Jekyll2020-08-24T10:40:28+00:00aelang.net/feed.xml| digital textualities | semantic cartographies | networked geographies |adventures at the interface of literature & dataAdventures in feminist collaboration: The Data-Sitters Club2020-04-16T12:50:41+00:002020-04-16T12:50:41+00:00aelang.net/blog/2020/04/16/dsc<p><a href="https://datasittersclub.github.io/site/"><img style="float:right;border:10px solid white" src="/DSClogo.png" width="50%" height="50%" /></a></p>
<p>This project started life in late 2019 as a bit of fun by a handful of feminist scholars - Quinn Dombrowski, Katia Bowers, Roopika Risam, Maria Cecire, me, and a little later Lee Skallerup Bessette - whose overlapping backgrounds in library science, modern languages, English literature, digital humanities and children’s literature turned out to be surprisingly handy when it came to turning a simultaneously critical and nostalgic lens onto a young adult fiction series we had all read and felt ourselves to be shaped by as girls and teenagers in the 1980s and 1990s: Ann M. Martin’s <em>The Baby-Sitters Club</em>. Quinn, our inexhaustible and endlessly creative Kristy, describes how the project got started on <a href="https://datasittersclub.github.io/site/chapter-2/">the chapter 2 section</a> of our website.</p>
<p>Since then, the Data-Sitters Club has gathered considerable momentum, winning places in two categories in the <a href="http://dhawards.org/dhawards2019/results/">2019 DH Awards</a>, being featured in <a href="https://insidehighered.com/news/2020/04/15/we-need-fun-more-ever-digital-humanities-meets-baby-sitters-club-books">Inside Higher Ed</a> (and subsequently picked up by <a href="https://bookriot.com/2020/04/16/introducing-the-data-sitters-club-critical-linking-april-16-2020/">BookRiot</a>), and getting a lot of love on Twitter for <a href="https://datasittersclub.github.io/site/covid19/">Quinn’s gallery of covid-19 public service announcements which pastiche BSC book covers</a>. At this point (April 2020), we have seven ‘books’ in total: four in the regular series - of which <a href="https://datasittersclub.github.io/site/dsc4/">#4, <em>AntConc Saves the Day</em></a>, is my contribution - and three in the Multilingual Mysteries series which investigates translations of the books. It’s a project I’m very happy to be part of: it’s an example of feminist pedagogy in action, it embodies the digital humanities ethos of openness and accessibility, it’s a reminder of the importance of taking popular culture seriously, and we see it bringing joy on a regular basis to those who discover it on Twitter, at a time when joy is in short supply. I am drafting my next DSC book, on some compellingly weird BSC-esque prose generated by a neural network, so that should be out over the next little while: stay tuned to <a href="https://twitter.com/search?q=%23DataSittersClub&src=typed_query">#DataSittersClub</a> for more.</p>Machine learning and international trade agreements2019-07-11T22:54:51+00:002019-07-11T22:54:51+00:00aelang.net/blog/2019/07/11/dataviz-tota<p><a href="https://dev.clariah.nl/files/dh2019/posters/0949.pdf"><img style="float:right;border:10px solid white" src="/TradeAgreementsPosterThumbnail.jpg" /></a></p>
<p>For the past few years I have acted as a “data host” for Benjamin Bach & Dave Murray-Rust’s annual <a href="https://sites.google.com/view/datafairs/home">Data Fair</a>, which they run for students on the MSc in Design Informatics here at Edinburgh. This involves bringing a dataset (or two or three) for a small group of students to work with, and producing a ‘data brief’ which explains the form and extent of the dataset, why it is interesting, particular challenges they may encounter while working with it and so forth. The students start by analysing the dataset and then must find ways to visualise it and tell the ‘data stories’ within it, in consultation with the data host who as the domain expert can help to orient them to the most interesting findings.</p>
<p>This year I was working with four students, Esteban Serrano, Oliver Ford and Xinyu Du, on a tranche of international trade law documents collected by the <a href="https://github.com/mappingtreaties/tota">ToTA: Text of Trade Agreements</a> project. They used word embeddings, topic modelling and network analysis to do some exploratory work investigating the similarities between these long and complicated documents. The work was accepted to the DH2019 conference as a poster: the abstract is available on the conference website <a href="https://dev.clariah.nl/files/dh2019/boa/0949.html">here</a>, and the poster itself can be downloaded by clicking the image to the right. There are some additional visualisations on <a href="https://ollieford.github.io/DS4D-Trade-Agreement-Project/">the students’ website</a>.</p>
<p>Students I have worked with over the years have come up with some beautiful and compelling ways to present the data visually, and if you are Edinburgh-based and have buckets of data lying around I would recommend <a href="https://sites.google.com/view/datafairs/submit-a-data-challenge">signing up</a>.</p>
<p><a href="http://creativecommons.org/licenses/by-nc-sa/4.0/"><img src="https://i.creativecommons.org/l/by-nc-sa/4.0/80x15.png" alt="Creative Commons License" /></a>
The material on this site is licensed under a <a href="http://creativecommons.org/licenses/by-nc-sa/4.0/">Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License</a>. Please also drop me a line at the address above to let me know what you’ve done with it. Thanks!</p>Lost in (Vector) Space: Seminar given at the British Library2019-04-29T10:14:55+00:002019-04-29T10:14:55+00:00aelang.net/blog/2019/04/29/britishlibraryseminar<p>Here are the slides from a seminar I gave at the British Library on 29 April 2019, titled <strong>“Lost in (Vector) Space: Using Machine Learning to Investigate Discourses of Spatiality in a 33 Million Word Corpus”</strong>. Thanks to <a href="https://twitter.com/mia_out">Mia Ridge</a> for the invitation.</p>
<p><br />
