A smarter network than you?

Weinberger makes the point, over and over again, that “knowledge is becoming a property of the network, rather than of individuals who know things, of objects that contain knowledge, and of the traditional institutions that facilitate knowledge” (182). By attacking our epistemological assumptions, Weinberger is showing us just what the internet is capable of providing seekers of knowledge. He goes out of his way to disavow “technodeterminism,” but holds that “there are some basic elements of the Net experience shared by almost anyone… that bear directly on how we understand knowledge” (174). Although Weinberger then describes several attributes of the internet which affect its scholarly utility, they are all more or less presented in a positive light. A truly objective, non-technodeterminist viewpoint, however, would be able to acknowledge that the internet’s evolution of discursive modes is a lateral shift which brings many potential problems as well as benefits. That these changes affect the very nature of our perceptions of knowledge makes this topic pertinent to all.

The abundance of material online creates a situation where paywalls, often the lifeblood of any publishing firm hoping to provide fair compensation for decent writing, lead to information “ghost towns”. Weinberger writes, rightly, that most people encountering a paywall are likely to use another similar, albeit free, source (175). There is nothing about the internet which makes it fundamentally biased towards open-source everything. Content creators will still need to eat, yet the abundance of similar-yet-slightly-inferior content means that audiences will have no real reason to pay. Additionally, Weinberger states that “links erode authorial control” (177), by providing any given work’s source material at the click of a mouse. Although facilitating quick cross-reference of source material is vital to verifying a scholar’s work, links can just as easily be used for cherry-picked counter-arguments and straw men. Ask any of the tutors at Oxford — the learning process frequently requires interface with one single authority. The idea is to digest the syncretic work of one, not only with more exposure and expertise, but optimally situated to guide the learner through their own particular society.

Oxford gets brought up by Weinberger himself, when he points out that “a notice that 99.9 percent of an eBay seller’s 14,000 transactions have been rated as satisfactory is a better guide than knowing that the seller teaches at Oxford” (188). However, this is a clear false equivalency. Weinberger mentions earlier that books came embodied with their own kinds of authoritative metadata, simply within the quality of their binding. What, then, could an eBay seller’s satisfaction rating compare to the physical market where you are able to fully inspect any and all goods you wish to purchase? True, the internet is eroding traditional notions of credentials. But the principles behind credentialing remain: there is a need to verify the truth by providing a certain richness of metadata.

Comprehensive metadata for the whole web is still just a dream, however. Even if the idea of a Semantic Web formed through “linked data” were available (187), it’s not clear that real knowledge would be more readily available — the network would simply be able to tell which knowledge would be most relevant to your own particular perspective. Although Weinberger has some bright ideas on how the internet is changing the nature of knowledge, he spends too little time pondering how this fundamental shift might, in many ways, leave people in the dark.

Tearing Down Skyscrapers for Interstate Highways

In reading the second half of Too Big to Know, one very important message shone bright – that the nature of science is changing, and so, our outlets for science must change with it. I believe that, with the formation of the scientific internet, that major scientific print publications must switch to the web or be left behind.

I believe such a dramatic shift must occur for three main reasons – firstly, that such a change would quicken the publication process, and thus the scientific process, significantly; secondly, that such a medium would allow for a bi-directional flow of information, thus improving the speed of discussion, iteration, and scientific advancement; and finally, that such a change would allow researchers and scientists to harness the massive untapped potential of amateur scholars.

Currently, research and publication is severely hindered by the speed of print. Whereas an article could be published instantly on the internet, a print publication can take multitudes longer. Victor Henning, creator of Mendeley, an online publication sharing platform, aptly states the problem in saying “if I’m writing a paper today it takes a year to get it peer reviewed and another year to get it published”. The fact that it takes two years in order for modern information to get to its consumers makes the entire process prohibitively expensive in terms of progress. Conversely, Mendeley, as previously mentioned, allows for immediate publication and discussion – a huge perk to the scientific community, as shown by its “450,000 registered users and 33 million articles” a mere twenty months after its conception.

