Monday, October 19, 2015

Birthday research project: Yet another data point

Last year, as well as five years ago, I made a quick list of the media through which I receive birthday greetings. This year, I got yet another data point.

Let's see:

In 2015:
Twitter - 0
e-mail - 2
physical co-presence - 1 [My birthday was on Saturday, and I haven't been to the office since, which certainly skews statistics.]
telephone - 0
Xing private messages - 1
Skype - 0
Facebook private messages - 8
SMS - 1 (+ 1 spam mesage)
text messaging (not SMS) - 0
snail mail - 2

In 2014:
Twitter - 0
e-mail - 1
physical co-presence - 1
telephone - 0
Xing private messages - 1
Skype - 0
Facebook private messages - 10
SMS - 1 (+ 1 spam mesage)
text messaging (not SMS) - 1
snail mail - 1

In 2010:
Twitter - 6
e-mail - 2 (spam e-mails congratulating me were not counted)
physical co-presence - 1
telephone - 1
Xing private messages - 1 (spam PMs were non counted)
Skype - 1

Quick summary:
My observation from last year, that online services that share "birthday reminders" with people (Facebook, Xing) increase the number of birthday congratulations sent through them, still holds. While I have many more followers on Twitter than friends on Facebook, Facebook publishes my birthday and Twitter does not, so that I end up with birthday greetings on Facebook, and none on Twitter.
Mail - both snail mail and e-mail - seem to be a family domain, while the other media are less limited to one 'social group' only.
Beyond that, not that much change. Let's see how the birthday greeting ecology looks like in a year :)

Thursday, September 24, 2015

Yummy research - or: Multilingual CALL

It is a truth universally acknowledged in the realm of desserts that combining two yummy things creates something even yummier. This is why cookie dough ice cream exists. Or chocolate covered pretzels.
We have something quite similar in research. Sometimes it is the intersection between two topics that makes our spider sense resarch interest tingle - even more than each of them individually. Personally, I am fascinated by all things computer-assisted-language learning. But when I look at CALL, I do not want to limit myself to the exceptional case of monolingual students engaging in monolingual practices - or pretend-monolingual practices in the currently studied target language. Yet, when I present my work, I usually have to focus either on the multilingualism aspect, or on the CALL aspect, as there is little overlap between the research communities interested in each of these topics.
This is why I'm so happy that as part of my work on MElang-E, I am able to co-organize a conference on just this topic:  Multilingual CALL: Multilingual Language Learning with Digital Media in Primary and Secondary Classrooms. It will take place in Frankfurt, Germany, on February 17-18, 2016. The call for contribution is currently open (and will be open till November 16th). I am very much looking forward to many chocolate covered pretzels presentations that breach these two topics.

Wednesday, August 26, 2015

Who is the peer in peer feedback 2.0?

Peer feedback - may it be in language learning contexts or in any other setting where individuals practice writing skills - is popular, and not only because one could frame it as a means to relief the teacher from the burden of correcting texts (replacing this, instead, with the responsibility to support learners in giving and using feedback, of course). A lot of the research looks at what the person whose text has been feedbacked-on can gain from peer correction and feedback, but few teachers would use peer feedback if they didn't believe that the process of editing another person's text supports learning just as much as receiving that feedback yourself.
When you move this into the web2.0 sphere, though, things can change. In a classroom context, whether the classroom is offline or online, a peer is an equal, an specifically not a teacher. Peers may be able to see problems with your texts that you couldn't see, but they are not infallible (which, incidentally, seems to reduce the acceptance of peer feedback quite a lot, as MOOC organizers can attest). In one study (Pham and Usaha 2015) the instructional design went so far to have peer corrections corrected by the teacher before they were passed on to fellow learners.
When you look at one of the many web2.0 language learning communities, the peers are usually native speakers of the language you are learning. Yes, they do not have the formal position of a teacher, but often take up or are pushed into a kind of "native speaker as teacher" role. A native speaker who provides feedback on your target language text is in many ways as much of an authority, at least in matters of content, as a teacher is. Providing you with feedback does not help them improve their own writing in that language, but is more a courtesy, a way to reciprocate the help they received with their target language.