<a href="https://aelang.github.io/BLSeminarLang201904PDF.pdf"><img src="/bl_seminar_thumbnail.jpg" /></a>
<br /></p>
<p><a href="http://creativecommons.org/licenses/by-nc-sa/4.0/"><img src="https://i.creativecommons.org/l/by-nc-sa/4.0/80x15.png" alt="Creative Commons License" /></a>
The material on this site is licensed under a <a href="http://creativecommons.org/licenses/by-nc-sa/4.0/">Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License</a>. Please also drop me a line at the address above to let me know what you’ve done with it. Thanks!</p>Here are the slides from a seminar I gave at the British Library on 29 April 2019, titled “Lost in (Vector) Space: Using Machine Learning to Investigate Discourses of Spatiality in a 33 Million Word Corpus”. Thanks to Mia Ridge for the invitation.Spatialising the structure of an argument with reveal.js2017-03-17T21:13:25+00:002017-03-17T21:13:25+00:00aelang.net/blog/2017/03/17/reveal-js<p><img style="float:right;border:10px solid white" height="160" width="200" src="/reveal-screenshot.png" />Inspired by some elegant slide decks by <a href="https://cradledincaricature.com/">James Baker</a> and <a href="https://www.martineve.com/">Martin Eve</a> for their <a href="http://www.digital.hss.ed.ac.uk/blackbox/">Black Box</a> workshops, I’ve finally got around to trying <a href="https://github.com/hakimel/reveal.js/blob/master/README.md">reveal.js</a>, in the low-stakes context of an undergraduate lecture. What appeals to me the most about reveal.js is its ability to structure slides in both ‘vertical’ and ‘horizontal’ ways, ie. to use spatiality to signal to one’s audience the structure of one’s argument (and my students will testify that when it comes to their writing, I am all. about. structure). The ability to quickly zoom out and locate another slide is useful, and not being dependent on a Microsoft product is a nice bonus.</p>
<p>I experimented with Markdown but ended up going with HTML so I could control the line spacing (so as to get decent-sized chunks of the Waste Land on the screen). For those who want to try it, reveal.js lives <a href="https://github.com/hakimel/reveal.js">here</a>, and you can look at the source code of my <a href="https://aelang.github.io/data/reveal/#/">lecture slides</a> as a template.</p>Inspired by some elegant slide decks by James Baker and Martin Eve for their Black Box workshops, I’ve finally got around to trying reveal.js, in the low-stakes context of an undergraduate lecture. What appeals to me the most about reveal.js is its ability to structure slides in both ‘vertical’ and ‘horizontal’ ways, ie. to use spatiality to signal to one’s audience the structure of one’s argument (and my students will testify that when it comes to their writing, I am all. about. structure). The ability to quickly zoom out and locate another slide is useful, and not being dependent on a Microsoft product is a nice bonus.Stylo and the Stevensons2016-07-13T15:24:09+00:002016-07-13T15:24:09+00:00aelang.net/blog/2016/07/13/stylo-and-the-stevensons<p>Along with my PhD student Robyn Pritzker, I’ve been looking into the authorship of <a href="https://babel.hathitrust.org/cgi/pt?id=uva.x000181310;view=1up;seq=9"><em>The Dynamiter</em></a>, a text by Robert Louis Stevenson and his wife Fanny van der Grift Stevenson. Although it’s known that the pair collaborated on the writing, there’s no certainty about which sections were written by which author, and how much of the book can be attributed to Fanny. (For more background on this, see <a href="http://thedynamiter.llc.ed.ac.uk/"><em>Deciphering The Dynamiter</em></a>, which Robyn and <a href="http://thedynamiter.llc.ed.ac.uk/?page_id=17">her MSc colleagues</a> put together for their major project in my DH for Literary Studies class in 2015. They did a huge amount of work preparing the corpora and getting to grips with both the historical context and the literature on stylometry, and I’m very proud of them.) Robyn and I are now at the point of having some results to share, which I am presenting at <a href="http://dh2016.adho.org/">DH2016</a> in Krakow on behalf of us both.</p>
<p>Since <em>Deciphering The Dynamiter</em> went live in March 2015, a serendipitous thing happened: Maciej Eder released two new functions for the R library <a href="https://sites.google.com/site/computationalstylistics/stylo">Stylo</a>, rolling.classify and rolling.delta, which are designed specifically for the stylometric analysis of collaboratively written texts. These scripts work by slicing a text into overlapping chunks and performing analysis on each chunk using a ‘moving window’ approach, in order to discern the author whose most frequent words (MFW) in their reference texts most closely match the MFW of the test text. The available distance measures are nearest shrunken centroids, support vector machines and several flavours of Delta. Technical details <a href="http://dx.doi.org/10.1093/llc/fqv010">here</a>, or <a href="https://sites.google.com/site/computationalstylistics/stylo/scripts/stylo_0.6.3.tar.gz?attredirects=0&d=1">try it yourself</a>.</p>
<p>Using these rolling classification scripts, I generated some graphs that give a sense of where Fanny’s authorial signal comes through the text most strongly, and which show how her signal varies as the number of MFW rises. To show how the signal varies with the number of MFW, the graphs below have been animated, and they run from 100 MFW to 1000 MFW in increments of 100.</p>
<h4 id="the-dynamiter">The Dynamiter</h4>
<p><a href="https://aelang.github.io/Dyn100to1000MFWRollDelta4000-per-sliceFadeLooping.gif"><img style="float:left;border:10px solid white" src="/Dyn100to1000MFWRollDelta4000-per-sliceFadeLooping.gif" /></a>
<br /></p>
<p><em>The Dynamiter, 100 to 1000 MFW, classic Delta distance, 4000w slices, 3500w overlap</em></p>
<p><a href="https://aelang.github.io/Dyn100to1000MFWRollNSC4000-per-sliceFadeLooping.gif"><img style="float:left;border:10px solid white" src="/Dyn100to1000MFWRollNSC4000-per-sliceFadeLooping.gif" /></a>
