But it got me thinking.

Why entrust to even a single group of administrators decision-making ability for hundreds of campuses? We can engage in a little creative destruction in the interest of leveraging the efficiency gains of an algorithmic approach to administrative decision making. No need for any humans to be involved at all. Netflix, eHarmony, Amazon and the rest of our technological overlords have already showed us the magical future of “the algorithm.” Sure, Siva Vaidhyanathan has warned us about such googlization. But, in the case of university administration, I can’t see how much more harm could be done by automation.

So, let’s design this algorithm.

Here’s a start:

adminHealth = 0
Policies = [policy1, policy2, policy3, policy4]

def policyChange(policy):
    x = policy()
    if x == helpsFaculty:
        return adminHealth -= 1
    elif x == helpsStudents:
        return adminhealth -= 1
    elif x == helpsAthleticDepart:
        return adminhealth += 3
    else:
        return adminHealth += 5 

while adminHealth <= 1000000:
    for policy in Policies:
        policyChange(policy)

Please, add some new functions so we can get this algorithm right. That way, we can dismantle the university much quicker, and the denizens of the algorithmic future can feel good about themselves as they save us all from the inefficiencies of, you know, values, morals, leisure, depth, thought, consideration, and all the rest.

I’m doing a lot of reading as well, and taking notes. It’s been a while since I spent a lot of time in DEVONthink, and in the interim I’ve been completely converted to vim and markdown. I very much like taking notes in vim, leveraging the pandoc vim plugin and its tab completion of bibtex citations. The thing is, for some reason I much prefer vim in the terminal to MacVim. I don’t know exactly why, though I suspect its because vim in the terminal is much more integrated into the programming workflow I’ve been using the past year.

It’s very easy to open a DEVONthink record in an external editor. I take all my notes in DEVONthink as plain text notes, part of a commitment I made a couple of years ago to using plain text whenever possible. MacVim is the default editor for .txt files on my machine. In DEVONthink, you can open any file in your database with the system default application either by clicking a toolbar icon or using the keyboard shortcut SHIFT-CMD-O. But, what about terminal vim? A little more complicated.

So, I cobbled together a Automator application from posts here, here, and here that opens a highlighted record from DEVONthink in vim in a new tab in the terminal.

For my own documentation, and maybe to help you out, here’s how it works (click on the images to make them larger):

1. Open Automator and select a new application.
   
2. From the Automator script library, first drag over the DEVONthink action Get Selected Records.
   
3. Next, drag over the DEVONthink action Get Item from Records.

   


4. Go to this page and down load the file Get File Path, which is an Automator action to get the full file path of any file.

   Once you file is downloaded, double click on it to install it in your Automator library. Then, drag into fourth place on the workflow.

   

5. Now it’s time from some Applescript. Once we have the file path, we pipe it to a script that opens a new terminal tab, and then opens the file in vim. So, drag a Run Applescript action to the workflow and enter this code:

   on run {input}
   	set the_path to POSIX path of input
   	set cmd to "vim " & quoted form of the_path
   	
   	tell application "Terminal"
   		activate
   	end tell
   	tell application "System Events" to tell process "Terminal" to keystroke "t" using command down
   	tell application "Terminal"
   		delay 0.25
   		do script with command cmd in front window
   	end tell
   	
   end run
   

   

6. Save the file in ~/Library/Application Support/DEVONthink Pro 2/Scripts. Name it something that makes sense, like OpenInVim.

That’s it. Now, you can select that from the scripts icon off the topbar. Or better yet, set an application-specific keyboard shortcut, which you can do in the Keyboard section of the System Preferences. I set mine to OPTION-CMD-o, so as not to clash with DT’s open-in-external-application shortcut.

Following last week’s announcements of new contracts between Coursera and a whole host of public institutions (coverage at IHE, CHE, NYTimes), I requested a copy of the contract signed between the UT System and the online MOOC provider. Tennessee has a public records act, which I mentioned in the email asking to see the contract. When I opened my email this morning, there it was. (Thank goodness for sunshine laws and the temporary good sense behind them.)

I have only skimmed it thus far, but I’m putting it up here for others to read. I’ll come back with more of my thoughts on the contract once I’ve had a chance to read it more closely.

A couple of notes of explanation:

1. The contract reveals the extent to which Coursera is looking for new opportunities to leverage its Platform (videos with embedded quizzes and a discussion forum — is that really a platform?). This contract is quite different, at least in the stated goals of the System, from others we’ve seen, like the one signed by the University of Kentucky system.

2. This contract was signed by the University of Tennessee system, which includes five campuses: UT Chattanooga, UT Martin, the UT Space Institute (Tullahoma), UT Health Science Center (Memphis), and UT Knoxville. The UT System is governed by a Board of Trustees. There is a second system in the state called the Board of Regents system, which includes campuses like East Tennessee State University. The Board of Regents system has also signed a contract with Coursera, though I don’t know how much it differs from this one.

3. The contract covers an 18-month period during which UT will evaluate Coursera’s platform as an alternative to its current online course technology provider (which I believe is Blackboard Connect). Thus, we’re talking about implementing two courses with multiple sections across campuses. I have a problem with that from the start house. Speaking of which, my favorite quote thus far…

The Platform will support cross-institutional simultaneous enrollment at Institution in a single-class instance to allow for the creation of larger cohorts.

So, without further ado, here it is: Coursera Contract

Please leave any thoughts in the comments below, or on twitter, where I am @parezcoydigo.

For a final project, I gave the students the option of writing a paper on three films, critiquing the films using Matt Restall’s Seven Myth. I also gave them the option of forming small groups to put together a choose-your-own-adventure text game on some aspect of the conquest. An intrepid few chose the latter, mostly because they were fatigued by the notion of writing papers. Given the short time span, I was particularly impressed with a couple of the games.

We used Twine to write the stories. Twine is nice because it is multiplatform, and uses simple wiki syntax to construct the story. That’s because Twine is essentially a wrapper around TidlyWiki. It’s intuitive, and easy to work with, as long as you remember to name the first entry “Start”.

The students essentially turned the classic document collection Victors and Vanquished into text games, and conveniently one group chose the Spanish perspective and the other chose the Mexica. Click through those links to see the finished products.

One note on doing this– the process of writing such a story lends itself to reproducing the myth of exceptional men, in which individual decisions and actions are preeminent in making the conquest. We talked so much about that, though, that the groups noted it when they presented their work today.

At any rate, as a short order experiment over just a few weeks of class (15 class days!), I think it was successful. I’ll definitely do it again.

During this break, I’ve been grinding through neglected inventories of cases that form the archival core of my current work on popular sexuality in the Spanish Empire. These cases are drawn from the Archivo Nacional del Ecuador (ANE), and are drawn from the Criminales, Civiles, Matrimoniales, Residencias, Fondo Especial, Prisiones, and assorted other Series from the ANE. The most important of the cases are from the Serie Criminales, of which my case set counts some 900 of about 3400 extant criminal cases from the colonial period. The Serie Criminal for the colonial period includes cases dating from 1589-1820, though through the middle decades of the 17th century, there are very few surviving cases. Of the 900 cases I’m using from the Serie Criminales, 333 are sex-related cases, in which the individuals being investigated are accused of sex-related acts. The other 600-odd cases relate to contextually interesting criminal prosecutions: 1. drunkenness; 2. domestic violence; 3. insults; 4. magistrate abuse of power; 5. gambling; 6. horse racing; 7. murder and assault; 8. etc.

I’ve chosen these cases in order to better understand the nature of sex-related prosecutions, and their potential for an ethnohistorical-like investigation of popular sexual norms. I decided in the process of putting together my inventories of cases to keep track of the relative frequencies of sex-related prosecutions in the criminal archive. I then plotted these frequencies across boxes and years. As with many archives, the ANE Serie Criminales is divided into numbered boxes (cajas) full of individual cases folders (expedientes). It is impossible to draw too many conclusions from the cases that have been preserved over the centuries because we have no way of knowing what’s behind the particular preservations. Was part of the archive damaged or destroyed in a previous age? Were some magistrates or notaries better or worse at collecting, preserving, and passing on their files? Likewise, we have no cases from the first 20 years of Quito’s existence as an Audiencia, and only a smattering into the early 18th century.

Nonetheless, charting the cases in my archive does show some interesting trends. So, two charts for you. (Click on each chart to get much larger images.)

