reference that travels from discovery to recording to study to catalog.  This traversal of the reference is different than in the linking explained above, because it involves in the case of Latour’s study, physical artifacts that travel from person to person as the work in the study is completed.  But there is a way that the circular reference is similar, in that as the artifact is passed, it is also described, though an academic or scientific process for cataloguing and citation.  This citation process is one of reduction, so that information about the artifact can be culled, and catalogued together with other artifacts, put together, to discover new sorts of information across many of the objects being studied.  In this way, the study references can resemble hyperlinks that are expressive about what is at the other end of the link, and in the case of a blog post or website that catalogs several links in one area or discussion, there is the ability to see the links put together, for the purposes of understanding and discovering more than what is contained in any one of the references or links.

 

Hyperlinks often bridge gaps between the information explicitly conveyed at the top level of a website or blog, and the linked information at some other location, by collapsing the distance between the two, creating a relationship where none may have previously existed.  This ability to bring what was very far away, to a state of nearness, through a linking relationship is different than offline physical communications because the physical distance can be removed by the act of linking.  The relationship making can be the same with online linking and offline citations, creating another kind nearness, the physical barrier to a reader in looking up the referenced material immediately does change the state of the thing being linked to for readers.  Suddenly, this ability to see the reference, alters the situation.  A reader might trust a hyperlinked reference to the writer, where simply knowing the resource is accessible, and can be moused over to check it’s expression for some kind of information, may be enough in most cases.  With a paper based article, the ability to check the resource might require travel or time waiting for the resource.  The quickness and lack of interference, barring a broken link, of a hyperlinked online resource allows users to, at one time, believe more the implication that the writer doing the hyperlinking is telling the truth because the reference is so accessible, as well as investigate more of the source for himself, if interested.  This linked item, in a way also becomes the “silent witness” that artifacts described by Latour in the scientific study rely on, though the interaction between the reader of the study, and the read of a hyperlinked text, may result in different actions due to the difference in accessibility of the referenced item.

 

“Capitalizing Linking To Mobilize Allies”

 

Latour talks about capitalizing inscriptions to mobilize allies.  What he means by this is that in science, a reduction of the items studied is made, in order to then combine inscriptions, to then mobilize allies or convince people of their findings. There are several ways that linking is being used in this same way, because of tools developed to understand the frequency, aggregate quantity, and kinds of linking being performed, as well as a way to understand authority and relevance.  But linking is, in the act of making a link, not a reduction by the linker as it might be in the act of a science researcher who takes a specimen, tags and inscribes the item, and then combines those inscriptions for analysis and conclusions, in order to then have their findings peer reviewed and accepted.  People, and for this discussion we should focus on bloggers as probably the most active linkers currently on the internet, do not usually create the linking expression in the same way that an inscription might be created by the scientist.  However, a blogger does choose documents, blogs, blog posts, news articles, etc. in order to support or in some way associate, even if negatively, with the writing in the blog or blogpost.  And it is this choice of the expression on the other end of the link, that leads to using the link, which in the aggregate, becomes interesting for many services such as Google, Technorati, Blogpulse and Feedster.  These services are able to show in distinctly different ways what the aggregations are toward the linked items. 

 

For example, currently, the New York Times has 32,000 plus inbound links, counted at the root URL level (nytimes.com, though all links are included because the search is for any link with this base), to any article, or to their site generally, compared to the Wall Street Journal with just over 400 (and which is behind a firewall and therefore, not terrible linkable in the minds of bloggers who would like to send their readers to supporting or associated artifacts that will be viewable by their readers).  The ability to combine all the links to the NY Times together is one way these link counting tools act as recombiners of what is essentially recombinable.  One reason the NY Times has so many links is that it has made itself available to be discussed by bloggers, and so in a way it has allowed itself to be mobilized by bloggers, who bring more allies to both themselves as they assert things about the Times, as well as simply attention, authority and relevance, to the Times itself. 

 

Attention occurs when users find top linked-to sites using services like Technorati and Blogpulse and Feedster, which point out the distinction between those that have heavy inbound link counts and those that do not.  In this way, allies to this system of understanding both link counts and linking more generally, further adopt the understanding that being linked to and linkable is a key to authority and relevance.  On systems such as Google, which uses page rank algorithms to rank the results, so that what is returned is sorted by the top ranked items, those with the most links are displayed at the top.  Google affirms this structure of understanding links both by using links to understand rank, but also in displaying them.  Technorati, in contrast, understands this as well, but then allows the results to be sorted by date, which means that people can see both the authority through ranking, as well as those results which are most recent.  By doing this, Technorati resorts the authority, making time a kind of relevance as well as inbound links.  In otherwords, they are able to associate an additional property with the link that others are not, and by doing so, have mobilized many bloggers to use and propagate their position as an aggregation authority over other tools that understand and redisplay links.