However, as if the fact that this information gets to the populace immediately weren’t enough, an online publishing platform would also allow for immediate fact-checking, criticism, and further progress. This is fantastic, because it allows for immediate feedback and development, as opposed to the long and arduous process of print publication. As witnessed in Deolalikar’s proof of P = NP, “within two days, Delolalikar’s paper suffered the fate of prior attempts … both armchair and professional math pundits proceeded to tear it apart in comments sections and subsequent blgo posts, finding major flaws.” Had Delolalikar’s paper been published via print, not only would it have taken years to enter the public sphere, but upon entry, it would have taken years further in order for a formal response to be published. Rather, we skip these four plus years of peer review and bureaucracy, and instead directly enter the meat of the matter.

But wait, there’s more! This process also allows for immediate and organic collaboration with hundreds, if not thousands, of interested scholars. While they may be amateurs, the probing minds of this horde strictly outperform the handful of brilliant minds belonging to peers and colleagues. While these amateurs may not be breaking any new ground or establishing any new theories, this teeming mass is more than happy to fact-check every last detail. Albeit with a slightly disappointed connotation, Weinberg puts it succintly: “These amateurs are not tearing down the wall of credentials within which the scientific community lives. They are extending the apparatus of science.” It would be foolish to let such a powerful resource go to waste.

I made a city metaphor in previous blog posts, describing the science of old as a city built tall and narrow, and the internet as one built flat and wide. To return to that metaphor, this shift would be the equivalent of taking a tower-top restaurant with elevators slow to rise, and swapping it out with a huge drive-through on an even bigger highway. Sure, it’s a little less fancy, but the tradeoff in throughput makes it more than worth it.

Changes in Knowledge Creation

When I was younger, I had a junior encyclopedia set. Everything presented in the set was knowledge that was agreed upon by whoever published the set. The books included all of what the authors deemed necessary and excluded everything thought to be unworthy of publication. The information in that set was the truth, and that was end of it.

With the rapid increase in use and popularity of the Internet, the amount of information that is presented as true has grown. Anyone can present their analysis of a subject online. Amateurs can argue for hours with professionals on social media websites on nearly any issue, but they also work together online. Either an amateur or a professional can post their findings online, but now it is up to the reader to analyze the research’s authenticity. This is an issue in online knowledge creation. In Weinberger’s Too Big to Know, he mentions the controversial figure Jenny McCarthy, whose large media presence has been used to convince parents that vaccinations are dangerous and can cause autism in their children. This is not based on anything resembling science, yet many viewers perceive the medical advice from the actress/model McCarthy as knowledge.

Another part of Weinberger’s discourse on open knowledge creation is the idea of public research. There is the example in Chapter 7 of Jean-Claude Bradley’s “UsefulChem,” which was an attempt to document Bradley’s lab’s work and relay the information to the public. This is an important development in the realm of knowledge creation for a few reasons. First, this exercise teaches the public how the research method actually works. An interesting observation that Weinberger makes is that journals rarely publish boring or negative test results, yet most research projects conclude with such results. Second, the exercise alters the traditional temporality of the research results. In the past, time consisted of two parts for the public in reference to a research project: the time before the research results were published, and the time following the publication. With a live experiment like Bradley’s, the public could now follow his research as it went, well before publication of the results. This would allow the public to use Bradley’s findings to theorize for themselves. Having such an open platform could take away credit from those doing the experiments, but more knowledge is being created and science may actually benefit from it. It is an interesting balance that will need examination in the coming years.

 

Openness

The Internet is like a big playground, and it just seems to be getting bigger. More and more people are joining conversations in the Internet and some are wondering if certain people should be allowed to play or not. This openness has allowed people that were previously untrusted with knowledge to gain access to tools, information, and ways of communication to collaborate and voice their ideas. As Weinberger says, this has not eliminated the need for professional scientists but has certainly grayed the area because of the mix of public and the “professionals” (131).

With more contributors, there is of course an abundance of knowledge and information with the Internet. As we have discussed before, there is an inherent problem that comes with the openness. We have to acknowledge that not everything is truthful. We often have to dig through layers or search a long time to find the “best” information. As summed up by Weinberger, knowledge has become a playlist and not a library. We search for something and we get results, similar to looking in a genre for songs.

Not only is there an abundance of information, as networks are created between people and groups of people, authorship is beginning to lose its grip. If more minds collaborate on an idea, everyone may be able to benefit because a better product can be produced. In the past, we have limited the production of ideas to the original people that thought them up, not allowing changes that could make it better until a certain time. Even now, copyright laws have been strengthened to encourage this thinking. We need to allow the public access to these ideas so that collaboration can produce a better idea for the general public.

