Looks good on paper…

Home » science » tech

Category Archives: tech

A year is a long time in running

Well, well, well… It’s been a looooong year, but here we are, at the very end of it. I thought I would reflect on some of my running goals, the peaks and troughs, and what running resolutions I might set for 2017.

So how did it go? All in all, I don’t think it was a bad year. There wasn’t a lot of progress in pacing, per se, but I was a lot steadier and more consistent in getting out for runs, and I had some really good months (unsurprisingly during the summer).

Volume Highlight Lowlight
January 32.4 km over 4:51 across 6 runs It’s hard to remember this far back… The usual New Year weather blues!
February 94km over 8:31 across 16 runs Picking up training volume again, including a 54:23 10k The number of runs that are just shy of 5k
March 90.4km over 8:13 across 14 runs First 10k+ run (13.7km in 1:13) That my average pace starts a slow decline here
April 94.2km over 9:06 across 15 runs Steady as she goes… Noticeable pace decrease, including two 10ks (57 and 59)
May 131.8km over 12:05 across 13 runs Two half-marathons (first ever at 1:57, second at 1:58) That I was too cheap to pay for the Runner’s World app training…
June 91km over 8:16 across 12 runs I at least managed an 8-miler I couldn’t face a long long-run this month, having done halves in May
July 112.82km over 10:06 across 17 runs A good increase in the number of runs Quantity =/= distance
August 136.4km over 12:25 across 22 runs Still on the up, including some 10-milers and some 8-milers Average pace is down, though…
September 113.1km over 10:47 across 19 runs Holding fairly steady and enjoying the sunshine Only one 8-miler
October 60km over 5:32 across 12 runs Despite travel and moving house, I’m still managing a few runs a week Most runs now pretty short
November 70.75km over 6:29 across 15 runs Got in a 10k and a 5-miler Pace is still trending down
December 76.18km over 7:05 across 14 runs Starting to pull back up again, distance-wise  Still a bit slower

I have found myself slightly frustrated with Strava after discovering that it didn’t have the delightful Excel download feature of RunKeeper, so pulling together my stats was a bit of a hassle. I’m going to have to think about whether I try Nike+ Run Club (which I think lets you download stats), or the paid-for version of Strava.

Highlights

  • The two half-marathons I did, just by myself.

I’m happy that my first one ever was supposed to be a 10-miler that I just pushed on with, and it came in sub-2:00. But the fact that I only managed two and then my half-marathon-a-month plan fell apart is indicative of a fairly basic fact: I’m not that interested in longer distances.

  • Running is more part of my everyday routine than in 2015.

There was almost no week when I didn’t run at all. I had some niggles, like shoulder cramps (yep, a weird one) or tight ankles, but still trundled on when and how I could.

Lowlights

  • That the half-marathon-a-month plan fell by the wayside so easily.

I’m partly baffled by quite why this happened, and partly glad that it helped clarify quite what I’m running for.

  • That I still haven’t figured out how best to get speed workouts into my routine.

I do these workouts intermittently, when I’m feeling guilty or in a clear rut. If I want to “progress”, which I guess means getting faster,  then I need to take these more seriously, as well as strength training, etc.

  • That my pacing seems to have decreased.

If this is a function of running longer, then it’s disappointing that I haven’t picked up a great deal across the year, which may reinforce the point above about speed workouts.

Online publishing and pedagogy: some thoughts from working on NAVSA’s COVE

I have been working on various parts of NAVSA‘s Central Online Victorian Educator (COVE) for six months now, and I thought it would be a good time to reflect on the experience, as well as on how COVE might serve educators in the future.

(more…)

Are programming languages languages? A question about digital humanities CVs

It’s that time of year, and I’ve been refreshing my academic CV to see how it’s taking shape. Although I’m a part-time PhDder and work full-time in another industry, I try to be as active as I can in academic endeavours, so there’s plenty of updates and tweaks to be made.

Taking a look at some colleagues’ CVs for tips and hints on structure, etc., it began to strike me that those of us who are interested in digital humanities face a challenge: where do our programming skills belong?