 

Other aspects of a comparison to Latour’s understanding of the generation of science and citation are interesting in this context of understanding linking by bloggers.  Use of the term “immutable” in science, is about a specimen kept in stasis so that it can be brought back to the lab from the natural world, to be studied further.  Bloggers, in their use of links, and the expressions they link to across the web, are in some ways relying on this quality, though they are not doing the preserving themselves most often.  However, sometimes, bloggers will cut and paste some or all, or simply save and redisplay, something on the web for some reason.  But most often, bloggers rely on the assumption that when they link to something, that item will be there for people to see for a while, even if it goes away at some later date.  Also, the link itself, even if it becomes broken, in some ways also is immutable too.  In the expression of many links, it is possible to gain some idea of what will be there on the other end, and if a link is broken, the change or break is a kind of information in itself because that item has been moved, removed, put behind a firewall or a website is down. 

 

Websites and blogs are not completely immutable, however.  They can be “corrupted” in one form or another as described above (as missing in some way), or if changed or replaced as well by their owners.  Those who link to sites often have no control over the sites to keep them immutable.  However, those linking do have the recourse of copying either part of all of the site, and redisplaying it.  There are various copyright and other restrictions against this practice, however, it is done somewhat anyway, in an attempt to maintain information and create a new set of links to the copied information.   This new set of links is determined by the linker/copiers, because they determine the copy location and naming.  In this instance, they become keepers of the immutable documents, in the fullest sense that Latour defines this.

 

There is also a sort of transformation of the linked item, by the linker as linking is performed.  Partly, this is due to the tools mentioned earlier that count links, understand them in some way, and report this back in search results along with statistics about those sites or documents being linked to and discussed.  But there can also be a realization across the two pieces of information in their home locations, as one pulls the other in, and creates a changed understanding of the previously understood state of the item that's been linked. In this way, trackback, if it is implemented in such a way as to quote from a part of the website that is doing the linking at the point where the link is, and then that quote is redisplayed on the end being linked to, can actually change the document on the receiving end of the link.  In this way, the link shows up as a quote (usually a few lines) with the link, and readers and the blog owner on the receiving end of the link, can see how they have been pointed to, and find the link back to the quoter as well.  This loop is transformative, and very new, in that in the past, citations were not so easy to find, and even with services such as Technorati or Feedster, would not necessarily cause all readers of a linked-to blog post to see those references, because it required knowing about those services and using them.  But trackback implementations such as that described above, make and explicitly transform the original posting by showing the reference.

 

However, this kind of mobility, where the trackback information is redisplayed, acts as a kind of mobility, where, if set up manually, the trackback process of redisplaying information can occur automatically.  Linking can function as a tool that goes sends a ping to another site, alerting it to go out, get information and bring it back.  But the act of the original link is also about getting information and bringing it back.  This can be as simple as brining back the linking expression to implant that into a blog post or website.  But it can also involve the copying of some or all of what is on the other end of the link, and in this way, the relevant information, when selectively quoted, can act as a permanent record, as well as a reduction of what the linker finds important.  These reductions can be recombined in posts, along with other quotes, and explanations about why these various quotes are important, why the author is pointing them out, and how the author would like the readers to see those quotes and documents.  In causing multiple links from a single post, a blogger can construct a sort of outbound link understanding or reduction of the overall post.  This too can be seen with tools such as Technorati (which at onetime, displayed outbound links and may again) and Google, in their advanced search tools.  However, a reader can simply mouse over several outbound links in a post, to determine quickly what the author is pointing to, to gain a quick understanding of this information.  And of course what is outbound on one end, is inbound to the other end, and those inbound links do become the readable, combinable basis for Technorati and Google.

 

 

 

 

 

Resistance Verses Transformation

 

There is an interesting question in light of the one way linking that occurs online.  Do people who resist shift things some without totally removing what is being resisted?  Certainly in the case of links that are made to from one site to another, or from Google to a site that someone doesn’t like or protests for some reason, there is resistance.  In light of the theory of enrollment, there is the question of who is enrolled. In the case of the entity linking, and Google as the indexer of the links, there is enrollment.  But for those being linked to, there is no need for enrollment, in a way, because of the lack of need for agreement in the linking process.  But because Google, and other search services for particular corpuses of data such as Technorati (technorati.com) allows for a search lookups of all those linking to a particular URL, and because some blogs effectively make trackbacks (links from one blog to another) apparent to the blog being linked to, it is possible for those being linked to or those who protest some particular material posted on a URL to understand who is linking to that URL.  The information is more noticeable, particularly on blogs where it is often posted in some top level way to the blog, either on the side bar under comments (see Copyfight’s left side bar, under comments for another  example of this:  http://www.corante.com/copyfight/), or under a particular post either explicitly as text or linked under a “trackback” hyperlink.  Because these tools now make these links explicit and reflected back to the linked-to post and the audience, there is more of an opportunity for the entity being linked to, to protest or respond, and because it is more able to be responded to in some way.  So there is an implicit enrollment of the object linked, because resistance is easier to make, simply because a kind of notice is easier to have.