We need to learn how to play well with others. We need to learn that as a group, we can make good ideas better. Of course, there are downfalls but the public will benefit greatly from this type of openness. There will always be a need for professionals but amateurs have much to offer as well. One thing is for sure, the playground is only going to get bigger.

Global Digital Humanities: Qs for 3.26 class discussion

  • Propose 1-3 questions, based on the practicalities, goals, rationales, and/or challenges of global digital humanities and/or Dr. Gil’s own pathway and interventions on this score, in response to the course prep materials for this week. Links below.
  • Share those qs by adding them in the comments section to this post.
  • The questions should be ones that would be useful for us to discuss with Dr. Alex Gil, this week’s Intro to Digital Humanities visiting expert at 1:30 p.m. Thursday 3.26
  • We ask that you post them here by 10:00 p.m. on Wednesday evening 3.25 
Links: Course Prep for 3.24 & 3.26 (links on d2l as well):

The Examining th Small Details in the Big Picture

A liberal arts degree in the 21st century has become a common point of mockery especially when speaking to my more scientifically enclined counterparts. I am constantly surrounded by engineers and I cannot state how many times that I have been asked “Why major in History?” or “What do you plan to do after you graduate?”. The answer for me is simple. Everytime, I simply reply I want to teach the young minds of america about the importance of recalling our past and the power that ability to remember and corelate has. As a liberal arts major, I have spent a lot of time justifying my existence which I why the introduction of a technology which can basically in the eyes of my non-liberal arts peers, can do my job and can do it much faster.

This introduction of what we liberal arts folks call distance reading, could very well spell the downfall of my profession. If a student can look up an in depth literary analysis of a historical account and gather a sort of pseudo summary of that account then what is my purpose within this relationship between student and their acquisition of knowledge? What benefit do I have against a computer?

The answer is simple for me, the distinction between close and distant reading is akin to how Ramsay views the differences between searching and browsing.  One allows a person to actually witness and fully experience a single piece, to be intimate with and understand it. The latter provides much the needed context regarding the general issue and give the viewer a more stable base. This of course allows one to know where to place and how to interpret the information of the former.

It is therefor imperitive that as a future scholar of liberal arts one has to master both ends of the spectrum. We have to learn how to use these scientific advantages in order to adapt to the modern times that we live in. After all, after speaking to older generations of historians they discussed how their generations boasting main skill wasn’t necessarily the in depth analysis of a data set, but instead they were more proud of their ability to just acquire that information. I had a professor boast about her ability to travel to Russia, translate a book into english and then use that translation as evidence for one of her theories in her book. In today’s generation our main ability is not the broadness of our research, that is a given thanks to the wonder that is the internet, instead we attempt to over analyze the data we discover. Hence the generation of the post-modern field of history. A field dedicated to a new found analysis of history that had at one point been believed to be completely true, and for the most part was not widely disputed.

Its therefor obvious that the debate of close versus distance reading isn’t necessary for liberal arts scholars. Instead we should hurry and accept the fact that they both have benefits in regards to research, and that if we plan to keep up with the pace of knowledge then we must adapt to the new methods of acquistion that these hard sciences can offer to social sciences.

It’s All Just Reading

In class, I made the assumption that distant reading only refers to the digital methods of textual analysis we’re being introduced to. When Dr. Purcell corrected me, saying that distant reading might also apply to the wide breadth of reading a traditional researcher focused on a particular canon does, the issue got a bit more confused for me. But it makes sense: the types of connections drawn, coupled with the far-sighted biases of a human working with at least a shelf’s worth of text, do indeed qualify some traditional, physical-book readings as “distant”. Where, then, the line between close and distant reading lies, I’m not so sure. What’s more, I’m not sure it’s very important.

Interpretation of readings is highly subjective. Whether or not close or distant reading is good or bad depends entirely on the situation and desired outcome. I think in this day and age, it would be fair to say that any type of reading is “good”. But good for what? And how?