1. Total and Selected Cases, 1601-1820:

This chart includes both raw counts and rolling 15-year averages of case counts for the total Serie Criminales and my 900-case subset. It’s divided by archival box, rather than by case year– which gives an impression of a much more consistent archive size over the years. Note that there are two periods where my selection of cases deviates a bit from the total case trend line: 1. from around box 45-55; and, 2. from box 185 onward. Temporally speaking, boxes 45-55 correspond to the period roughly between 1755 and 1765, the early years of Charles III’s reign (1759-1788). Box 185 onward takes us from 1802-1820, through the end of colonial rule. (Ecuador officially became part of Gran Colombia in 1822, but I’ve truncated at 1820 for convenience only.) This was an era of liberal ascendance and crisis. That said, in the raw counts its visibly the case that cases I find interesting begin to decline in the Box 140s, or around 1790, which corresponds to the end of Charles III’s reign. I should have plotted this with a different trendline. Why a 15 year rolling average? I don’t know, it’s my own experience that 15 years might be a chunk of time that produces a sort-of epoch of prosecutorial priorities.

2. Total and Sex-Related Cases by Year, 1601-1820:

Here I’ve changed the x-axis from box to year, and plotted only sex-related cases. It’s much more obvious in this chart that the Serie Criminal is really an archive of the 18th century. The first real uptick in cases begins around 1722, the second around 1743, and the big explosion amps up in 1779. By the way, the second peak corresponds to the first decade of the 19th century. I’ve also included 4-year rolling averages in this chart. Sex-related prosecutions peak in the same general time frame as overall prosecutions peak, but their percentages are anomalously high. As a percentage, sex-related prosecutions peaked in 1788 as 27% of the extant criminal cases in the Serie Criminales. This comports with jail censuses.

You’ll note there is a little bump in the trendline for Sex-Related cases earlier in the 18th century. Right around 1745, when for a cluster of a few years those cases again reached into the 20% range of total cases. What do the 1740s and the 1780s have in common? Well, for one thing they have periods of epidemic disease and natural disaster in common. Could natural disorder account for a rise in sex-related prosecutions, when a reformist government might be interested in making order?

The devil in that question is in the details, and it’s interesting to note that prior to the 1780s, sex-related prosecutions were dominated by instances of forced rape, incest, and cases involving young women. In the 1780s, cases are dominated by descriptions like illicit relationship, concubinage, and adultery. This suggests the state takes a very different interest in sex acts at the point that prosecutions dramatically increase.

What am I doing with these cases?

This project has evolved over time. What I’m actually interested in is both methodological and historical– understanding popular attitudes towards sex through prosecutions. So, I’m taking especially cases of concubinage and adultery and looking at a number of factors. How do magistrate and defendant portrayals of these relationships differ? How long did relationships last before they were brought before legal authorities? What were the circumstances that made relationships prosecutable– what accounted for their transition to notoriety? What I’m finding is that defendants use arguments from customary practice that are similar to legal defenses across the spectrum of criminal and civil litigation. I’m also finding that relationships were very frequently long-lived, often producing children and never drawing the attention of the state until something outside those relationships sparked a conflict. This could be violence, nightly patrols by judicial authorities, commercial disagreements, and any other number of conflicts that had little to do with sex and much to do with other aspects of neighborhood tranquility. What to make of that? The cases paint a picture of toleration without tolerance in popular sexual attitudes in the eighteenth century.

I’m also finding that as the 18th century wears on, the insults people hurl at one another become increasingly tied to racial terms, including those hurled at women. More and more, phrases like “(s)he treated my like a zamba/india” show up. But, that’s another post altogether.

I want to get this working because, in part, I want to reproduce the lessons Matt Jockers presented on text analysis in R at the Digital Humanities Winter Institute.¹ The Python Data Analysis Library provides the kind of data structures needed to do the kind of analysis offered in R. At any rate, it brings me back to matplotlib again.

So, I have tried the prescribed solutions to the problem of installing in every imaginable combination– installing and uninstalling dependencies in different orders, to no avail. I was doing this using OS X’s built-in python, which with Mountain Lion is 2.7.2 and serves my purposes, generally.

I finally decided to try using a virtualenv, and magically matplotlib started working. So, here’s how I did it:

1. Using the package manager homebrew I installed libpng, freetype, and pkg-config.

   $ brew install libpng freetype pkg-config

2. Using virutalenv and the truly awesome virtualenvwrapper to set up a fresh python instance just for this kind of text analysis.

   $ mkvirtualenv text_analysis
(text_analysis) $ pip install numpy
(text_analysis) $ pip install scipy
(text_analysis) $ pip install pandas
(text_analysis) $ pip install nltk
(text_analysis) $ pip install ipython
(text_analysis) $ pip install tornado
(text_analysis) $ pip install zmq

And then, finally,

(text_analysis) $ pip install matplotlib

The only package of those listed above that is required for matplotlib is numpy, but the others are useful for what I’m working on.

Whereas doing this exact same thing with Mac’s built-in python produced constant Import Errors, regardless of whether I installed matplotlib with pip, with another package installer, or building from source.

I’m glad to have it working again. If you’re having trouble getting matplotlib working, then try installing in a virtualenv, and by all means use virtualenvwrapper to manage your python envs. This may well require a change in your workflow, but it will be a change for the better. It’s good working practice to isolate your development environments, and better than sudo site package installs on your system python.

A couple of words on managing python environments:

1. To ensure your scripts will work with the activated env, make sure to use the correct shebang line at the top of the file:

   #! /usr/bin/env python

2. As you get used to using envs, you may find a few standard sets of site packages you use again and again. pip allows you to install packages from a special requirements file. While working in an environment, such as my text_analysis environment, you simply use a pip freeze command:

   (text_analysis) $ pip freeze > requirements.txt

Then, in a new env you simply bootstrap with pip:

(new_env) $ pip install -r path/to/requirements.txt

If you’re interested in using ipython or ipython notebooks, you’ll also need to install readline. The ipython folks recommend installing it with easy_install instead of pip, because it works better for reasons I’m not entirely sure of. The process is the same as above:

(text_analysis) $ easy_install readline

Finally, virtualenvwrapper makes it very easy to list site packages (lssitepackages), make new envs (mkvirtualenv), and list and switch between them (workon). For more, see the documentation linked above.

One might claim to see a parallel between the dotcom bust and the general loss of academic standing by critical theory….

and

[Roland Barthes, Jacques Derrida, and other critical theorists'] approach to textuality remains very helpful in understanding our experiece of hypermedia. And vice versa. *

 in connection with my recent piece on long-form historical narrative. The context of these comments comes with Landow’s admission that poststructuralism waned in the academy, even as the read-write world of Web 2.0 exploded. He’s right in his contention that poststructuralists provide excellent analytical tools for understanding hypertext and the read-write web. It’s hard to imagine a more perfect example of Deluezian/Foucaultian emergence than the explosion of hypertextual writing in the post-dotcom era, materially manifest in the LAMP stack (linux, apache, mysql, php), and served to the world through the well-masked control of Protocol. 

Historiography always maintained a tenuous relationship to poststructuralism. And the demise of the latter has been celebrated by many Historians if only in continued silence and a smug sense of having “correctly” ignored it all along.  

Historians, to generalize to a ridiculous extent, simply struggle to incorporate theoretical challenges that simultaneously lack historicity AND that set out to undermine narrativity.  

Which brings me back to my discontent with current options for long-form narrative online. Is my discontent related to the theoretical implications of hypertextuality itself?

*George Landow, Hypertext 3.0 (Johns Hopkins Press, 2006):xiv.

The session began with a reminder that accountability in higher education has “increased over the past decade and accelerated with the recent recession.” Due to pressures from accreditation agencies, like SACS, as well as I’m sure tons of other interest groups, “we nee to transition to a culture of assessment.”

Cultivating a culture of assessment in Higher Education is, apparently, a new thing. All the forms of assessment that professors use with their students aren’t what we’re talking about here. That sort of assessment doesn’t ensure students, parents, and legislators that learning is happening. What will ensure it is choosing a set of learning goals, writing a rubric to assess them and the programmatic level, and then of course make changes based on the result. I love how agencies assume as an opening gambit that such things aren’t already happening, and assume such because the iterative process of education isn’t written in the correct formal language.

What is that formal language, you might ask? It’s a language that is shared amongst institutions undergoing the process. To write learning goals and assessments correctly, one can simply google sacs accreditation and look at materials institutions put up on the web. A “Culture of Assessment” is really about a “Language of Assessment.” It’s one of many newspeaks in operation on campus today, and together with MBA-Speak and associated corporatization-of-academia-speaks, I find it simultaneously vapid and invidious.