<br /></p>
<p><em>The Dynamiter, 100 to 1000 MFW, NSC, 4000w slices, 3500w overlap</em></p>
<p><a href="https://aelang.github.io/Dyn100to1000MFWRollSVM4000-per-sliceFadeLooping.gif"><img style="float:left;border:10px solid white" src="/Dyn100to1000MFWRollSVM4000-per-sliceFadeLooping.gif" /></a>
<br /></p>
<p><em>The Dynamiter, 100 to 1000 MFW, SVM, 4000w slices, 3500w overlap</em></p>
<p>There are a few things to note about these graphs:</p>
<p>First, although the three distance measures are in rough agreement about the places where Fanny’s authorial influence emerges in the text, these vary depending on the number of MFW chosen, especially when Delta is used.</p>
<p>Second, the three distance measures give quite different pictures of the <em>extent</em> of Fanny’s contribution. Tomoji Tabata noted in <a href="https://www.conftool.pro/dh2016/index.php?page=browseSessions&form_session=174&presentations=show">yesterday’s Digital Literary Stylistics workshop</a> that in his tests of the three measures, SVM came out ahead at 0.975, followed by NSC at 0.9375 and Delta at 0.9025. When used on <em>The Dynamiter</em>, however, the SVM graph only allocates one tiny slice of the text to Fanny, and that very fleetingly (at 300MFW). The Delta graph identifies many different sections in which her style predominates, while the NSC graph is more consistent in the sections it assigns to her, but is often at odds with the Delta and SVM graphs.</p>
<p>Third, the collaboration seems not to have been governed by the sections of the book (shown by the dotted vertical lines, which represent the starting and ending points of the stories and stories-within-stories that constitute the book). Where the signal changes from one author to another, in other words, this does not necessarily align with any breaks between sections.
Seeking comparators to understand what these signals might tell us about Fanny and Louis’s collaborative writing process, I ran the same analyses on three volumes that Louis co-authored with his stepson Lloyd Osbourne: <em>The Wrecker</em>, <em>The Ebb Tide</em> and <em>The Wrong Box</em>.</p>
<h4 id="the-wrecker">The Wrecker</h4>
<p><a href="https://aelang.github.io/Wreck100to1000MFWRollDelta5000-per-sliceFadeLooping.gif"><img style="float:left;border:10px solid white" src="/Wreck100to1000MFWRollDelta5000-per-sliceFadeLooping.gif" /></a>
<br /></p>
<p><em>The Wrecker, 100 to 1000 MFW, classic Delta distance, 5000w slices, 4500w overlap</em></p>
<p><a href="https://aelang.github.io/Wreck100to1000MFWRollNSC5000-per-sliceFadeLooping.gif"><img style="float:left;border:10px solid white" src="/Wreck100to1000MFWRollNSC5000-per-sliceFadeLooping.gif" /></a>
<br /></p>
<p><em>The Wrecker, 100 to 1000 MFW, NSC, 5000w slices, 4500w overlap</em></p>
<p><a href="https://aelang.github.io/Wreck100to1000MFWRollSVM5000-per-sliceFadeLooping.gif"><img style="float:left;border:10px solid white" src="/Wreck100to1000MFWRollSVM5000-per-sliceFadeLooping.gif" /></a>
<br /></p>
<p><em>The Wrecker, 100 to 1000 MFW, SVM, 5000w slices, 4500w overlap</em></p>
<h4 id="the-ebb-tide">The Ebb Tide</h4>
<p><a href="https://aelang.github.io/Ebb100to1000MFWRollDelta5000-per-sliceFadeLooping.gif"><img style="float:left;border:10px solid white" src="/Ebb100to1000MFWRollDelta5000-per-sliceFadeLooping.gif" /></a>
<br /></p>
<p><em>The Ebb Tide, 100 to 1000 MFW, classic Delta distance, 5000w slices, 4500w overlap</em></p>
<p><a href="https://aelang.github.io/Ebb100to1000MFWRollNSC5000-per-sliceFadeLooping.gif"><img style="float:left;border:10px solid white" src="/Ebb100to1000MFWRollNSC5000-per-sliceFadeLooping.gif" /></a>
<br /></p>
<p><em>The Ebb Tide, 100 to 1000 MFW, NSC, 5000w slices, 4500w overlap</em></p>
<p><a href="https://aelang.github.io/Ebb100to1000MFWRollSVM5000-per-sliceFadeLooping.gif"><img style="float:left;border:10px solid white" src="/Ebb100to1000MFWRollSVM5000-per-sliceFadeLooping.gif" /></a>
<br /></p>
<p><em>The Ebb Tide, 100 to 1000 MFW, SVM, 5000w slices, 4500w overlap</em></p>
<h4 id="the-wrong-box">The Wrong Box</h4>
<p><a href="https://aelang.github.io/Wrong100to1000MFWRollDelta5000-per-sliceFadeLooping.gif"><img style="float:left;border:10px solid white" src="/Wrong100to1000MFWRollDelta5000-per-sliceFadeLooping.gif" /></a>
<br /></p>
<p><em>The Wrong Box, 100 to 1000 MFW, classic Delta distance, 5000w slices, 4500w overlap</em></p>
<p><a href="https://aelang.github.io/Wrong100To1000MFWRollNSC5000-per-sliceFadeLooping.gif"><img style="float:left;border:10px solid white" src="/Wrong100To1000MFWRollNSC5000-per-sliceFadeLooping.gif" /></a>
<br /></p>
<p><em>The Wrong Box, 100 to 1000 MFW, NSC, 5000w slices, 4500w overlap</em></p>
<p><a href="https://aelang.github.io/Wrong100to1000MFWRollSVM5000-per-sliceFadeLooping.gif"><img style="float:left;border:10px solid white" src="/Wrong100to1000MFWRollSVM5000-per-sliceFadeLooping.gif" /></a>
<br /></p>
<p><em>The Wrong Box, 100 to 1000 MFW, SVM, 5000w slices, 4500w overlap</em></p>
<p>Here again the signal is far from conclusive across the three distance measures, though it seems fairly safe to say that Louis appears to have heavily edited the texts he co-wrote with Lloyd. Taken together, the graphs suggest that Louis had something of a different mode of collaboration with Lloyd compared to the way he and Fanny worked together. For <em>The Wrong Box</em>, there is a relatively strong signal to suggest that Lloyd was the person who played the largest role in writing the ending of the book. For <em>The Ebb Tide</em>, there are well-delineated sections in the NSC graphs where Lloyd’s signal is clearly present, so the question here is not whether he wrote those sections but to what extent his style emerges over Louis’s generally more dominant voice. <em>The Wrecker</em> is the most confounding of all: the NSC graph suggests that Lloyd had influence only over a few very small slices of the text, while the Delta and SVM graphs tell a different story in suggesting that Lloyd’s influence is present, even if not dominant, across much of the