Sometimes, it’s easy. There is a role or project that involved particular development work. But increasingly people put HTML, or CSS, or Java alongside French or Japanese, with a “fluent” beside them (I’ve yet to see anyone using an ILR or CEFR scale marking, though). The first time I saw this I smiled a little smile to myself. Well, we’ve got to try to get these things on the CV somehow. But after a while, I started to wonder: when we say this, do we mean it? As a literary scholar, I have strong feelings about the fact that words mean things. And as a spouse of a professional developer, I feel sceptical.

What does it mean when we say we are “fluent” in a programming language? Is coding ever so smooth, graceful and effortless as speaking or writing in a ‘natural’ language? Do we ever code easily and accurately in the same way as we might speak our mother tongue?

The rate of bugs or errors in your average professional code—the type produced by developers at Apple or Google—is about 1 every 10 lines. If one were to have a typo or lexical error in every 10 sentences of written work, one’s supervisor would raise a red flag!

Now, perhaps we may quibble about whether a line of code is equivalent to a sentence (or, indeed, the comparison to a mother tongue), but I think the point stands. The rate of error is much higher in professional “fluent” code than in academic written prose.

Similarly, professional coders do not for the most part store the whole or majority of the language intuitively in their brain in the same way as one’s mother tongue. The reliance on coding reference materials, such as Stack Overflow, is higher than the reliance on linguistic ones, such as dictionaries. Tales of developers finding a perfect and helpful answer on such repositories, only to discover that they themselves submitted it years before, are legion. What would be the equivalent in terms of a ‘natural language’? Stumbling across the idea of using consonance to convey a certain effect, having forgotten that it existed?

All in all, reading “fluent in HTML” on an academic CV strikes me as a category mistake. And it makes me wonder about their French, too…

Teaching without a class

553697

Yesterday, Amazon announced its steps into the online education market with Amazon Inspire. The growing expansion of online education, from the early days of VLEs to MOOCs in all disciplines and languages, has me thinking a lot about teaching and how it is increasingly escaping the physical classroom.

As a PhD student, ordinarily I would be expected to undertake some teaching, in particular if I aim to enter academia after I earn my PhD. As a part-time, long-distance student with a full-time job, however, the reality is that regular teaching isn’t possible, even if my institution’s admin team were working proactively to try to make it happen for students in my situation (spoiler alert: they’re not).

Being a sometime MOOC fan, and now working with Dino Felluga and his team on BRANCH (Britain, Representation, and Nineteenth-Century History), I’ve been thinking about how digital media might make it possible to teach outside the traditional classroom.

Ordinarily, we think of online academic provision from a learner’s perspective: how well are their needs met, how robust is the learning and examining process, how does such an education compare with traditional university experiences? However, what if we diverge for a moment from models of academic provision that remain centred on traditional institutions?

MOOCs mostly follow traditional course structures with weekly homework, readings complemented by video lectures, often delivered with PowerPoint, and forums to replace office hours, where teaching assistants answer queries. Coursera restricts itself to organisational “partners”—for-profits, universities, not-for-profits, etc.—and edX does the same. That means that, for teachers, MOOCs do not remove barriers but reinforce the priority of academic institutions over the individual educators that make them up.  Academics, particularly PhD students and ECRs often talk about blogging, online-only journals, etc., as a way to get our scholarship out there, but what about the opportunities for getting our pedagogy out there too? The articles on BRANCH, for instance, are certainly educative, but they are scholarship, not pedagogy. 

Lots of lecture courses from traditional universities make their way onto YouTube or Udemy as platforms, but it seems that the possibility for cross-fertilisation of ideas—between students, students and lecturer, etc—is limited. Such platforms are “stuck on transmit”. TED Talks, although maligned, feels less staid, but the form is much the same from an online perspective (the live experience may be different). Discussion forums underneath videos are not much different to comments under news article, and we all know what’s said about those.

Teaching online is an increasingly prevalent mode of educating that merits PhD students’ and ECRs’ attention. What is expected of us now as teachers, especially in cases where traditional teaching opportunities might be closed to us because of the changing HE landscape? I don’t have an answer to this, but I do have some more questions:

What are some of the ways that you have seen effective teaching being done outside traditional environments, in particular online? What sort of distinction is there now between online scholarship and online pedagogy? Is the act of making scholarship freely available and accessible—to those who can engage with it—an act of pedagogy? Are attempts at online pedagogy—particularly those that are fairly low-engagement for the teacher, as MOOCs often are—inevitably less effective than classroom-based teaching? Will moves towards more online or open-access teaching detached from ‘real’ (classroom) teaching only exacerbate an attitude in academia of treating PhDs and ECRs as contingent labourers by replacing more of our traditional work? Would promoting one’s ability to teach through, for example, one-off Prezis/YouTube videos or Udemy courses actually be counterproductive for a job candidate?