Assessment-speak, like corporatist language on the university, claims to promote efficiency and accountability. But how and to whom? Institutions facing SACS re-accreditation are encouraged to construct an elaborate set of learning objectives using verbs tied to Bloom’s taxonomy of learning. That’s right, a significant portion of SACS approval hinges on the facade of a set of terms related to educational research published in 1956. The taxonomy is, of course, hierarchical and posits that students move to higher levels, from remembering to creating, as they learn more.

My experience is that students don’t move from lower to higher levels in a linear fashion, but rather iteratively and messily. Nonetheless, we get to write learning objectives that draw on the higher levels, and demonstrate movement to them. These are supposed to follow a particular form:

The student will verb object.

Despite the learning objectives drawing on the higher end of the taxonomy, they’re also supposed to be simple and measurable. I figure, the easiest way to do this, appropriate to the rigor of the process, is to automate it. So, as a service to history departments across the country facing regional accreditation, I present to you Bloom’s Picker, a simple web app that will write your learning goals for you. Just refresh the page. Three to five are usually sufficient!

The thing is, the sentences produced by the app almost always work and could be copy and pasted straight into an assessment rubric. This isn’t a commentary on the rigor of academic history. I’ll leave it to you on what it does measure.

DHWI was as a very interesting experience. I took a class on large scale text analysis in R taught by UNL’s Matt Jockers. There was tons of code, and a learning curve for me in getting used to R’s data structures and syntax. I’ll admit I struggled a bit with visualizing in my head the various types of matrices that constitute much of R’s data structures. I have more to say about this class too, but in another post.

I drove each day to College Park from Baltimore together with Alex Galarza. Alex and I talked a lot sitting in traffic on I-95. At some point during the week we ended up in a discussion on something I’ve been thinking a lot about in the last year or so– frameworks for long form (historical) narrative online. We both share some misgivings about many of the platforms currently in use in Digital History projects, including WordPress, Drupal, Omeka, KORA, etc.

I’m not a designer, and I think that I lack a vocabulary to explain exactly what my misgivings are about the current platforms and how they affect the potential for presenting long form historical argument. In the case of the platforms I listed above, in each case the idea of constructing a long form narrative on the platforms always feels like a hack. WordPress is a great platform for writing blog posts. It doesn’t take much, though, for an individual post to get to the tldr level. The UI of a blog sets limits, at least of expectation, on the length one will spend reading an individual piece. That’s not, of course, the condition of an Omeka or Drupal, both of which might be used in a blog-like manner, but which are intended at a more fundamental level as expressions of modeled relationships between entities that exist in their databases. The way that the elementary entities of the various CMSs are conceived affects the organization and delivery of information using that platform. Raf Alvarado gave an excellent talk on this very thing a few years ago at THATCamp Prime. I put up my notes from that session a few years ago, which was the best presentation I’ve ever heard of the connection between the ontology of a CMS and its resulting product. I bring it up here because I think that Raf’s idea about the way platforms (CMSs) model content has direct bearing on my own discomfort with them as current options for long form historical narrative.

Let’s take Omeka as an example. In Raf’s presentation he noted that for Omeka, the elementary unit is the ITEM. Items belong to Collections, and have Dublin Core metadata, keywords, and tags associated with them. Combinations of items, organized by membership in collections or linked through metadata, keywords, or tags allow information to be accessed through Exhibits. Raf described this, borrowing a term from Landow’s Hypertext, as “Axial Hypertext,” or a “sequential content model.” On the face of it, it may seem that a sequential content model would be the best for historical presentation, encouraging the reader to go through a sequence curated by the scholar through the construction of exhibits. Reading that sentence, I imagine you imagining a museum, which is appropriate given the genealogy of Omeka. But, it highlights what remains an unspoken divide in historical scholarship– between what is traditionally called “public” vs. academic history. The narrative of a public history exhibit, and that of an Omeka site as well, is intimately tied to Item objects. This isn’t the case in the kind of history I and most of the History professoriate write.

Long form narrative is not object oriented, to butcher a phrase associated with philosophy and computer programming. While historical writing is certainly evidentiary, it’s not a sequential presentation of evidential objects. Omeka doesn’t force one to parade objects, but it is predisposed to organizing information in that manner. I’m not trying to pick on Omeka either, because I have problems with all of the major platforms and the models they impose on the information-knowledge complex. They can also all be hacked for the purpose of long form historical narrative.

Why does this matter, even? We have epubs for tablet reading. Isn’t that the electronic substitute for the book? It is I guess, but it falls short for me for two reasons. 1. Epubs truly do cede design elements to the platform. 2. Epubs do little to utilize the interactivity made possible by hypertext. They’re an ugly simulacrum of the book. I also don’t like the movement of more and more information to apps, as opposed to on the open web. There isn’t too much I can think of about the epub experience that can’t be replicated on a well designed website that is by intention responsive to screen size and computer type.

This discussion cropped up on twitter a couple of weeks ago. Alex storified it. It was started by a query about allowing users to create their own narratives from archival collections. It’s a cool question, and one that is really similar to the idea behind Scalar or Scholars Lab’s plugin Neatline for Omeka.

I’m really less interested in existing or new platforms that can be coerced into long form historical narrative, especially those where evidence objects are the central focus of the platform’s “data object.” I’m more interested in general design principles for presenting either wide-ranging or deeply analytical historical work online. Digital History has focused more on the object of evidence, in part because of its roots in public history and in part because really the Web provided for the first time the capacity to share sources ubiquitously. As a result of that orientation, and with exceptions of course, DH has not offered much to historians working in a more academic mode. And that’s a shame. And that’s a shame that has affected the perception of DH in the traditional academy. If you doubt that, read the Introduction to Allan Megill’s *Historical Knowledge, Historical Error, which is essentially a scathing review of Valley of the Shadow. The problem, if it is one, is the extent to which Digital History has turned it’s practitioners into curators or archivists to the neglect of long-form narrative and analysis.

It’s my contention that part of that is platform driven. I’d like to call, instead, for an html/css/js framework for historical narrative. The success of a css framework like Twitter’s bootstrap is in part its platform independence. There has been a backlash against bootstrap, in part because of its success. And its success is rooted in providing a set of icons, css layouts, javascript ui that makes it easy for people without design chops to prototype a site quickly. I don’t, though, find bootstrap to be particularly readable for the kinds of sustained interaction historians are looking for. We’re not talking about web applications here, as much as a newly imagined book.

What would a framework for historical narrative include?

1. A minimalist interface.
2. Good typography that conveys the seriousness and temporality of the narrative. With today’s screen resolutions, I think we can go with serif fonts and large type. We’re not selling SAAS or any commodity, so why take our typographic cues from that world?
3. Navigation. Finding one’s place, and finding it again is important for scholarly reading. Together with navigation would be, I’d say, the ability to cite by paragraph. And, to be honest, I’m neither a fan of scrolling or pagination. But maybe that’s just me.
4. Javascript popups for footnotes/endnotes, and maybe for evidence objects as well. One of the things that hypertext has done to sustained reading is present a huge array of interesting and distracting linked elements. But, every time I leave a page to look at something, finding my way back and refocusing can be difficult. Better embedding would help this.
5. I’d really like some means to visualize embedded metadata/rdfa/microdata as well.

What else? What else would you want to see in such a framework? The upside of such a framework is it could be used with any platform, or by itself with book-length works written in Markdown. This isn’t exactly the direction the conversation on twitter was taking, but as Jeremy Boggs tweeted in the storified conversation:

@parezcoydigo @galarzaalex @anneperez @tjowens @wragge @miriamkp we’ve only scratched the surface of what we can do with good ol HTML & CSS

— Jeremy Boggs (@clioweb) January 23, 2013

With the ease of authoring with Markdown, I think he’s exactly right that there’s still much to explore with just HTML and CSS.

Flask is great, from my perspective, for a couple of reasons. 1. It’s in python. (I’m comfortable with python, so I don’t have to learn a bunch of new language conventions while also learning to build the web app.) 2. Even though as a microframework, Flask doesn’t handle a ton for you, there are already a bunch of extensions out there that help quite a bit.

The past two years or so, I’ve been building my course sites with hyde, a python static site generator much like ruby’s jekyll. I think that most of what we do with course sites can be done statically, with just a little client-side javascript. And, in fact, there is nothing on this new site that I couldn’t do that way.

So, why bother with Flask then? I don’t know. Shits and giggles? The chance to learn how to deploy a wsgi app on Dreamhost? Learning at a very basic level about routing views, app configs, etc. I still didn’t want to mess with a SQL database for the site, because other than tracking student posts there’s no point in it. Everything else my course sites need, at least for now, are better served as, essentially, static files.