text. If we accept Tabata’s idea that NSC is the most reliable distance measure, Lloyd’s contributions to <em>The Wrecker</em> and <em>The Wrong Box</em> are very minimal, and his influence in <em>The Ebb Tide</em> more considerable. But the story told by the Delta and SVM graphs is more complicated, and suggest that Lloyd’s input into the drafting process was more significant than the NSC graph allows. So, if NSC can be considered somewhat too black-and-white in the way it attributes authorship, then what these graphs suggest is that where the process of authoring has been an intricate and intermingled one – ie. where the authors have drafted and edited together, and/or edited and rewritten each others’ materials, rather than each writing discrete sections – NSC may not be such a useful measure. Rather, the greater nuance of SVM and Delta may be better suited to observing the relative prominence of authorial signals.</p>
<p>Robyn will be working in Fanny’s archives over the next few months and we hope she will turn up evidence there which will help to illuminate – and complicate – the information in the graphs above. For now, my tentative interpretation is that the presence of Fanny’s signal at lower MFW and its tendency to disappear at higher MFW is an indication that she played an important role in the early conceptualising and drafting, while Louis contributed proportionally more in the later stages of redrafting and editing, which accords roughly with the (sparse) evidence we have in front of us now. Eder comes to <a href="http://dx.doi.org/10.1093/llc/fqv010">a similar conclusion</a> for this pattern in Conrad and Ford’s novel <em>The Inheritors</em>, and Mark Algee-Hewitt presented yesterday on some preliminary work on co-authored Stanford Lit Lab pamphlets which is particularly tantalising, as the authors of these co-written texts are still around and can be asked how the writing and editing actually occurred.</p>
<p>For more detail on the findings and how we have interpreted them – as well as why this kind of approach is exciting for feminist literary historians seeking to recuperate women’s contributions to literary history – come along to our DH2016 paper (<a href="https://www.conftool.pro/dh2016/index.php?page=browseSessions&form_session=38&presentations=show">“All In the Family: Testing Burrows’ Delta on Robert Louis Stevenson’s Collaboratively Authored Volumes <em>The Dynamiter</em> and <em>The Wrecker</em>”</a>).</p>
<p><a href="http://creativecommons.org/licenses/by-nc-sa/4.0/"><img src="https://i.creativecommons.org/l/by-nc-sa/4.0/80x15.png" alt="Creative Commons License" /></a>
The material on this site is licensed under a <a href="http://creativecommons.org/licenses/by-nc-sa/4.0/">Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License</a>. Please also drop me a line at the address above to let me know what you’ve done with it. Thanks!</p>Along with my PhD student Robyn Pritzker, I’ve been looking into the authorship of The Dynamiter, a text by Robert Louis Stevenson and his wife Fanny van der Grift Stevenson. Although it’s known that the pair collaborated on the writing, there’s no certainty about which sections were written by which author, and how much of the book can be attributed to Fanny. (For more background on this, see Deciphering The Dynamiter, which Robyn and her MSc colleagues put together for their major project in my DH for Literary Studies class in 2015. They did a huge amount of work preparing the corpora and getting to grips with both the historical context and the literature on stylometry, and I’m very proud of them.) Robyn and I are now at the point of having some results to share, which I am presenting at DH2016 in Krakow on behalf of us both.Introduction to Digital Humanities: Lecture at QMUL summer school2015-07-23T21:13:25+00:002015-07-23T21:13:25+00:00aelang.net/blog/2015/07/23/qmul-lecture<p>On Thursday 23 July I gave an introductory lecture on digital humanities at the ADEPT summer school at Queen Mary University of London (the second such presentation on DH that I’ve done in the past month). Here are the slides, and the Storify is below.</p>
<p style="text-align: center;"><iframe style="border: 1px solid #CCC; border-width: 1px; margin-bottom: 5px; max-width: 100%;" src="//www.slideshare.net/slideshow/embed_code/key/DxtKrJxbMsyEn9" width="510" height="420" frameborder="0" marginwidth="0" marginheight="0" scrolling="no" allowfullscreen="allowfullscreen"> </iframe></p>
<p><a rel=""><img alt="" /></a><br />
Licensed under a <a rel="">Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License</a>.</p>
<div class="storify"><iframe src="//storify.com/Ed_Dev_QMUL/introduction-to-the-field-of-digital-humanities/embed?border=false&template=grid" width="100%" height="750" frameborder="no"></iframe><script src="//storify.com/Ed_Dev_QMUL/introduction-to-the-field-of-digital-humanities.js?border=false&template=grid"></script><noscript>[<a href="//storify.com/Ed_Dev_QMUL/introduction-to-the-field-of-digital-humanities" target="_blank">View the story “Introduction to the Field of Digital Humanities” on Storify</a>]</noscript></div>On Thursday 23 July I gave an introductory lecture on digital humanities at the ADEPT summer school at Queen Mary University of London (the second such presentation on DH that I’ve done in the past month). Here are the slides, and the Storify is below.Thirty-three ways of looking at a DHSI week2014-06-08T21:20:43+00:002014-06-08T21:20:43+00:00aelang.net/blog/2014/06/08/dhsi<p>(Cross-posted from the <a href="http://editingmodernism.ca/2014/06/thirty-three-ways-of-looking-at-a-dhsi-week/">Editing Modernism in Canada</a> website)</p>
<p>In 2012, after a brilliant week of taking the GIS course at DHSI, I wrote a post with <a href="http://editingmodernism.ca/2012/06/twenty-two-reasons-to-go-to-dhsi/">twenty-two reasons to go to DHSI</a>. I did not think it was possible, but this week has been even better than that one, in part because I am now much further in to my work with digital methods, and my comfort levels are much higher. So, herewith a followup post, two years on, with eleven further reasons to convince anyone who hasn’t yet drunk the kool-aid, or to choose a more database-appropriate metaphor, taken the red pill.<br /><br />