 I wonder how this impacts the uptake of peer feedback. Pham and Usaha interviewed learners about instances in which they used, and did not use, peer feedback to change their text. Indeed, students regularly refused to follow peer recommendations, for a wide range of reason, from the peer being considered wrong, to the learner not knowing how to make the suggested change. In a web2.0 community, it is easy to consider your 'native speaker buddy' infallible. Does this increase uptake? [I'm still number crunching, but, so far, the numbers of suggestions taken up are MUCH, MUCH lower than what Pham and Usaha found.] How do learners react when they receive feedback from multiple native speakers, and the feedback is contradictory? [I have, so far, found one example in my data set where this happened, and the learner expresses surprise and confusion about this fact.] How much of the work done on peer feedback applies to language learning communities?

Thursday, August 20, 2015

Language as puzzle

There are many ways to integrate language learning into a game. Personally, I prefer games that primarily tell a story, rather than focus on language teaching per se (something that is strongly reflected in my contributions to the MElang-E project, as well as in my own work on "Death comes to Hamsterley" (workign title), a Ren'Py-based game for young English language leaners).
As a linguist, though, rather than as expert for computer-assisted language learning, I also find games that treat languages and language learning as a puzzle very intriguing.
A good example of this (according to my resident Myst expert) is the Myst series. Language and its power to create select from possible worlds (Sapir-Whorf, anybody?) is a major theme across the different games. It is very rare that players actually need to understand something about a language to solve a puzzle and to continue with the game, but the games continually puzzle players with language, and encourage them to puzzle language out.
A more extreme example seems to be Sethian. It has not yet been released, but the kickstarter concept description is quite detailed:
In Sethian, you play as an archaeologist from thousands of years in the future, exploring a distant colony, which has been abandoned for centuries. In the ruins, you discover a functional computer, which operates in the native language. The core gameplay of Sethian consists in deciphering and studying this language, and using it to communicate with the computer, to find out what happened to the colony.
If anything, Sethian is a puzzle game, where the puzzles involve interpreting and utilizing the game's language in order to uncover the mysteries of the planet. But Sethian is not a game of decryption--the game uses its own language, unrelated to English, which must be translated by the player.
When we learn a new language - or use a language we are already well-advanced in - a lot of what we do is solving puzzles. I know this kanji, and I know that kanji, and I know the context the combination of them appears in - can I guess what the two of them mean together? I know this word, and I know that word, but together, they seem to form a proverb that should not be taken literally. What could it refer to instead? Potentially, this "puzzle" element of language learning could find its way into (language learning) computer games, too - though catering for different levels of preexisting language skill might be difficult, and the puzzles could end up much too easy or much too difficult for most learners.
But even fooling around with artificial languages in games such as Myst or Sethian might be helpful. I dabbled in Klingon as a teenager, and throughout my first years as English language major, my limited knowledge of Klingon helped me to understand linguistic concepts. Why not introduce learners to concepts such as agglutination, or different writing systems through a language-puzzle in a computer game?
And of course, there is one more aspect to it: Motivation and self-confidence. If you puzzle out the alian language while playing Sethian, you know that you have the potential to deal with complex language puzzles. How, hard, really, can French, Japanese or Latin - languages not nearly as 'alien' - be after that?

Monday, August 17, 2015

Native speaker in a box?

As part of my research on computer-assisted language learning - and as part of my own language learning habit, I admit - I try to keep on top of current language learning communities. Very often, what I find is that communities 'commodify' native speakers. A product is [insert positive adjective here] because it offers access to native speakers. Lang-8, for example, uses the slogan "Let our community of native speakers support your language learning." Livemocha promises "Registration, online language lessons, and access to native speakers for FREE."
Accordingly, many language learning networks are organized around speaker status - connecting language learners with native speakers, and discouraging other types of connections. As I argue in a forthcoming paper*:
"language and linguistic correctness are the social object of Lang-8, i.e. what “mediate(s) the ties between people”, or constitutes the “reason why people affiliate with each specific other and not just anyone” (Engeström, 2005): Two bloggers connect because one blogger is a native speaker of the language the other blogger is currently learning and can provide corrections, not because they share, e.g., the same hometown or hobby."
I wonder if this trend is now on the wain. At least one community seems to have a slightly altered approach. At Speaky, users not only indicate their language status as learners of some, and (native) speakers of other languages, but also their interests. And the list of interests users can chose from is quite detailed: I choose "Education, Linguistics, Social Medias , Language Learning, Video Games, Teaching, EdTech, Languages" from the provided list. And unlike other communities, where similar info can be provided in a user's profile as well, at Speaky, you can base your user search on these interests. When you search for other users, overlap of their interests with your interests is highlighted, so that it becomes easy to find somebody in the community who speaks French and is interested in the Internet of Things (see screenshot below). 
Another interesting element is that you can easily look for tandem partners among native speakers - and among non-native speakers. While the presentation of the website, like most others, focuses on "Meet Native Speakers from more than 180+ countries", the user interface does not reflect a sole focus on the native speaker as teacher. It is not only possible to include non-native speakers in one's search - it is possible to look specifically for non-native speakers of a specific level! 