Close reading, forging an intimate connection with a text through sustained, focused attention to its subtleties, is probably best for individuals looking for singular diversions and/or wisdom. But historical inertia has steadily pushed us further and further away from the raw essence of a text. What, exactly, was lost when humans stopped memorizing and reciting vast tracts of oral literature, like the Iliad or Odyssey? Something in the rhythm of the words, the rhyming, their specific cadences and patterns. Something intangible that nonetheless added important meanings to the text. A similarly undefinable quality of the text was probably lost when, in the Middle Ages, readers shifted from mouthing out the words they read to reading them silently. Now, people complain that e-books are stealing the distinct odor of books, especially old books, which has spiced the study of bibliophiles the world over for millennia. All of these qualities are comprehensible in a system of “close reading”, and they speak to the principal benefits of close reading: the ability to envelop the reader in an all-encompassing world, to stimulate not just the mind but the body itself. Texts have become disembodied in our digital age, to the point that there are no real qualitative differences between the words of Marcus Aurelius on Project Gutenburg and the words of a middle-school blogger on Tumblr. This may seem like a strange thought, but considering the mostly subjective benefits of close reading, I think it’s relevant.

Distant reading, on the other hand — the work of scholars for millennia — offers different benefits. Distant reading allows contextualization and synthesis of many disparate texts and ideas. We go from a small ship exploring the vagaries of the coastline day by day, to a plane with a bird’s eye view, or even a satellite taking full-resolution scans with each orbit. The analogy makes even more sense when we realize the cartographers on the boat would be able to give us information on the type of fish available, the nuances of the ocean current, the exact quality of the weather and many more things, while the plane or satellite could only give us a less detailed, though far more broad and probably accurate, idea of the nature of the coastline. Distant reading is perfect for evaluation and analysis of many different texts. However, it has a downside in the sheer scale of its domain: the more information you have, the harder it is to put it all in a comprehensible order, and the easier it is to draw false conclusions from what’s before you*. Still, distant reading is great if you’re trying to understand a large body of text. Virtually every class I’ve ever taken has relied on distant reading, as the professor tells us the many interesting and relevant things s/he has picked up in her/his research. Distant reading, then, is probably better for groups trying to make sense of textual collections that no single individual wants to read themselves.

*This website, Spurious Correlations, gives a great example of one of the pitfalls of distant reading: http://www.tylervigen.com/

Close and Distant Reading

In the history of humankind, there’s only been a very brief span in which individuals could be well-read enough to be familiar with every book written; this happened so long ago that the concept today seems completely ridiculous and laughable. Since that time, we have constantly struggled with which books to read in our limited time, a distinctly human condition.

However, in the past 30 years or so, we’ve made a new advancement, in our ability to aggregate and analyze huge quantities of books. This art of “distant reading” allows for understanding to be drawn from millions of books at once, but brings forth another dilemma.: do we read a hundred books closely, or a million books distantly? It’s a difficult question to answer, but I’ll do my best to briefly explore it in the next 500 words.

I view the distinction between close and distant reading similarly to how Ramsay views searching and browsing: the former allows one to witness and experience a single piece, to get truly intimate with it and understand it, while the latter provides much needed structure and context, allowing one to know where to place and how to interpret the information of the former.

When I think of close reading, I think of reading in the classical sense – reading by candlelight, with notes scrawled in the margin, reading until your eyes are weary and your thoughts grow dim, falling in love with a book, or perhaps not, and trudging through it anyway, to see the road less traveled. As to be expected from its multi-millenia reign, close reading has plenty of merits: it captures a first-person interpretation of the period, it holds near-limitless qualitative depth, and it’s a ton of fun (an often underrated quality).

However, close reading doesn’t come without drawbacks – without either some existing knowledge or research done during reading, it’s pretty hard to pinpoint what inspired this piece, and harder still to know the piece’s impact after its creation. Drawing inter-piece and cross-cultural parallels becomes significantly harder, without an understanding of the material at hand, and even then, requires very precise knowledge to do it well.

Of course, this is a rather narrow interpretation of close reading, basically limiting the act of “close reading” to what happens between the covers of the book, with the existing knowledge and expected research that goes into the analysis considered separate actions – this is for good reason, which will hopefully make sense in a second.

And so we come to distant reading – the antithesis, in which a book is never touched and a story never weaved, all for the tradeoff of “reading” books thousands of times faster than humanly possible. It’s wickedly efficient, and takes all the cons of close reading pretty easily in stride, but at what cost? A purely distant reader, with their brilliant understanding of historical context, cultural parallels, and far-reaching effects, never knows the amazement of exploring a world, the excitement of talking it over with a friend, the satisfaction of closing a book. They entirely remove the human element of reading, making it a cold, calculating, exacting science.