1) First of all, the feeling of being stretched and challenged in entirely unpredictable ways: of being taken way outside of my comfort zone, having my object of study taken away from me, exploded into unrecognisable pieces, conceptually reassembled, and handed back to me. It’s a cliche that DH forces us think differently about what we do and how we do it, but the generative experience of actually doing this intensely for an entire week with experts in the room is an immense privilege, in a world where the imperatives to knock out REF-able publications, achieve ‘impact’ with our research, climb ever higher on departmental and university rankings tables and so forth - these are our UK imperatives; there are others in the north American context - militate against the deep immersion that we need to do our best work.<br /><br />
2) I’m going to give the Databases course its own reason to go to DHSI. Yes, really. Databases. It sounds dismal. It’s anything but. Take this course if your research involves even a moderate amount of data-gathering (hint: if you are using Excel and find yourself adding extra columns, putting in lots of null values and becoming dissatisfied with the way your tables represent whatever it is you are studying, a relational database is probably what you want instead). Harvey is hilarious, and long impromptu riffs on an eclectic medley of pop culture texts turn into serious points about the theoretical and practical exigencies of working with relational databases. The pedagogical component of this course is top-notch: he has thought very hard about how to make the material intuitively graspable by people who are not trained as computer scientists. Come for the standup, stay for the profound transformations to your thinking.<br /><br />
3) Almost as good as databases is … free beer! A prize for accidentally walking up to the registration table at exactly three o’clock. I didn’t think I could be happier about arriving in Victoria under deep blue skies, reconnecting with EMiC friends and starting a week of databases, but turns out free beer will do the trick.<br /><br />
4) An antidote to years of finding the command line intimidating. Decades, in fact: our first family computer had an MS-DOS prompt into which my brother typed arcane magic words, and the helplessness I felt watching the screen fill up with glowy green type is a visceral memory. After a 45 minute unconference session with Jonathan Martin, it is intimidating no more, and in fact to my surprise somewhat intuitive. For those who want to try, the resource we used is called <a href="http://cli.learncodethehardway.org/book/">Learn Code the Hard Way</a>.<br /><br />
5) The vault of collective knowledge that it is possible to tap into via the #dhsi2014 hashtag. Trying to quickly clean up a data set, I sent out a plaintive cry for help with a regex to remove URLs, and three lovely people answered in under a minute with suggestions. (Happy ending: I figured it out myself! Woohoo! It is the world’s ugliest regex, but it works.)<br /><br />
6) Twitter, which this year was something of a different experience. There was absolutely no chance of following the #dhsi2014 backchannel, given that new tweets popped up on it roughly every three and a half seconds, and in fact the TAGS spreadsheet keeping the twitter archive broke. So, because I couldn’t follow it completely, I dipped in and out. While I know I will have missed many things, I feel like I still got something of a decent sense of many of the most interesting conversations - the conversations about gender happening in #femdh, or Susan Brown’s wisdom about managing large and long-running projects in the #cwrcshop, for instance - which could then be followed up in meatspace by collaring friends who had been sitting in those classes. I do feel marginally less tapped out brain-wise than I have in previous years, so that’s a plus. DHSI is obviously not getting any smaller in the near future, so we’re all going to have to find other ways of filtering the backchannel.<br /><br />
7) Serendipity, which always happens at DHSI: it is like some kind of magic fairy dust that Ray arranges to have sprinkled around the campus. Discovering that the person standing behind me in the line at the Monday night reception was a topic modelling guru who graciously let me pull out my laptop so he could show me around a topic modelling tool with a nice GUI. Finding that Emily Robins-Sharpe also teaches transnational modernisms and swapping notes on texts we put on our syllabi. Giving a paper at the colloquium and getting tweets from people working on similar areas, and meeting afterwards to share resources.<br /><br />
8) Victoria and its serene beauty: the fir trees, the beach, deer grazing on the cluster lawn in the twilight, a faun trotting after its mother, a lone bunny hopping past a cluster house. #comebackbunnies<br /><br />
9) Meeting people, all the people, all the time. People who I have gotten to know virtually and whose work I have admired from afar - Scott Weingart, Paul Fyfe, Alex Gil - and whom it was a pleasure to finally meet in person and continue conversations that had already started online. People with whom there’s lots of common ground research-wise: Paul Barrett, for instance, who is doing some intriguing work on topic modelling the archive of Austin Clarke, and Alana Fletcher, who knows the Queen’s University archives like the back of her hand and who is, happily, willing to work on digitizing the Crawley materials. And of course picking up with everyone with whom the Atlantic ocean gets in the way of hanging out on an everyday basis: my excellent Twentieth-Century Literary Letters project collaborators; my fabulous housemates Hannah, Lee & Karis who have all three had wonderful professional successes since the last time we caught up, which make me happy to know them and to be associated with them through EMiC. Let me have a shot at doing this as a left join …<br /><br /></p>