So, individual pages for the site are all written in Markdown, and served using a flask extension named flat-pages. It’s a nifty plugin. When the app receives a url request, the path is checked against a folder named pages. So, for example, http://history475.chadblack.net/syllabus/ checks for a file named syllabus. Since it’s there, it takes the text of that file, which is still in Markdown, converts it to an html snippet and serves it through a page template. This is, in many ways, what a static site generator does, and what Frozen Flask does to produce static sites from a Flask app. (I’m going to do that when this course is over. There’s no need for an in-active site to be anything but a static site.)

This brings us back to a page of student post feeds. I require my students to each start a wordpress.com blog for my course. They write weekly, and I always have a place on the site that aggregates their feeds. There are many easy ways to do this. For my previous static sites, I’ve used a number of javascript options. For my personal site, I use feedjs for an up-to-date feed of this site. For course sites I’ve been using Google’s Feed API, which lets me embed a whole bunch of feeds into a singe frame. It’s also nice because Google caches feeds and lets you control a number of features. Sharon Leon is using a Google Reader bundle for The Digital Past, which is very nice given the ease of putting together and sharing bundles in Reader. I always actually track student posts for grading purposes that way.

But all of that is too easy. For my Flask site, I first put the feed retrieval directly in site logic with a function that checked feeds each time someone visited the student work page. By the way, I’m using a python package named feedparser for getting the feeds. This works. But, it’s also very slow and requires a whole bunch of resources to do retrieve things that are rarely updated. A better solution would be to store student feeds in an SQL, or even NoSQL database. But, I only want to show the most recent post on the site, so all the caching and checking, and then the overhead of adding an ORM to the app seemed like more trouble that it’s worth.

Solution? Why not write a script that gets the most recent post of a student site and have it directly write Markdown to a file in the pages directory? So, that’s what I did. I run the script with a cron job. Speaking of which, that’s another thing I learned doing this– how to write a crontab entry and deal with script path problems and the like.

feedparser returns a dictionary-like object that is fairly complex, so it took a bit of time figuring out where the elements of the feed were that I wanted to display. Then, I had encoding problems. Because Python 2.x always ends up in an encoding problem. feedparser returns unicode objects, and Python’s normal file writer tries to encode everything as ASCII.

The result is this page, which is essentially programmatically-written Markdown. All the code for the site is on github.

Next semester I’m teaching a grad/undergrad seminar on Digital History. That site will also be a Flask app, but this time I’m going to have students write posts directly on the site through an admin interface. Yes, I’m going to reinvent the blog, because doing it the hard way is better for me.

This evening, a retweet originating with Brian Croxall came across my feed highlighting an ad that should have come from the mlajobs tumblr. But it didn’t. It’s real:

Penn State U, DuBois – Full-Time Non-TT Assistant Prof. of English (DH, Lit, & Composition) – INTERVIEWS SCHEDULEDEdit

Full-time Nontenure-Track Assistant Professor of English

Teach three courses (9 credits) in Digital Humanities, Literature, and Composition a semester using traditional and hybrid delivery modes. Publish in refereed journals. Participate in professional organizations and in course, curriculum, and program development. Stay current in pedagogy appropriate for discipline through scholarly activities. Advise students and provide career guidance. Participate in campus, university, and community service activities. Position may be renewable. MLA interviews. Qualifications: Ph.D. in English, with specialty in Digital Humanities and expertise in British or American Literature. Evidence of potential in scholarship and professional development is expected. Commitment to high-quality instruction in a student-centered environment is desired. Interest in the instructional use of technology is required; interest in active and collaborative learning is an advantage. Prior college-level teaching is preferred. Enthusiasm for teaching in a multidisciplinary environment is important. Applicants should include a list of links to their significant digital humanities work, with appropriate access information. Finalists will be contacted and asked to provide a list of references.

How offensive. Penn State Dubois wants to hire an English PhD, with a specialty in Digital Humanities and British or American Lit, who has published in refereed journals, who’s up on their pedagogy, and who will have to advise students, do service, and is expected to be active in program development. That’s the description of a tenure-track professor. But, is this job for a tenure-track professor? No, of course not. It’s for a lecturer whose contract “may be renewed.”

The level of experience this ad describes doesn’t even fit the bill for many new PhDs who haven’t yet had a tenure-track job.

Now, Penn State Dubois offers Associate Degrees, which I guess means it is as much a community college as an extension campus of PSU. And it may be that community colleges are more used to and more likely to hire off track. But, the description here is the that of an R2 or quality SLAC. What does PSU Dubois expect? Do they simply think that DH people are all into that #alt-ac thing, and so don’t need the promise of traditional academic promotion in return for a commitment to research, publication, and teaching? Or is it simply that PSU Dubois knows there’s a large enough pool of surplus academic labor that they can get away with this? Or is this ad written for an inside candidate? If its the latter, then shame on you for putting other people through the MLA job wringer. If its any other reason, then shame on you for helping undermine the profession.

The conference interview is such a strange, and artificial part of the hiring process. It’s uncomfortable for everyone involved, on both sides of the table/couch/chair/cattle-pen. Whose idea was it originally to have awkward interviews in hotel rooms? I once had an interview where the school sent nine people to interview their candidates at the AHA. It was such a big crowd that we had to sit in a closed circle of chairs. This made it very hard to make eye contact. There are much, much worse stories out there than that. Suffice to say, we all need to figure out a few good strategies to make best on a less than ideal situation.

Advice posts circulate around every year, but I figure– why not one more? So, here are a few tips from an interviewer to an interviewee:

1. Relax. I know, that’s easier said than done. But, as a committee we are really interested and excited to talk to you. Think seriously about strategies to manage your nervousness. If coffee and caffeine make you nervous, don’t double down on it. If sleep really helps you, then do you best to make it an early night before. New Orleans will be a tempting place to take in a bar and restaurant scene. You’ve made the first hoop, and the committee wants to talk with you, learn about your work, and learn about you.

2. Be prepared to talk about your dissertation/book/work in a few sentences. You are guaranteed that you’ll be asked about this. If I’m interviewing you, I promise this question won’t be turned into something silly like, “What would you say to someone at a cocktail party if they asked what you do?” I hate those kinds of questions. But, some committees will feel like a gimmicky question gives them insight into you. So, be prepared for this, or any other question, to be asked in a gimmicky way. You should have a short and sweet answer to the dissertation question, and also one that is more extensive.

3. Be prepared to talk about teaching. Committees want to know that you’re interested and excited about teaching, and that you’re prepared to design and implement your own courses. We don’t just want to know that you can teach courses already listed on a course description page or in a curriculum. If an ad mentioned some specific type of course, be prepared to answer questions about how you will approach that class. But, don’t go do to much research on course catalogues– you have no idea how long courses have been in the catalogue, if they aren’t taught, if they’re the bailiwick of a particular faculty member, etc. Be able to speak to both general and specific approaches. (BTW, everyone who teaches history likes to use primary sources. Saying you do that isn’t enough.) If the program has both graduate and undergraduate offerings, be prepared to talk about both. Have a seminar in mind that you could teach next semester.

4. For schools with high research expectations, be able to talk about what needs to be done to your dissertation/manuscript for it to be publishable, and what presses you think would be a good fit (even if you haven’t contacted them yet). We’re in a transition period, and for historians the single-authored-monograph is still the gold standard for promotion. The AHA interview is not the place to litigate its efficacy. Many times readers of this blog are more technologically sophisticated and forward looking that those who they’re interviewing with. Just remember that initial interviews aren’t the place to grind axes, and at this stage in the game you have to speak in a both/and frame, rather than a either/or one.

5. Have a couple of questions to ask the committee. I think something like, “Where do you see your department in five or ten years” is a good one, because it gets the committee talking about themselves. Be careful not to craft a question that might be interpreted wrongly by people who could feel defensive about their school, student body, or geographic location.

6. Don’t bring extra materials to the interview. If committees want to see more from you, they will ask for it.

7. Scope out the layout of the hotels before the morning of your interview. Know how long it takes to walk from one to the other. Know where the right set of elevators are ahead of time so you don’t have to stress about it five minutes before your interview starts.

8. If the subject of service comes up, connect it to something you’re particularly passionate about– undergraduate teaching? graduate teaching? outreach?

9. I’ll say it again– try to relax. I’ve listed a bunch of stuff above, but remember first and foremost that the hiring process is about finding colleagues. People want to hire people they want to work with. And that goes both ways. It’s more the case in an on-campus interview, but you’re interviewing us as well.

Any other recommdations? Please share them in the comments below.