Is this a conscious move away from language as social object? A reflection of ongoing debates about English as lingua franca? Or just a way to deal with a small user base during beta testing? And will these affordances impact actual user behavior? 

* Buendgens-Kosten, Judith (forthcoming) “Please check for grammar.”: Code-alternations in a Language Learning Blogging Community.  Journal of language contact.

Thursday, July 2, 2015

"Oh la la! Un Game de Computer pour le language learning!"

Games are a bit like movies: Most of the time, magically, everybody speaks the same language. If you are lucky, a short segment in an additional language is used for stylistic effects. Or a fantasy language is used as part of puzzles (cf. D'ni in the Myst series). But that's about it.
Choosing in what language to address a non-player character (NPC)? Code-switching mid-dialogue? Translanguaging? Things that are a fundamental part of the daily life of non-monolinguals are weirdly absent from most games.
At the MElang-E project, we are trying to create a language learning game that is, as far as feasible, sociolinguistically realistic. A game in which multilingualism does not merely become visible through 'decorative' accents and stereotypical phrase (see title of this blogpost), but is part of the gameplay.
This is easier said than done, though. How do you create an interface that does not just reflect WHAT the player character says, but also, IN WHICH LANGUAGE he/she says it? Without confusing the player with too many options, of course... How do you create branching structures for dialogues in multiple languages? Can you offer the same basic dialogue in multiple languages, just by translating it? Or do we have to assume that a dialogue would run a different course in another language, perhaps depending on the NPCs language skills and language attitudes? How do we deal with the fact that the player character may have much stronger receptive than productive skills? How do we signal NPCs' non-comprehension, and what consequence does NPCs' noncomprehension have for gameplay? Is there any chance to implement translanguaging in an adventure-style game?
A lot of tough questions that we tried to answer last weekend, during our second transnational project meeting in Barcelona. In the end, we might not be able to do all that we envision right now. But we will try! I can't wait to see the finished product!

Monday, May 25, 2015

Reading blogs in EFL - but which blogs?

In an earlier research project (forthcoming), I looked at how teachers introduce learners to blogging. I have been surprised to see that reading blogs often plays only a minor role. Nobody would ask students to write poetry or novels without them having experienced reading poetry and novels beforehand. By reading such texts, you develop a feeling for the expectations held concerning these types of texts. How long does a text have to be a novel? Is it OK to include dialogue in a novel? What kinds of stories do people tell in novels? Reading novels (on its own) does not make you a good novelist, but it prepares you for the expectations future readers might hold if presented with a text labeled as 'a novel'. This is even more helpful if you move from the level of 'novel' to the sub-level of 'detective novel' or 'romance'. The same, I'd argue, applies to blogs, and especially to blogging in EFL.
Of course, reading blogs is not only relevant when you plan to have students write blogs in class. In current English textbooks, publishers like to include 'modern' text types such as blog posts or emails. Yet, the imitation-blog posts and fake e-mails are sometimes so badly done, it is nearly hilarious (Saskia Kersten gave a fascinating talk about this on the last GAL conference). Adding a blog post or two - printed on paper, if need be - can give students a better idea of how these texts look like 'in the wild'. The best part about this is that blog posts - unlike novels - tend to be fairly short, and that if students like what they see, they can easily continue reading other posts from the same blog at home.
The questions remains: If we read blogs in class, which blogs? I have started a Google Docs file to collect suggestions. I will add my own discoveries, but would be very happy if others added their favorite blogs-for-EFL, too!

Click here for the list of blogs for EFL.

Life, Language and E-Verything

So Long, and Thanks for All the Ghoti.