<blockquote>
<div><span style="font-family: Courier New">USE dhsiclub</span></div>
<div><span style="font-family: Courier New"> </span></div>
<div><span style="font-family: Courier New">SELECT firstname, lastname, ", ", role_name</span></div>
<div><span style="font-family: Courier New">FROM role, person_role, person, dhsi_attendees, dhsi_class</span></div>
<div><span style="font-family: Courier New">WHERE dhsi_class.class_id=dhsi_attendees.class_id, dhsi_attendees.person_id=person.person_id, person.person_id=person_role.person_id, person_role.role_id=role.role_id</span></div>
<div><span style="font-family: Courier New">AND dhsi_class.class_year=2014</span></div>
</blockquote>
<p><br /><br />
10) Developing my thinking about the letter as cultural artefact, and about the work of transmission that it performs. Thinking beyond its existence within document culture (see, Harvey, I’m learning) and reimagining it within database culture, its magical power to connect not just people but places, texts and ideas comes to the fore. And tabulating those elements in a database enables us to query them, ie. relate them, in ways that we cannot if we look at them within document culture.<br /><br />
11) Finally, if I had to pick one overriding quality of DHSI, it would be generosity. I’ve had the privilege of participating in DHSI over the past six years because Dean was generous (farsighted? foolish?) enough to invite me, as an early-career scholar, to be a co-applicant on EMiC. The ramifications of this act of generosity will continue to ripple out for a long time for me, and I have a difficult time articulating how important my involvement in EMiC has been, especially as I write this at the end of a week in which all the words have gone because all the brain is full. Along with Dean’s generosity, there’s the generosity of other Canadian faculty and graduate students who have been welcoming to an interloper with a peculiar interest in the literature of their country. (‘Why would anyone outside Canada want to study Can lit?’ is the question that emerges, politely, sooner or later in most conversations. The answer, for the record, is: because awesome.)<br /><br />
EMiC colleagues, from the most eminent senior scholars to just-beginning graduate students, I can’t tell you how much your friendship and your intellectual company means to me, at DHSI and the other occasions when I’m fortunate enough to hang out with you physically, virtually, or on the page. Thank you all, and may we continue to be <del>left joined</del> connected for many years to come.<br /><br />
<br /><br />
<br /><br />
<a href="http://creativecommons.org/licenses/by-nc-sa/4.0/"><img src="https://i.creativecommons.org/l/by-nc-sa/4.0/80x15.png" alt="Creative Commons License" /></a>
The material on this site is licensed under a <a href="http://creativecommons.org/licenses/by-nc-sa/4.0/">Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License</a>. Please also drop me a line at the address above to let me know what you’ve done with it. Thanks!</p>(Cross-posted from the Editing Modernism in Canada website)Twenty-two reasons to go to DHSI2012-06-13T20:50:15+00:002012-06-13T20:50:15+00:00aelang.net/blog/2012/06/13/dhsi<p>(Cross-posted from the <a href="http://editingmodernism.ca/2012/06/twenty-two-reasons-to-go-to-dhsi/">Editing Modernism in Canada</a> website)</p>
<p><strong>Things you don’t even know about yet that will make your life better.</strong> <a href="http://pressbooks.com/">PressBooks</a>, which lets you produce beautifully formatted e-books in multiple formats: thank you <a href="https://twitter.com/isleofvan/status/210788123012120577">@isleofvan</a>. Free digitized historical maps from the <a href="http://www.davidrumsey.com/">David Rumsey collection</a>. <a href="http://www.paigemorgan.net/visibleprices/">Visible Prices</a>, a tool by <a href="https://twitter.com/#%21/paigecmorgan">@paigecmorgan</a> which uses the SIMILE scripts to help render historical prices in literary texts comprehensible to modern day readers. <a href="http://oxgarage.oucs.ox.ac.uk:8080/ege-webclient/">OxGarage</a>, a tool for converting just about anything to (lightly marked up) TEI. The beautiful <a href="http://vangoghletters.org/">Van Gogh Letters project</a>, whose elegant interface has helped me to think through some of the design quandaries I’d been puzzling over for my own EMiC project: thank you Beth Popham. Many other resources flowed past me on presentation screens and on Twitter, too many to list here.<br /><br />
<strong>Things that make you think differently about how you do your own work.</strong> Using text mining in the attempt to map conceptual space alongside geographical space in Olaudah Equiano’s <em>Narrative</em> (a project by one of my GIS classmates, Elizabeth Maddock Dillon at Northeastern). Using tags to get students to think about the thematic content and structure of their writing (@chris_friend).<br /><br />
<strong>Things that will make you more literate</strong>, not just as a digital humanist but as a member of a mediated society. One unconference session taught me the basic principles behind machine translation (= statistics, not linguistics). Another drew back the curtain on the curious inefficiencies in transforming printed books into e-books, and challenged those of us in the room not just to rethink the structure behind the model, but to imagine ways to change it.<br /><br />
<strong>That moment when the technology you’ve been wrestling with the whole week actually works</strong>, and <a href="https://twitter.com/a_e_lang/status/211144979312160768">you manage to do the thing that you came to DHSI specifically to do</a>. Euphoria!<br /><br />