I’ve been doing most of my writing, coding, notetaking, and other work in Vim for a few months now. There has been a slow accretion of wisdom and workflow, and muscle memory. (I frequently find myself reaching for Vim keyboard shortcuts when using other text editors or word processors, and find myself frustrated when they don’t work.)

I’m going to start posting an irregular series of posts on using Vim for the kind of long-form writing historians and other humanists do. In part, I’m doing this because I learn a tool better when I write about it. (That was certainly the case with Devonthink.) In part it’s because Vim gets better with customization, and when ever I move to a new machine I forget half the things I’ve done, or how I did them. So, it’s also an exercise in personal archiving.

My favorite tool for writing has long been Scrivener. What I most like about Scrivener is the ability to split the screen and view source material as I write, and export to a variety of file formats of a finished draft from its constituent parts. I also like its file navigation and little bits like distraction-free writing mode. I’m less interested in its summarization and the many complex forms of metadata one can add to files. Another thing I don’t like about Scrivener is that the package contains full copies of all the files in one’s research folder, as opposed to symlinks. This means for a large research project you can end up with a very large file, and one that is duplicative. With a few plugins, Vim can easily handle many of the things that I like about Scrivener (if not most). There are a few bumps in the road, though.

Most of my research files, notes, etc. are text files. But, of course, as an academic, I also have tons of pdfs and legacy files in .doc or .docx. These files are, obviously, not accessible by vim. I’m working on a couple of lugins that will make them accessible by converting their contents to text or markdown automatically with filetype detection. (As it stands now, using NERDTree, it’s easy to open these files in their external system editors (Word, Preview, etc.). At any rate, what do I want from a Vim Scrivener set up?

• multilingual spell checking
• file navigation
• split screen editing
• PDF text layer viewing
• web page downloading/markdown conversion
• export to pdf, docx, or rtf
• sensible line wrapping
• directory search

Much of this is already available from Vim and a few good plugins. So, I’m going to start an irregular series of posts for the Vim Humanist on how to customize Vim for our writing and research needs. If you have ideas or tricks, let me know and I’ll include them in the series.

If you read the comments on the gawker and jezebel pieces, you’ll find a bunch of commentators excoriating gawker, Chen, etc. for violating the privacy and anonymity of redditors, an act so vile as to bring them to the defense of people like Brutsch and subreddits that celebrate the anti-consensual consumption of women and girls as objects.

In a coincidence of timing, this story was breaking in the days surrounding the UN’s new International Day of the Girl on Tuesday. Erwin C. wrote a post for the Day that provides just a few snippets of data on how deep gender inequality continues to be in Latin America. (h/t to Colin) From the post:

• In 2009, 15.5% of Mexican girls between the age of 15 and 19 had at least one child partly as a result of 61.2% of teens who did not use birth control.
• 69% of the 7551 reported cases of domestic or sexual violence in Peru during the first eight months of this year involved girls under the age of 17.
• A survey taken two years ago showed that 43% of women in the Dominican Republic aged between 20 and 49 had married before reaching the age of 18.
• 70% of Bolivian girls do not got to school while 30% of girls residing in rural areas finish elementary school.
• 248 girls under the age of 14 were killed last year in Colombia,which is up from 176 in 2010.
• 14% of Ecuadorian indigenous girls between the ages of 5 and 17 do not attend school.
• Roland Angerer of the Plan International NGO told EFE that for Central American girls the “quality of education is worse” than other regions and the levels of malnutrition are “worrying”.

I’m struck at the cognitive dissonance of these two bits of news. While redditors debate the value of their anonymity to act out misogynistic fantasies, real girls around the world struggle daily for access to school, with sexual and domestic violence, for control over their fertility, for economic opportunity.

I find this enraging. I think there are good cases for internet anonymity– for whistleblowing, for political dissent in countries with real consequences for such dissent, etc. This is simply not the case here. In the bubble of anonymous internet communities, though, the real and daily struggles for life and justice for women and girls aren’t compelling. Trolling and violation are. These people deserve to have their identities outed. Their sense of outrage is rooted in the same gender privilege that justifies their daily violation of women’s consent on those subreddits. It’s abhorrent, and should be called such.

There is a page on the homebrew wiki that explains how brew handles installing Python and its package managers easy_install and pip. Following the instructions, including prefixing your PATH with /usr/local/bin and /usr/local/share/python, is supposed to leave you with both easy_install and pip installing packages for Python2.7, without the need to use sudo. It didn’t work for me. So, here’s the simple fix I had to make to get pip (and easy_install) working correctly:

If you type which pip at a bash prompt, and followed the instructions correctly, you’ll find it (and easy_install) in /usr/local/bin. If you open that file in your favorite text editor, though, you’ll find a shebang line at the top:

#! /usr/bin/python

that points pip to the system python. Edit that line to:

#! /usr/bin/env python

and pip will then correctly install packages for homebrew’s Python instead of system Python. I banged my head against the wall for a while trying to get pip to install packages correctly, until I thought to look at the pip file itself. Maybe now you won’t have to do that.

This is why the MLA, the AHA, and other professional organizations need explicit language on acceptable criteria for professorships, especially at the entry level.

Yesterday (Sept. 11) I attended a talk at UTK by current MLA President Michael Bérubé. A member of the audience asked him about the CSU job ad, and he responded that he knew about it, thought it was horrible, put it on the next MLA Executive Committee Agenda, and that the MLA was going to un-approve the ad.

Today, it appears that CSU has changed the ad, and removed the offensive language. I’m glad that they were shamed into doing it. I doubt, though, that changing the ad will affect what goes on in search committee deliberations.

Unfortunately, what CSU put into writing (inadvertently or not) reveals a set of assumptions that operate often, and sometimes explicitly on search committees. The further one is from degree the more that date penalizes them (unless they already have a TT job). It’s so much like the recent practice of companies not interviewing the unemployed for jobs. And it’s pernicious, especially in an age of waning opportunity.

I frequently tell our graduate students that excelling in grad school qualifies one for a job market lottery ticket. But, especially given that one cannot forecast 6-8 years out what the economy will be doing, and what the hiring situation will be like, it’s not much more than that. And that is if one is excelling. And, on top of all of that, we already cede the reality that new PhDs will have no control over the geography of the professional career or its impact on family.

I should have named the original post “Stale PhDs Need Not Apply”, because regardless of the implications of the ad for protected categories, what the ad did was, in the words of a friend, “invent a new epistemology of exclusion.” And as Robert Townsend documents in Figure 1 of his aptly named post The Ecology of the History Job, it’s quite obvious who the target of that exclusion is.

Required qualifications:
1. Ph.D. in English or American Studies or closely related area awarded between 2010 and time of appointment.
2. A promising record of scholarship/research in pre-1900 American literature and culture.
3. Ability to teach a range of subjects in American literature and culture between 1600 and 1900.

It’s no surprise, of course, that CSU wants a colleague who shows promise as a scholar and who can teach widely in their area of expertise. But, I (and others) find it quite astonishing that the department expressly limits their pool to those with a PhD “awarded between 2010 and time of appointment.” I find this so ethically shady, especially given the recent history of the humanities job market, that I think it must be called out. Why would CSU write the ad this way? Trevor Owens suggested that it’s so strange, they must be writing for an inside candidate. I thought maybe they were just trying to keep the applicant pool small. Rather than continue to speculate, I wrote to the Chair of the Department and of the Search Committee and asked them why that language was included. I haven’t heard back yet, but I’ll update this post if I do. (I’m not hopeful though, as academic job search committees frequently can’t even be bothered to communicate with their own candidates, much less some nosey professor who’s not even in their discipline.)

A little more on why I find that language offensive. The effect of the ad is to say to PhDs who, in the wake of the job market crash of 2008, have held out hope for full employment that they should simply give up. It’s been both conventional wisdom and empirically backed up for a while that a new humanities PhD has three years to land a tenure track job, a time frame outside of which it gets inordinately more difficult to get a job. And, while we all may know that to be true, to see a job ad that codifies that as a hard and fast rule is very depressing. The CSU English Department is saying that if you were unlucky enough to enter the job market in 2008, 2009, and quite possibly 2010, you need not apply because apparently you’re not current enough. Those years adjuncting or lecturing while working to get a publication or two? Not good enough. Or, if you did get a tenure track job, but if you have ambitions to live in a different part of the country or advance your career in other ways, too bad for you. On its face, this ad expresses a total callousness to the state of humanities work in 2012. And, given all the rest of the capriciousness of the academic job market, anything that makes it more callous should be called out as unacceptable.