 

Accidental blogging: some thoughts on academia.edu’s sessions format

I’m a relatively infrequent user of academia.edu (aren’t we all?), but I noticed recently the ‘sessions’ feature. And by “noticed recently”, I mean that I logged on one day to find that academia.edu was telling me I had an “expired session”, and I had no idea what the hell they were talking about, so I went away to find out!

It turns out that a session is a piece of writing that is opened up for comment and discussion. To open a session, one uploads a draft paper. One can ask for feedback from specific people in your network, but the draft is automatically open to comment by certain groups of followers (e.g. supervisors or mutual followers). I now recall that when I uploaded the paper in question, I did upload it as a ‘draft’ because it was a record of what I had presented at a conference, which was still a work in progress. I didn’t realise that labelling it ‘draft’ would throw it open in the way it did. 

With this particular paper, I don’t mind too much that confusion during the upload process meant it was accidentally opened up, but: 

  1. I don’t particularly like the way that academia.edu is trying to redefine collaborative processes as inherently a public one; and 
  2. I think it will only lead to the site becoming a partial blog for people workshopping general ideas.

Rather than, say, uploading your paper to Dropbox, or iCloud, or Google Docs, or just plain ol’ emailing it around to people whose insights would be valuable, this proposed model of academic collaboration is open by default. There are some pluses, perhaps. Your paper might receive attention from those in your extended network, or beyond, who can offer valuable insights that you might otherwise have lacked. 

However, the process fails to take into account the fact that publication is a key metric for academics hoping to progress in their careers. There are relatively few disciplines or cases where the model that academia.edu is offering is a desirable one for academics. This is not necessarily a good thing, but it does mean that the incentives to share so publicly early drafts of exciting and publishable work is pretty low. 

In addition, particularly for those who are on the very lowest rungs of the academic ladder, for whom publishing might be a hurdle they are seeking to jump, academia.edu muddies the water. By taking steps to link one’s rate of “producing quality content” to the metrics of how much pieces have been viewed and commented upon, it constructs a competitive blogging environment. 

In terms of how the feature will be used in practice, perhaps it remains to be seen, but what I have seen of the feature so far doesn’t make it a great method for collaborative working. One session that I joined, out of curiosity both about the form and the subject matter, was on The Study of English Literature, based on a short paper by John Xiros Cooper of the University of British Columbia about the future of the discipline. I did not actively seek the paper out; nor did Cooper seek me out to solicit my feedback. His session simply appeared in a list of open sessions related to my research interests, floating on the right-hand side of my home page. I requested access and was granted it. (Many thanks, John!)

To me, the paper was the sort of thing one might expect to find on a high-calibre academic blog as an extended post, and the comments were similar to those one might expect on such a post. The feature doesn’t allow for in-line editing and commenting, such as might be seen on scribophile.com or even in OneDrive, that old corporate lag. In general, the way that papers are presented is a bit undesirable; they rely on scribd, and there is always a delay in viewing/downloading a paper. If the purpose of the ‘draft paper’ function is actual collaboration, then it lacks the features necessary for nuanced comment and discussion.

Personally, if I wanted to throw open ideas for comment by anyone with an interest, I would use a blog post with moderated comments and tweet the link… 



Disclaimer: I should also say that this is a very separate issue to open access to academic literature, which is one of academia.edu’s main aims, and a laudable one. As part of its general mission to open up more and more content to be available for free on its site, academia.edu encourages academics, particularly doctoral students and early-career researchers, to publish PDFs of their work. In 2013, when Elsevier issued a take-down notice for numerous papers, the site was clear and unequivocal in criticising the move, as were many academics (Michael Clarke’s analysis of the incident is particularly worth a read).

Library of Babel: a Borgesian black comedy made real

Borges’ short story ‘The Library of Babel’ was one of my favourite pieces from my sophomore lit-crit tutorial, so I’m weirdly excited by the fact that someone has now devised a website that can make it a “reality”. Programmer and author, Jonathan Basile was stirred by the resonance between Borges’ description and the capabilities of modern technology and surprised to find the code not already written, so he set out to produce a digital version of the library, which re-mediates Borges’ library. Once an imagined place whose literary content could only be inferred by an imaginative reader, it is transformed into an online on-demand production of literary content abstracted from the physical sense of a library. 