<strong>Vicarious pride</strong> when @jmhuculak gives a demo of the the EMiC modernist commons and <a href="https://twitter.com/heatherfro/status/210527310045122561">virtual flutters of delight ripple out on the tweetstream</a>. This is completely unjustified on my part as the construction of the commons is nothing to do with me, and everything to do with the brilliant folks at Islandora, but when Matt gave the demo, being a part of EMiC <a href="https://twitter.com/a_e_lang/status/210528354842054656">made me happy</a>. What a fabulous bunch of people, and what an awesome resource this will be.<br /><br />
<strong>Collegiality.</strong> The ability to explore research questions in new ways with technology might have been what initially brought me to DH, but what keeps me here is the friendliness of the community. In addition to an emphasis on mentoring, there’s also a close critical attention to questions of academic labour, and the material consequences of institutional practices, especially on early-career scholars (on edit: see <a href="http://editingmodernism.ca/2012/06/tenure-lack-alt-ac-and-generally-talking-back/">Alana’s post</a>).<br /><br />
<strong>Sharing the love.</strong> Three of us got talking on the shuttle from the airport; one person had not heard of <a href="http://www.zotero.org/">Zotero</a>, and her eyes got rounder and rounder as we explained how it pulled bibliographic data directly from the web into a browser plugin where it could then be sorted into libraries, published to the web or dropped into a Word document where it would appear in an already-formatted output style of her choosing. @benwbrum, who I don’t think was even at DHSI, was <a href="https://twitter.com/benwbrum/status/210117034741211136">happy to hear about the Google doc from the regular expressions session</a> for his regex-for-humanists workshop (and actually … where can I sign up for that?). Meanwhile <a href="http://neatline.scholarslab.org/">Neatline</a> will be exactly the thing that a non-DH colleague of mine at Strathclyde needs for the C18th correspondence project she’s in the middle of grant-writing for, though she doesn’t yet know it.<br /><br />
<strong>Generosity.</strong> This time last year <a href="https://twitter.com/#%21/mbtimney">@mbtimney</a> graciously gave up a lunch hour to do a WordPress install on my server. And I’m particularly happy to have been invited to bring the data from my map of <em>Memoirs of Montparnasse</em> to the Paris map app that Paul Hjartarson, @hquamen and others at UoA are developing for the <a href="http://www.cwrc.ca/projects/infrastructure-projects/pilot-projects/editing-the-sheila-watson-wilfred-watson-letters/">Sheila and Wilfrid Watson letters</a>. It’s a great example of the connections between editorial projects that EMiC has helped to facilitate.<br /><br />
<strong>The fact that there’s a decent chance that the person teaching your class, chatting to you in the coffee queue or retweeting your observation is at the top of their field.</strong> I had the chance to learn humanities GIS from Ian Gregory. The Omeka demo at the unconference was given by <a href="https://twitter.com/#%21/clioweb">@clioweb</a> who (I think) helped to develop Omeka. And I didn’t realise in my first year at DHSI that one of the people out the front of the class, Julia Flanders, was a world expert in TEI, because she was so approachable and no question was too basic for her to answer.<br /><br />
And leading on from the two points above: <strong>an emphasis that we are all learning.</strong> Expertise in this field can only ever be in process, and those with that expertise are almost always more than happy to share it.<br /><br />
<strong>Things you’ve been longing to learn which someone volunteers to teach during the lunchtime unconference.</strong> <a href="https://docs.google.com/document/d/1sYtTOZNNZGDpynB9FR_h6WkhTdKJ2mzo3r49K-L3KE4/edit">Regular expressions</a>. <a href="http://www.qgis.org/">QGIS</a>. <a href="https://docs.google.com/document/d/1EENEmyxLRmDvGrGsexb226N8-DXoLImmNJwE-8nakJQ/edit">Python</a>. While an hour or so isn’t enough to learn these in any systematic way, it is still amazing how much easier it is to figure something else out on your own once you’ve had a quick introduction and the chance to ask questions.<br /><br />
<strong>The serendipitous possibilities</strong> that open up when you move outside the orderly conventions of regular academic conferences and ask people to organise themselves into sessions based on what interests them right at that moment, rather than nine months earlier when everyone wrote abstracts based on what they imagined their findings would be. There is a place for carefully written conference papers and formally structured sessions, of course. But when you take a group comprised of, say, librarians, publishing interns, university press employees, TEI devotees, TEI sceptics and academics, and give them a chance to debate the ideas raised in a colloquium session about academic publishing and shifts in the infrastructure of book production and distribution, you create the conditions for all sorts of intriguing insights to arise from the meeting of different disciplinary viewpoints and different forms of practical expertise.<br /><br />
<strong>Twitter.</strong> This post is a testament to the many useful things I’ve discovered through Twitter, and indeed for many people it needs no introduction. If you are yet to use it for scholarly purposes, though, then there are two reasons to start. First, as one participant pointed out - on where else but Twitter - having conversations with other people on <a href="https://twitter.com/#%21/search/realtime/%23dhsi2012">the #DHSI2012 backchannel</a> makes it a fair bit easier to talk to them in person. (And also convenient: someone I was exchanging messages with turned out to be sitting behind me, and quickly walked me through the tool we’d been discussing when the colloquium session was over.) Second, for those who feel overwhelmed with the sheer number of projects, clever tools, programming languages etc that it feels like you have to keep a handle on, Twitter can function as a useful filter. Once you are following enough of the right people, you can more easily keep track of which resources are being used and are generating excitement, and you can prioritise those.<br /><br />