At UT, we have to have our job ads and the whole search process cleared by our Office of Equity and Diversity. The language here doesn’t, as far as I can tell, violate discrimination policies for protected categories of race, ethnicity, age, gender, or religion. But from an ethical standpoint, I find it just as bad. Moreover, as my department’s Director of Graduate Studies and advocate for our graduate students, I would be chagrined if this approach were to catch on widely in the humanities job market.

Update (3:30pm): 
I heard from the people at CSU. I’m waiting for permission to post that email communication in full. Suffice to say, the search committee feels the language denotes the position is entry level and for people with no more than 3 years experience on the tenure track or at the true beginning of their career. Again, I find the language to be astonishingly dismissive of the reality of the humanities job market. As Eduard Gans states below, there are any number of reasons (besides already being in a tenure track job for 3 years) why someone might be 3 years or more out from their degree, completely qualified (eh hem), and looking for work. What is more, this language passed CSU’s Office of Equal Opportunity. It may be fine from a solely legal perspective, but I see this as an institutional failure by CSU.

Update 2:
Anand Sarwate contacted CSU’s Office of Equal Opportunity directly. Apparently because Fort Collins is a popular place to live, and the Department doesn’t want/can’t pay for an experienced professor, OEO cleared the ad. So, OEO is as tone deaf as the department?

Update 3:
Here is the response I got from the chair of the Search Committee:

Dear Prof. Black:

Thank you for your questions concerning our job description. Our reasons for requesting applicants who received their Ph.D. in English or American Studies between 2010 and time of appointment have to do with clarity and fairness regarding our expectations. We seek to recruit the greatest number of qualified applicants for our particular position. Because the position is “entry-level” with an entry-level salary and expectation, we believed it necessary to define that term in some way. By specifying “between 2010 and time of appointment” we indicated that we are interested in applicants with up to three years in a tenure-track position as well as those who are just beginning their careers. In examining the pool of applicants, we have actually given the true “entry-level” applicant an advantage in that such applicants will not have to compete with others who have as much as six years more experience. Our language has been approved by the Office of Equal Opportunity’s federal standards for fairness and clarity. I hope this response clarifies our choice of phrasing in our job description. We hope you will distribute this position description to any qualified applicants you know.

Sincerely,

Prof. Paul Trembath

Update 4:
One last update in a new post here.

The flow for an edit function needs to open vim with a named temporary file in the buffer, allow the user to write what needs to be written, save and close vim, and then capture that input. In my case, I want the capture to return a dictionary of key:value pairs to use with my task script. The easiest way I can think to do this is to use yaml as the file format. Thus, this script needs pyyaml installed.

Following good python practice, import the needed modules with standard and third party modules on separate lines:

1
2
3
4

#! /usr/bin/env python

import tempfile, subprocess
import yaml

tempfile is part of Python’s standard library, and provides simple support for spinning up temporary files that are automatically flushed by the script. That means, of course, that your task won’t be stored permanently on disk as a file. It also means the file needs to be read before it’s flushed. We do this with a while block:

while tempfile.NamedTemporaryFile(suffix='task') as temp:
    subprocess.call(['vim', temp.name])
    text = open(temp.name, 'r').read()

This code creates a temporary file object with the suffix ‘.task’. temp.name passes the name of the temp file to vim, which is called by subprocess.call(). Note that call expects a list as input, which includes the command and any arguments passed to it. At this point, vim will open and we can write whatever we want as per usual, saving and exiting vim as per usual. Upon exiting vim, and before the while block ends, we assign the contents of the temp file to text.

In my case, text is a series of key:value pairs in yaml form. To get a dictionary, we simply load text as yaml:

params = yaml.load(text)

Now we have a Python dictionary of key:value pairs that can be used by other functions in the script.

I want these yaml files to be consistent, so I’m using a vim plugin that will serve templates based on filetype. After installing the plugin, I put a file named taskedit in ~/.vim/templates like this:

title:    
context:    
tags:
note:

# cursor: 1 12 del

So, when the script opens a new vim buffer, it’ll be pre-populated with this template, with the cursor waiting after title for input. Nifty. Those keys will be the keys of my parameter dictionary as well.

Today, I was playing around with argparse again, trying to cleanup and simplify some code and stumbled across what seems to me to be a better way.

The reason I used all those if statements in the past is because the parser returns a Namespace which isn’t iterable:

import argparse

p = argparse.ArgumentParser(description="parse some things.")

p.add_argument("cmd", help=argparse.SUPPRESS, nargs="*")
p.add_argument("-d","--date")
p.add_argument("-p","--project")
p.add_argument("-c","--context")
p.add_argument("-l","--list")

opts = p.parse_args()

print opts

Run this code with input like this:

add "A task to be done." -d=today -p=myproject -l=mylist

(obviously, I’m dealing with task management in the real script) and the parser will return a Namespace object:

Namespace(cmd=['add', 'A task to be done.'], context=None, date='today', list='mylist', project='myproject')

Type Namespace isn’t iterable, and I want a convenient way to get rid of arguments that returned None. Finally stumbled across the answer today– vars. We can extract from the Namespace object to a dictionary by passing the object to vars:

opts = vars(p.parse_args())

Why bother? Well, now we have a dictionary that can more elegantly be dealt with, and we don’t have to hard code all possible keyword arguments for the script. opts is now:

{'date': 'today', 'project': 'myproject', 'cmd': ['add', 'A task to be done.'], 
'list': 'mylist', 'context': None}

Now, in two quick steps we can pop the positional argument (cmd), and cull the unused optional arguments:

command = opts.pop('cmd')
options = { k : opts[k] for k in opts if opts[k] != None }

That gives us a list of the positional argument and its associated text, and a dictionary of the utilized optional arguments. Beginning with Python 2.7, dictionaries get comprehensions much as lists had for a long time. The second line of that couplet uses a dictionary comprehension to get the key:value pairs from opts that aren’t empty. This is nice, because we can call the correct function now using the list command, along with passing it arguments from both command and options.

command[0] gives us a string that names the function we need. So, how can we then call that function (without using eval/exec)? It’s easy with a dictionary that points to the functions we’ve defined. By way of example:

def add(task, *args):
    print task
    for arg in args:
        print arg

functions = { 'add':add }

The add function here obviously doesn’t really add a task. Just for show. Nonetheless, we can call it with our command/options variables thusly:

functions[command[0]](command[1],options)

I like this, because it makes parsing a wide range of potential positional arguments without having to add each one of them to the ArgumentParser. That does, however, cause problems for help output. We won’t have a help script for every possible command this way. My solution for this is to move the help cues into the program description, and exclude the “cmd” argument from the help output:

p = argparse.ArgumentParser(formatter_class=argparse.RawDescriptionHelpFormatter,description='''
    Parse tasks. Options include:
        add: add a task, followed by the task
        list: list the tasks. Filter with list, project, or context options
        del: delete task by number
        clear: clear tasks
''')

p.add_argument("cmd", help=argparse.SUPPRESS, nargs="*")

That’s it for now. And really, most of this is just so I’ll remember it.

• Introduction
• Installing Python
  • Linux
  • OS X
  • Windows
• 1. Interactive Mode
• Integrated Development Environments
  • IDLE
  • Other IDEs
• Text Editors
  • Linux
  • OS X
  • Windows
• Conclusion

Introduction

In my spare time this summer, I’ve been working through CS101: Building a Search Engine on udacity. I’ve written critical things about the new wave of massive online courses before, in part because I’ve never finished one. That’s not exactly a fair position from which to criticize, even if it’s a reality shared by the vast majority of people who sign up for these things. And, given this week’s announcement from Coursera that a dozen universities have signed on to offer courses through their platform, a little more critical participation may be in order to see the order of things to come. Or at least the order for the 5% who finish these classes.

So, I’ve been working through CS101, answering a few questions in the forum, and getting a feel for the community that exists around the class. One question that has come up repeatedly in the forums is how to access the Python Interpreter to work on solutions for problems before entering them into the course’s interactive programming environment. (BTW, that environment is pretty nifty, and appears to run on a virtual machine that is spun up and persist per session.) There seems to be a fair amount of confusion at times one how to access Python on student’s home machine. Surprising, the python reference included on the course site never covers this. And, the instructor (Dave Evans from U.Va.) frequently references trying things out in the interpreter. I find it a bit curious that the course materials essentially assume students will do all their work on the course site, which means that code snippets don’t get saved and built upon on the student’s own machine. (It’s very easy to take the class in this way, but I think weakens it.) Of course, CS101 isn’t a programming class, but an introduction to computer science principles. That means that most students likely have no programming background, which makes the absence of a tutorial on setting up a local programming environment all-the-more unusual.