This too comes with drawbacks – despite being automated by computers, even algorithms have biases and interpretations, but with the unfortunate scientific weight to be “non-debatable.” However, that’s a discussion for another day

With their pros and cons so conveniently complementary, I can only believe that reading is done best with a combination of the two strategies. Understanding the structure, and appreciating the art. Browsing the library, while finding the book.

How Beneficial is Distant Reading

Distant reading: it’s something that I’m sure we have all done.  Quiz on an assigned reading?  Let’s speed read it.  Want to refresh your knowledge of a given subject?  Let’s glance over it.  Distant reading is something that we likely do every day.  But how is there any merit to it whatsoever?

Close reading is obviously a much better tool to use.  In class on Thursday, I exemplified how close reading can more accurately determine context when text mining.  A distant read of the data might show the context is cuisine.  While cuisine could definitely play a role, it is the way that the context of the joke is actually delivered.  A close read is able to ascertain that the true context of the joke is the economy.  While the distant read did not reveal what the close read did, does that mean that its use is without merit and, thus, a waste of time?

I would suggest that this is not the case.  In jokes, it is important to understand how the comedian delivers the joke.  In this case, he ironically delivered the joke by disguising the true context with Chinese food.  The method he delivered the joke is not as important as the context, but if you understand both, you might be able to determine more info, such as if there are themes present in the way things initially look and how they really are.  This is definitely beneficial in many different ways.  Also, if you have a strong understanding of a subject, you might be able to do a more distant read than a close read on a paper or project.  But, it is extremely important to not only do distant reads, but close reads.

In other words, my main point is this: distant reading can be very beneficial when used in conjunction with close reading.  With that being said, in academia and research I do not think that distant reading is good.  Research takes time.  You cannot adequately give the subject justice by working on it for only 5 hours.  It takes a lot more time than this.  In fact, it might cause your field to be less-respected.  As it was said in class, any conclusions that comes out of distant reading is by chance and “b.s.”  As hard as this might be for students, this is something important to hear.  You can’t do your work justice by going “half-a****” on it.  Work hard on the project and give it the amount of time that it actually needs.

Distant vs. Close Reading

Distant and close reading techniques for discovering new information from known texts have their advantages and disadvantages.

Distant reading is a technique that is used to scan large amount of texts for information. This is a technique that is useful when one wants to examine volumes of work over an extended time period. Computer programs can be used to scan thousands of works with millions of words and phrases to spot common trends. This is a large advantage for distant reading. Distant reading can be automated. This method involves less manpower than close reading. More information can be found at a quicker pace.

However, there are complaints that this information is shallow. If a program searched works in a given time period for some specific phrase, the returned works or excerpts from the works may not provide enough context to give adequate information about search terms. There is also the issue of our ever-changing vocabulary. Words do not necessarily have the same meanings as they did in the past.

Close reading examines a smaller amount of texts. Close reading is used to find “hidden” trends or meanings in texts that a method like distant reading might miss. This is a method that is used when examining smaller amounts of text than in distant reading. Currently, close reading is very human dependent. Computers are not currently able to spot some hidden messages or decipher human metaphors. For example, the many hidden meanings in the Shakespeare sonnet that we read in class would likely not have been found by using some computer program to scan the text. The technique of close reading also allows the reader to have context. Again, if a reader knows that he or she is reading a sonnet instead of a historical novel, the reader can look for different trends or meanings. The sonnet may have a specific rhyme scheme or theme that would not be picked up by a text-scanning computer program.

However, as great as close reading sounds, the technique has a cost. This technique takes a lot of time and manpower. The close reading experiment we did in class with the Shakespeare sonnet was great, but we took almost 30 minutes to examine a fourteen line sonnet. Shakespeare himself wrote 154 sonnets. If we were to examine each of these sonnets in the same amount of time, this would take 77 hours. This amount of time, spanning over three days, would be devoted only to the examination of Shakespeare’s sonnets. While Shakespeare was a great poet, and these are probably the highest quality sonnets, these are only a percentage of all sonnets, or poetry for that matter.

In any case, these methods have their advantages and disadvantages. I am not sure if one method is better than the other. Which method being used should depend on the situation. If one wants to examine a lot of information for key words or phrases, distant reading should be used. If one wants to discover new meanings or patterns in a work or group of works, close reading would be the better option. These are both interesting, useful techniques that should be used when each is necessary.