<strong>Using your brain in a different way.</strong> Attending non-DH conferences now feels to me like a curiously one-dimensional affair. Following three threads of a lunchtime unconference on the backchannel while sitting in a fourth, with part of my head trying to figure out a problem that came up with my data that will need to be fixed that afternoon, I emerge feeling like I’ve been to not one but multiple DHSIs. It can be draining, and requires a different form of attention to the kind one gives to a task such as immersive reading, but given that travel and time are both expensive, I very much appreciate the way that DHSI packs a great deal into a short space of time.<br /><br />
<strong>Excellent, inspiring presentations</strong>, many of them by talented graduate students. Digital humanists think hard about the way they present information, whether it is part of a Prezi slideshow or a digital edition. (And even when what is being presented is inspired silliness, <a href="http://vimeo.com/43692769">the production values are still high</a>). I learned much this week about how to engage an audience without compromising the substantive content of the message.<br /><br />
<strong>Access to tools you might not have at home.</strong> ArcGIS is proprietary software, and expensive, but with a little bit of running between sessions on the final day, I got it to crunch through a massive dataset of Paris roads, polygons and points and extract the bits I needed. Seven minutes before my shuttle to the airport was due to leave on the other side of campus, it finished its export and now I have a set of shape files I can work with in the open-source program QGIS. Result.<br /><br />
<strong>Twice-in-a-lifetime astronomical events</strong>, conveniently-placed telescopes on top of the astronomy building through which to view them, and charming UVic astronomy profs and grad students to explain in lay terms what is going on. (OK, so the transit of Venus may not happen every year …)<br /><br />
<strong>The joy of building.</strong> Lots has been written about <a href="http://projectroomseattle.org/2012/03/making-things/">the ethos of building in DH</a>, and this is amply in evidence at DHSI. Of the apps that were showcased in the final session, my favourite came from <a href="https://twitter.com/#%21/mchlstvns">@mchlstvns</a>: in a few days he built an app for walking around Dublin in the footsteps of Leopold Bloom. And the #digiped folks didn’t just talk about digital pedagogy for a week, they built <a href="http://web.uvic.ca/%7Eenglblog/pedagogydhsi/">a resource for the rest of us</a>.<br /><br />
<strong>A chance to make your own work meaningful.</strong> I had fun building <a href="http://aelang.net/projects/canada.htm">a SIMILE map as a sandbox</a> for other lit folks to use if they wanted to explore the geographical dimension of a text without going to the bother of constructing a map interface themselves. After various conversations at Monday’s reception, it will now have a few more users, including some at Algoma who I may never meet, but I’m happy if they can make use of it.<br /><br />
<strong>Connections.</strong> A grant call crossed my desk the other day for bringing people together for expert meetings, and I couldn’t quite see how to use it at the time. After the mapping and visualisation session, though, I now know exactly how it could be used not just to drive forward my own EMiC project, but also support the research agendas of several other people and projects.<br /><br />
<strong>Old-school face-to-face conversations.</strong> Learning that another EMiC project needs to investigate the same archive I do, and that we can perhaps team up and save some time and labour. Comparing notes on how to keep on top of an ever-shifting array of technologies alongside the research and teaching and service we’re expected to do as part of our jobs. If I’m lucky I’ll see EMiC folks here and there throughout the year at conferences, and exchange the odd email with a few people, but DHSI brings home to me that nothing substitutes for the sustained conversations and relationships that are made possible when people come together in the same physical space. (Which is entirely appropriate for a project which is interested in the social networks and geographical placedness of Canadian modernism.)<br /><br />
<strong>New friends as well as old.</strong> Hello, fellow Aussie Anya, fellow Londoner Bo, and fellow Canadianist-outside-Canada <a href="https://twitter.com/#%21/readywriting">@readywriting</a>. Hello, Georgia Tech librarians in my GIS class who were surprised to learn that their Vertically Integrated Projects had been exported to Strathclyde as one of the ways we teach DH. Hello, talented literary folks scattered across the US and the Pacific who are pursuing the same kinds of research questions I am with maps, networks, visualization and modernism. I am so happy to have made your acquaintance!<br /><br />
And this is not even to begin on the breathtaking beauty of Victoria, tame deer grazing casually outside the student residences, local BC beers, and the impressive quality of EMiC’s collective karaoke chops.<br /><br />
Thank you, Dean, and thank you, EMiC colleagues and friends. It’s been a fabulous, invigorating, beyond exhausting week. And now I’m going to crash.<br />
<br />
<br />
<a href="http://creativecommons.org/licenses/by-nc-sa/4.0/"><img src="https://i.creativecommons.org/l/by-nc-sa/4.0/80x15.png" alt="Creative Commons License" /></a>
The material on this site is licensed under a <a href="http://creativecommons.org/licenses/by-nc-sa/4.0/">Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License</a>. Please also drop me a line at the address above to let me know what you’ve done with it. Thanks!</p>(Cross-posted from the Editing Modernism in Canada website)