In the interest of giving back to a free course, sharing both with those taking the course, and with other first-time programmers, I figure I might as well provide a tutorial of a few options myself. I’m not a professional programmer, but picked up these bits and pieces along the way when I first started to learn some Python.

For CS101 (and similar) students, you have three main options for interacting with the Python Interpreter:

1. In interactive mode from the command line.
2. With an Integrated Development Environment (IDE).
3. With a text editor.

But first…

Installing Python

A side note on which version of Python to use. The programming environment for the class uses Python 2.6.6. You can check it for yourself the next time you’re answering a question on the site. Just add this code and hit run:

import os
print os.system("python --version")

Given that the course is taught using a 2.x version, it’s highly suggested to install that on your local machine as well. Python 3.x makes a number of backwards incompatible changes to the language. This in one instance where the “latest” isn’t actually preferable.

Some operating systems have python already installed. Some don’t. Before one can work with/in python on their local machine, we have to make sure it’s there!

Easiest to hardest:

Linux

Python comes pre-installed on all the different linux distributions. If you use linux, I’m guessing you already know that! Nonetheless, to prove it to yourself. Simply open your Terminal application and enter python at the prompt:

$ python

Assuming you’re distribution isn’t too old, your version of python is likely fine for the CS101 course. If not, use your distro’s package manager to upgrade python.

OS X

Guess what! Python comes pre-installed on Mac OS X too! Open up Terminal.app and do the same as described above.

$ python

You’ll get the same. Depending on which version of OS X you’re using, you’ll likely have either 2.5.1 (Leopard), 2.6.1 (Snow Leopard) or 2.6.7 (Lion). For CS101, any three of those is fine.

If you’d like to use a different version of python DO NOT trash the default installation. That will cause real problems. Instead, follow the instructions here to install python with Homebrew.

Windows

Microsoft does not deliver its Windows product with python pre-installed. The easiest option for a Windows user is to take advantage of python.org’s binary installer. Simply go to the download page and chose the appropriate installer. This installer will provide access to python, including the at the command prompt, from your start menu. The installer will create a C:\Python27 directory. You can also simply open a command prompt and type:

c:\>; python

That will put you in the interpreter.

So, now to use the interpreter…

1. Interactive Mode

Python can be used in two different ways. Normally, one would write a script in an IDE or a text editor, an then have the interpreter run the script. In interactive mode, the interpreter evaluates each line of code as you type it in, giving immediate feedback. This is done on the command line (or prompt) as you’ve seen above.

When started, interactive mode will look something like this:

Python 2.6.7 (r267:88850, Jul 31 2011, 19:30:54) 
[GCC 4.2.1 (Based on Apple Inc. build 5658) (LLVM build 2335.15.00)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>;>;>;

The prompt will take your input, immediately evaluate it, and return the result (if any) to the terminal:

>;>;>; print "I'm enrolled in CS101."
I'm enrolled in CS101

If you make a mistake, you’ll get back immediate feedback as well:

>;>;>; print 3+' Three'
Traceback (most recent call last):
  File "<;stdin>;", line 1, in <;module>;
TypeError: unsupported operand type(s) for +: 'int' and 'str'

Here, the error is trying to concatenate an integer and a string.

There are advantages and disadvantages to using interactive mode.

Advantages:

• Get immediate feedback as you type a script in, including errors.
• No need to write files to disk, and then run them. From the terminal this means not having to call the specific file like this:

  $ python path/to/your/script.py

• Quick and easy to try something simple.

Disadvantages:

• No syntax highlighting. You don’t get color to mark parts of your code.
• If you make a mistake with, for example, indentation while writing a function, you’ll have to start over.
• There’s no persistence. Once the code has run, you’ll have to type it in all over again to retry or tweak it.
• It’s not suitable for longer scripts.

For CS101, interactive mode can be a useful way to try out solutions to simple bits of code, especially in answering Quiz questions. But, for the problem sets, the limitations noted become much more important. So, how to get persistence?

2. Integrated Development Environments

An Integrated Development Environment (IDE) is a software program that, well, integrates all the tools you would need for software development. IDEs range from very simple to highly bloated with support for many different languages, code completion, debugging, source control, and more. Most IDEs are probably overkill for CS101 students.

IDLE

For an experience most like the examples demonstrated by Dave in the course lectures, and the option to work in either interactive or normal mode, give a IDLE a try. IDLE is a very simple IDE bundled with python distributions. It offers both an interactive shell, and the ability to write scripts as files and run them from within IDLE. Moreover, in both interactive mode and normal mode, IDLE gives syntax highlighting as seen in CS101. This is very helpful, especially for making sure you close your strings in quotes! Also, in interactive mode IDLE will highlight the spot where syntax errors occur.

Errors show up in red, like a professor marking something wrong.

To write to a file, simply open a new window and start typing. To run that file, on a Mac at least, simply press F5. Output of the script will show up in IDLE’s interactive shell. Files can be saved and opened in the way you’re used to with other programs on your system. For CS101, this means you can maintain a file with all the functions that form the search engine. Or, you can maintain separate files for individual Problem Set assignments.

To open IDLE on a Mac, type idle in the terminal:

$ idle

On Windows, you can chose IDLE from the start menu, as part of the python bundle.

Many distros of Linux don’t include IDLE, but it can be installed by your package manager as either python-tools (Fedora) or idle (Ubuntu).

Other IDEs

I would argue that full-scale IDEs are overkill for CS101, especially as they usually involve build tools for languages that need to be compiled before execution (which Python does for you). Some cross-platform IDEs include Eclipse (targeted mostly at Java and C/C++ development), Apple’s Xcode (targeted mostly at Objective-C development for OSX and iOS), and NetBeans (also rooted in the Java world). Here’s a much more complete list by language.

3. Text Editors

In my opinion, text editors offer the best option for those new to programming, and especially to people like me who come from an academic humanities background. Some text editors have very high learning curves, like Vim or Emacs. Others are simple to use and quickly powerful. Text Editors tend to be very personal decisions as well, resulting in the kind of fanatical devotion that personal decisions elicit.

The workflow on a powerful text editor is still quite simple. Open a new file, enter your code, and run it from within the editor along the way, either sent to the terminal or within the editor. This may sound a lot like IDLE, but a good text editor is much more powerful. You get code folding, tab completion, line numbering, automatic indentation, some level of debugging, syntax highlighting, and some level of code templating. And not only for python, but for a number of languages. All that, without the complexity or bloat of a full-scale IDE.

I use Textmate, a program I’ve come back to many times after trying to leave it. (This post is written in Textmate with markdown.) When I start a new Unit in the class, I open two windows in Textmate — one for notetaking, and one for code. While following along with the lectures, I take notes in the one file, and start building the code snippets in the other. To try a piece of code written in Textmate, I have two options–>; CMD-R runs the code and sends the output to a Textmate pop-up window while SHIFT-CMD-R will open a Terminal window, and run the script there. There are bundles for all kinds of programming and markup languages for Textmate, each of which makes writing in the editor easier and more efficient. These instructions are particular to Textmate, and each editor will have it’s own quirks.

There are many text editor options, both free and for purchase, available for all three platforms. Some to consider:

Linux

• Redcar -free
• SublimeText2 — a very good, cross-platform editor
• Geany — free

OS X

• Textmate
• TextWrangler — free in the App Store
• SublimeText2
• Komodo Edit — free, and the little brother/sister of the Komodo IDE

Windows

• NotePad++ – free
• SublimeText2
• E-TextEditor — aiming for the Textmate experience on Windows
• SCiTe — free

These are only a few of many options.

Conclusion

Once you have Python installed, and have experimented with the Interactive Terminal, IDLE, and maybe a Text Editor, you should have a sandbox to play in on your own machine while working through the course. I hope that you’ll find making progress through CS101 a bit easier.