On the home page, as soon as one hovers over a link, the ‘counter’ of letters begins to whirr, rendering the word one had chosen quickly gibberish, but also exactly the same (because the ‘About’ page will be the end result of a link, even when the link text itself has been transformed into ‘Abqkj’). The changing digital text oddly offers a more fixed link between the symbolic and the real than the printed text of a short story’s page might.

The LibraryofBabel.info ‘About’ page notes that the site does not, in fact, “contain” the full Library of Babel, but only a small proportion of it (all permutations of 3,200 characters, rather than 1,312,000 characters). The question becomes, however: If the necessary algorithms are already there (as in, they could be written), is the website, in fact, as complete as Borges’ textual library? The website might be said to “contain” the full Library, but render only part of it accessible, without compromising the Library’s integrity. Indeed, the website adopts this claim elsewhere, such as on the ‘Reference Hex’ page: “Borges has set the rule for the universe en abyme contained on our site” (emphasis mine). 


Not content with a digital re-mediation, the website offers helpful suggestions for how readers might engage with the website, including the use of a screen reader. The screen reader itself being programmed with grammatical rules would not offer the same content, however, but a re-mediated version distinct from the version that might result from my reading the text (to myself or to others, aloud or mentally). Engaging us in interpretative activities, LibraryofBabel.info playfully suggests, following Borges, that the website and its forums should be a place for conducting and archiving research: “We encourage those who find strange concatenations among the variations of letters to write about their discoveries in the forum, so future generations may benefit from their research.” These include searches such as “The Complete Works of William Shakespeare”, which the searcher records with “Doesn’t have a ‘Hamlet'” “but it does have a ‘hrlbikst’. Good enough for me!”, or searches for ASCII art, led by Basile himself. 

The forum also contains spontaneous outpourings such as, “I think it’s incredibly important that [the site] never be taken down”, or “So incredibly happy that you brought this into being”.* It is fascinating the sort of emotional response that the promise of comprehensiveness can generate, even when what is on offer is comprehensive gibberish. Borges’ blackly comedic take offers a thought-experiment, but LibraryofBabel.info claims to offer a fully fledged product, monumentalising the Library. Whereas Borges offers an exemplar of how human interpretation might constitute reality, Basile’s website enacts human interpretation in a guided form. “Anglishise” is an option on any page, highlighting English words and drawing (suggestive but meaningless) allusions from the overlapping of words. 

* Luke Chrisinger, 14 May 2015; Christian Arthur, 17 May 2015: http://libraryofbabel.info/forum/?topic=lob.

The Culture of Connectivity: A Critical History of Social Media (José Van Dijck) — book review


The Culture of Connectivity is a few years old now (the acknowledgments are signed June 2012), but it caught my eye as an interesting endeavour to chart critically the origins of some of the companies and platforms dominating our online experiences today, and probably for the foreseeable future. 