 

In the case of Digital Millennium Copyright Act protests, where a link or site is deemed to violate copyright or some other provision of the act, those whose work is being violated can resist the linking and request a “take-down” by giving “notice” to the linker.  Under the DMCA, the links if removed within 14 days, are protected with “safe-harbor.”  However, due to the number of requests, Google then passes the “cease and desist” notices to Chilling Effects (chillingeffects.org) who processes the notices, returns them to Google, but also posts the notice including the offending material on the Chilling Effects site.  Google then links to the Chilling Effects sites links to the offending material, notifying searchers that the offending material was removed from Google due to a DMCA protest.  It is Google’s act of removing the direct link, yet replacing the link with another link that effectively maintains people’s ability to find the material with an additional click or two, that creates a transformation in the relationship of the links, but not the complete removal.  The information is still there, still findable, and yet the link construction and expression are now different, pointing to Chillingeffects.org’s copy of the material. 

 

Because of the nature of this process and Chillingeffects.org’s redisplay of the offending materials, the rules are followed, and there is a kind of partial acquiescence to the power of the DMCA and the Internet Service Providers who enforce the rules.  However, it is only partial, because it results in Chillingeffects.org’s redisplay.  The information never really goes away, and so it is a way of not giving over power completely, though the agreement to transform the information and link does mean some power is held by those who use the DMCA to alter what they object to on the internet, refusing to enroll in the idea that either linking or presenting information is an undisputed right. 

 

The original information on the site is also transformed by inbound linking.  This is because the documents linked to and located at Chilling Effects include not just the original, but also because of this resistance, the C&D notice, and information on the parts of the law that apply to the C&D, the DMCA and the offending materials.  In this case, the answer to the question above about resistance points to a shift and transformation, but not the complete removal as the resister would like.  Instead, there is a compromise, between those linking and those linked to, and between those linking and those outside the linking relationship but with an interest in the linking, accommodating some but not all of the interests of each of those in the linking relationship.  The actors, tools for linking, the objects being linked to and the network that moves and transforms due to linking, does shift, though they do not disappear.  They do not remain as they were before there was any linking, nor are they in the state that occurred after a simple link was made, but rather they rest at the point where the resistance has noticeably transformed them into different objects, related with different links, and with vastly different implications and meanings, while still maintaining the original information.

 

Other Aspects of ANT and Linking, For Future Work

 

I am less sure about parts of the ANT theory that have to do with human and machine made technologies or artifacts, as useful comparable relationships.  Links can be machine generated, though they are most often man made on blogs, though bloggers use links to other pages that have been machine generated, such as those from large traditional media publishing sites.  However, I am still investigating what those links mean, discovering what the machine processes are and how to treat them, and what the interaction is, because I do not yet understand how the links are generated.

 

I plan next to address these issues to try to discover whether there is some way to see these differently generated links, and the consequences, both intended and unintended, to understand how they act, and change the networks the linking creates which feeds back through tools to users and affects their perceptions of what websites, blogs and information is important and authoritative.

 

 

Bibliography:

 

Latour, Bruno.  Pandora's Hope: Essays on the Reality of Science Studies  (Cambridge: Harvard University Press. Chapter 2: Circulation Reference: Sampling the Soil in the Amazon Forest.)

 

Latour, Bruno.  ‘Visualization and Cognition: Thinking with Eyes and Hands’ (from Knowledge and Society: Studies in the Sociology of Culture Past And Present, 1986, JAI Press Inc., 6: 1-40.)

 

Ted Nelson on Two Way Linking:  http://ted.hyperland.com/buyin.txt. 

 

The Free On-line Dictionary of Computing: http://dictionary.reference.com/search?q=00-database-info&db=foldoc.  © 1993-2003 Denis Howe

 

Tim Berners-Lee Biography:  http://www.w3.org/People/all#timbl

 

XML Linking Expression Schema:  http://www.w3.org/TR/xlink/

 

Vote Links Discussion:

http://www.corante.com/many/archives/2004/02/14/vote_links.php

http://journalism.nyu.edu/pubzone/weblogs/pressthink/2004/04/07/atrios_kos.html

 

 



[1] Tim Berners-Lee Biography:  http://www.w3.org/People/all#timbl

[2] Ted Nelson on Two Way Linking:  http://ted.hyperland.com/buyin.txt. 

[3] Latour, Bruno.  Pandora's Hope: Essays on the Reality of Science Studies  (Cambridge: Harvard University Press. Chapter 2: Circulation Reference: Sampling the Soil in the Amazon Forest.)

[4] Latour, Bruno.  ‘Visualization and Cognition: Thinking with Eyes and Hands’ (from Knowledge and Society: Studies in the Sociology of Culture Past And Present, 1986, JAI Press Inc., 6: 1-40.)

[5] The Free On-line Dictionary of Computing: http://dictionary.reference.com/search?q=00-database-info&db=foldoc.  © 1993-2003 Denis Howe