One final note, I’m frequently surprised at the number of questions asked in the forum whose solutions are very simple to find with a google or duck-duck-go or bing search. It should be taken as a truism that if you are interested in learning in the MOOC setting of a udacity or coursera course, you must be willing to hone your google foo. For MOOCs to be successful at all, the participants must approach them with a hefty sense of D.I.Y., though hopefully mixed together with a sense of community — a community of D.I.Y.’ers

Doctor Don Ygnacio Martinez del Escobar was awakened by the heat of the midday sun. He opened his eyes to find himself lying in the quebraba just south of the Monastery of the Nuns of Carmen de la Antigua Fundación. Don Ygnacio sat up, and took stock of himself. He noticed immediately that his hat was missing, his cloak was soiled, and a combination of vomit, spittle, and escabeche stained the cuffs of his linen shirt in a manner that did not befit a man of his station. But, these were simply distractions from the doctor’s chief concern— the searing pain that cut back from his deep-set eyes into the recesses of his head, and radiated down, exiting in the back at the top of his spine, and in the front issuing from his cheekbones, as if the deeply-tanned cracks and wrinkles of his face that marked the fifty years Nuestro Señor had granted him on this earth were channels, rivulets, ravines for his suffering. His jaw ached. The pain came in pulses as Don Ygnacio fought to get his bearings; it was encouraged by the glare of the noon sun, which was surprisingly warm and bright despite the depth of the ravine. The air was thin in Quito, and offered little resistance to the intensity of the sun. The doctor ached for the shadows, but each movement set anew the pain flowing through his head. His only relief came in the form of the swirling breezes that released the cool air that had settled in the Andean night in the nooks and crannies of the deeply cut gully.

Luckily for Don Ygnacio it was August, which meant that the wind was constantly blowing and the relief came in continual waves. August was a dangerous month, when the saints and the Virgen de la Merced and the Lord Jesus fought for control of the atmosphere with the huacas and their mountains and lakes and rivers. The instability was disconcerting, and often manifested in the narrow, steep streets of the city when the barrios of Quito felt an irresistible urge to shove off authority. Normally, the natural order of the world prevailed, shorn up by the whitewashed stone walls of the Sagrario’s impressive block houses, churches, and jails. There were times, though, when the balance between Spanish institutions and Pachamama’s temples of lava, steam, and stone was uncertain for a fraction too long. In those moments either the barrios or the earth would explode in violent convulsions. August was dangerous, because the winds testified constantly to cracks in the veneer of normalcy that defined daily life for the forty-thousand inhabitants of a city cloistered by towering volcanoes and cavernous gorges. When the winds finally abated, the rains would come and travel would go from difficult to nearly impossible. News that took weeks or months to travel from Guayaquil or Cuenca, more-less Lima or Bogota, would take many more months to arrive, if the mail made it at all. The isolation gave a sense of timelessness, of geologic immobility that was a welcome relief from rapidity of August’s winds.

Such isolation was useful for a man like Don Ygnacio, and news of his past indeed was slow in catching up with him. At the moment, though, any fear of his past was subservient to the craving need to defeat the pain. As a doctor, Don Ygnacio knew the medicine he needed. Or at least, if not as a doctor, then certainly as a boracho he knew. He knew the pain would never abate until he found his way to a bottle of aguardiente. And he was right.

Red Emma’s, as one might expect, has these books in spades. I can’t tell you how refreshing it was to walk into a bookstore and find Laclau and Mouffe on the shelves, or a whole section of Althusser and Kropotkin and Marcuse and Etienne Balibar. I had discovered those authors and many more local bookstores in Raleigh, NC in the 1990s, but also on the shelves of the majors back then. I have fond memories of lounging throught the stacks at Borders in Raleigh and finding books worth reading.

Daedalus is a whole other wonder– overstock copies of smart books sold at used-book prices. Again, I can’t remember the last time I went to a bookstore in the other cities I’ve been living in recently (Albuquerque, Knoxville, frequent trips to Quito) where I wasn’t bored almost instantly. Retail bookstores have been so devasted by the Amazon phenomenon that they’ve become completely predictable, and little better than the mall bookstores of my childhood.

So, what I bought today:
MacPherson, Myra. All Governments Lie! The Life and Times of Rebel Journalist I.F. Stone
Laclau, Ernesto. Politics and Ideology in Marxist Theory: Capitalism – Fascism – Populism
Hess, Charlotte and Elinor Ostrom, eds. Understanding Knowledge as a Commons
Bohning, Don. The Castro Obsession: U.S. Covert Operations against Cuba, 1959-1965
Burnett, D. Graham. Trying Leviathan: The Nineteenth-Century New York Court Case That Put the Whale on Trial and Challenged the Order of Nature.
Hardt, Michael and Antonio Negri. Multitude: War and Democracy in the Age of Empire.
Illich, Ivan. Deschooling Society.
Anderson, Jon Lee. Che.

There’s a whole lot of fun reading for me in that list.

I cracked open MacPherson’s biography of I.F. Stone first. The epigraph is just so perfect for the events of the last four years (or, indeed, of the last few weeks at UVa!):

And, it turns out, neither did droves of other people.

One of the classes I was most excited to follow was Natural Language Processing, offered by a pair of Stanford professors. The course just recently concluded after its 8 week run. Interestingly, in a final email, the professors included statistics about course participation:

• Preregistered: ~70,000
• Registered: 42,223
• Watched at least 1 video: 24,287 This group included me!
• Watched all videos: 4,030
• Got a statement of achievement: 1,466

• Total video downloads: 860,854

There were eight programming assignments. The stats on submission of those assignments:

• PS1 8,425
• PS2 4,527
• PS3 3,721
• PS4 2,994
• PS5 2,529
• PS6 2,005
• PS7 1,911
• PS8 1,654

These numbers represent quite astounding levels of attrition, and bring to the fore, I think, a number of questions for the efficacy of this particular form of MOOC. First, I wonder what the difference in participation rates is between a course like this on coursera or udacity, and Jim Groom’s Digital Story Telling? Granted, the two courses offer very different experiences and to very different ends. But, those differences point directly to tensions in the move to offer open, online courses. On the one hand, Groom and his merry band of pranksters are engaging in acts of creativity that have next to nothing to do with credentialing. The CS and other courses offered in this Spring’s explosion of MOOCs feel, to me, much more pragmatic, much more credential-oriented. It’s like the difference between genuine DIY/edupunk learning and Kamenentz’s retail version, analyzed so well in Stephen Downes’ review of Kamenentz’s The Edupunk’s Guide.

Certainly, many people in the cast of thousands who dropped out of the NLP class did so because the content was hard. Hard in a technical way that’s different from the ‘hard’ of a course on digital storytelling. And, there is no denying that for the 1400 people who finished the course successfully, the NLP MOOC was a success. Heck, I know more about NLP (and command line utilities, for that matter) just from the handful of videos I watched. I’m sure that many others learned some subset of the course’s objectives as well. But, the attrition statistics point to some real problems with the online model still proffered by ventures like udacity, coursera, and the other elite university ventures. I know that if I had attrition rates of 97% in my traditional face-to-face courses, I would have intervention from my department, from my college. Is it fair to compare the two settings? I think it is, only to the extent that advocates for online education trumpet the internets as a source for revolutionary change in the higher ed model. Or, it could just be that the NLP class is an outlier, and the material is hard enough that only 3% of those who were motivated enough to enroll themselves in the class were capable of getting it. I doubt that’s the case.

More likely, the current model of user-generated forums, videos with embedded multiple-choice questions, robo-graded assignments, and credential-oriented outcomes really don’t generate the kind of engagement needed to keep the 97% involved for just 8 weeks. Is this a glass-half-full/glass-half-empty type of question? I wonder. Groom and ds106 raised $12,643 from kickstarter in a few weeks, in part because they lacked institutional support of a Harvard, Stanford, Michigan, MIT, or Penn or the ‘startup’ feel and Silicon Valley ties of udacity’s Sebastian Thrun. And yet, as an outside observer of this process (and small time kickstarter supporter of ds106), the educator in my finds one of these two models much more exciting than the other.

Did any of you finish one of the coursera or udacity courses?

The people at This American Life were endpoint collaborators with some amazing people, who worked on the story with dedicated prosecutors in Guatemala. ProPublica has their report on Dos Erres here. You can also read declassified US documents on the Massacre at the National Security Archive’s blog, Redacted.

I would very much encourage you to listen to the story. But, I have a word of caution, or rather frustration with This American Life being the venue for its telling. This story is very heavy, and the seriousness and somberness of its reality is hampered by Ira Glass’s tone of ironic detachment. While that may work well for the kind of frequently poignant and entertaining, if self-absorbed stories that generally pass for This Life fare, it is completely inappropriate to the gravity of this story. I was shocked and dismayed that Glass finished the piece with his formulaic joke at producer Torey Malatia’s expense using a quote from one of the Massacre’s participants (a cook who failed with the Kabiles because he couldn’t physically handle it) and relating it to NPR pledge drives. That kind of crassness is inexplicable in its inappropriateness, except maybe to say that the producers of This Life, and Glass himself just doesn’t get it.

I can remember a recent episode, the retraction of the Apple/Foxconn story, when Glass just didn’t feel like the formula joke was fitting and left off without it. That was a story of This Life’s failure. I guess if it’s not about them, then they don’t need to take context and appropriateness into consideration.