Overall, the book is a firmly theorised version of the many critiques written about how social behaviours can be (and are) co-opted by capitalism under the guise of ‘the sharing economy’, ‘social networking’, etc. (see, most recently, yesterday’s Guardian long-read by William Davies).
Interestingly, as I was writing this review, I caught sight of Wired’s reporting of yesterday’s Science, which published a study into the impact of Facebook’s algorithms on interactions between users from different political backgrounds. However, as Wired and others note (e.g. Zeynep Tufekci on Medium), the study’s methodology is not robust enough to allow us to reach firm conclusions about how technological decisions impact social experiences, although the study suggests both that Facebook’s algorithms decrease the amount of ideologically challenging content users see and that users limit their own exposure to such content via their active choices between content.
A few brief comments on the individual chapters themselves:
I joined Facebook in 2005 and have paid attention to some of the dust-ups over Terms of Service, the introduction of Timeline, etc., but it was still interesting to read an overview of how Facebook has repeatedly pushed users to stretch the norms of data sharing. Van Dijck summarises well the various criticisms over the company’s asymmetrical attitude towards transparency (whereby it has few obligations and users many), and I think her analysis is persuasive.
The book next explores how Twitter’s filtering and weighting algorithms can produce, as well as reflect, trends, and can create a circular pattern of popularity. Twitter is the main SNS that I use currently, and the site whose future choices will most directly impact my online experience (at least in the short-term, while I’m still a loyal user!).
The latter chapters I found mainly of academic interest, as they analyse sites that haven’t attracted me as an active user. Van Dijck explores Flickr’s difficulties establishing itself as a leader through indecisive leadership and a fractious relationship with the core user base, and she identifies how YouTube, ostensibly a ‘democratic’ UGC site, has been on a convergence course with PGC (professionally generated content) producers, tracing users’ own conversations about the lines between self-expression and marketisation.
The chapter on Wikipedia was surprisingly fascinating. Van Dijck teases out the nuances of Wikipedia’s user levels, including the increasing bureaucratisation of the site behind the scenes, with strict control over the content of the site exercised in a veiled layer of interactions below the familiar layout of which page. I have previously dallied with editing Wikipedia, but doing so is not particular intuitive for a newbie, and I have been genuinely anxious about incurring the wrath of other users for ‘not doing it right’, so I’ve not bothered.
From a personal perspective, I find looking at the impact of embedded, often invisible structures — including language (“friending”, “sharing”, etc.), algorithms, and “logged-in experiences” — in forming our online experiences is an intuitive approach to exploring this new social ecosystem in which many of us live. Her Foucauldian exploration of the “power of norms”, and how social media has changed online norms around privacy and sharing information, seems productive. She finds (unsurprisingly, although nonetheless usefully) that platforms are “mediators rather than intermediaries: they shape the performance of social acts instead of merely facilitating them”.
What is most interesting in the book is how it clearly debunks some of the myths about the utopic potential of social media, which elide the technological and human choices involved in constituting social practices online. Only a few days ago, for example, Accenture’s “Pulse of Media” argued that “consumers are now both kings and kingmakers, fully in control of which bundles, which brands and which content succeeds”, naively suggesting that the bundles being offered, the brands “promoted” and advertised through personalised ads, the content encouraged to “trend”, have no impact on consumer choices. Van Dijck goes some way to shining a spotlight on software which, as DM Berry (@BerryDM) has argued, is becoming increasingly ubiquitous and yet “also withdraws”. The research on hard data that is being undertaken by researchers with access to data held by the big SNSs, such as the Science article mentioned above, is only the beginning of the work that needs to be done.



Van Dijck‘s book offers chapters on Facebook, Twitter, Flickr, YouTube, and Wikipedia, as five of the major players in various digital niches (social networking sites (SNSs), user-generated-content (UGC) sites), although she steers clear of addressing either trading and marketing sites (TMSs) like Amazon or eBay, or play and games sites (PGSs) such as FarmVille or Angry Birds. Her individual analyses are interesting for anyone who hasn’t actively followed the development of these sites, but at times the chapters are dry. Organising the book this way makes sense, allowing Van Dijck to focus on chronological historical analysis, but I wondered several times whether a thematic structure (e.g. considering governance as a whole) might have been more productive. 






Around these five central chapters exploring the growth, development, governance and users of these platforms, Van Dijck offers a methodology for writing such histories, combining [academic nerd alert!] actor-network theory and political economy. The main aim is to draw together theories about the social and the technical in order to develop a theory of how they can be mutually constitutive. 



Despite finding the book informative and persuasive, I also found myself frustrated as a reader. In her search for a more objective historical narrative voice, Van Dijck doesn’t tackle the impact of the structures and strategies that she is charting. Although she offers the example of the Alvin family to describe different user reasons and strategies for engaging with social media, she tries to stay away from offering any value judgments. At various times, the book refers obliquely to privacy concerns as though the reader would be so thoroughly familiar with the arguments that they could be taken as read. This is probably true of her intended audience, but the book could benefit from a greater focus on the individual and the specificities of individual experiences, even if this might make the text more polemical.

At the risk of a hackneyed metaphor, there can be something Kafkaesque in the increasing technocracy of leading platforms, in the diminishing freedoms for the individual user who interacts online only through increasingly closely mediated channels. As platforms grow, commercialising or merely becoming more complicated, the rules for users interacting with them grow more detailed and more unfathomable. Wikipedia, for example, still open to (almost) all, is nevertheless more complex for a first-time user who wishes to modify it than it once was.