Latest Twitter Threads by @c_borstell on Thread Reader App

Apr 29, 2020 • 4 tweets • 2 min read

Last night I was playing a little with Openpose data in #RStats. I realized it's not too hard to wrangle the Openpose output and plot signing directly using #ggplot2 and #gganimate, like so:

But I decided to make some tweaks so you can change the color of the signer+clothes, which makes seeing the hands a bit easier (contrast!)...

Apr 23, 2020 • 9 tweets • 4 min read

You guys know that IKEA products are basically just #Swedish words and place names, right? Walking around an IKEA store is like walking through a dictionary.

This is a script simulating the idea in Swedish and other places/languages: github.com/borstell/fakea

🏴󠁧󠁢󠁷󠁬󠁳󠁿

Apr 10, 2020 • 4 tweets • 2 min read

And version 1.1.0 already up!
> devtools::install_github("borstell/signglossR")

New stuff:
– make video examples (repeated and/or slow motion playback)!
– make GIFs!
– automatic face blurring/censoring!

Feedback welcome!

github.com/borstell/signg…

https://twitter.com/c_borstell/status/1248093992136384512

So you can now input a video and it outputs it slower and/or repeated. Here's an example of a sign for 'deaf' in STS rendered with a repeated 30% speed playback!

(Oh, and passed to the make_gif() function as well!)

Apr 5, 2020 • 4 tweets • 3 min read

So, it's like *very* easy to process and reconstruct actual images with only a few lines of code. As in plotting software redrawing the image, pixel by pixel.

Here's is a gif of me made with #ggplot2 and #gganimate. Sunday = fun day!

Like this is also #ggplot2 omg it's amazing!

Mar 27, 2020 • 5 tweets • 2 min read

#LinguistTwitter: What is your current level of proficiency for common notational systems in linguistic subfields?

(Polls below)

#linguistics International Phonetic Alphabet (IPA): internationalphoneticassociation.org/sites/default/…

Mar 17, 2020 • 4 tweets • 1 min read

So in Swedish, some island nations/territories would mostly get the preposition "på" ('on') rather than "i" ('in').

I mapped a blog corpus search to see the relative distribution between på (red) and i (yellow).

Basically, "på" is preferred for the following:
🇩🇲🇫🇰🇫🇴🇬🇬🇬🇵🇬🇺🇰🇾🇸🇨🇦🇽🇮🇸🇬🇱🇨🇺🇲🇻🇲🇹🇲🇬🇯🇲🇲🇺🇦🇼🇯🇪🇫🇯🇮🇪🇧🇧🇨🇾🇧🇸🇲🇶🇬🇩🇭🇹

🇦🇶 and 🇵🇭 are slightly in favor of 'in' (52% and 68%, respectively). Rest are definitely 'in'.

Nov 21, 2019 • 8 tweets • 4 min read

In class today we had read @ortega_ger (2017) on iconicity and sign acquisition. Students were asking how iconicity could be positive and negative for learning. Sometimes showing is better than telling, so I had them play a round of telephone across 12 generations (=students)... @ortega_ger First person was presented with 10 signs from a "new sign language": 5 iconic (e.g. PEN, AIRPLANE, TELEPHONE) and 5 arbitrary signs for colors. The iconic signs had been spiced with some marked phonological form detail (e.g. marked handshape), but a clear iconic mapping.

Nov 3, 2019 • 4 tweets • 2 min read

Here's a methodology finding for you:

I was running Openpose on some sign language data, estimating hand positions across frames in the single-sign videos (dictionary data). A reviewer asked to cut out transport movements, so I manually trimmed all videos to sign only... After re-running Openpose on the new trimmed video files (≈800), the results were completely different. Wow, did the transport movements really affect the results that much? Looking into the issue, the general patterns looked strange...

Oct 9, 2019 • 12 tweets • 2 min read

I'm in my "field of dreams" (lol) – #linguistics – and languages fill me with joy. But, I'm a very privileged person in all respects, also linguistically (native speaker of a European language with official status).

This is a thread of language "incidents" that made me sad... 1. Heritage language loss:
When I realized my grandparents' local languages ("dialects") were deliberately not passed on to my mom+aunts. They would want to speak it (me too), but cannot.

Sep 29, 2019 • 18 tweets • 11 min read

On the topic of #NoTalkTISLR started at #TISLR13 (a thread)... I'm a hearing linguist. "Why did you get into sign languages?" is a question I get. Answer: I wanted to do linguistics, went to Stockholm Uni, got exposed to Swedish SL at the dept, got curious, enrolled in a class, got hooked, and continued parallel linguistics + sign language.

Sep 28, 2019 • 9 tweets • 5 min read

Very excited about Hope Morgan et al.'s talk on phonological (wiggle-fingers) complexity and frequency distribution! Relevant to much of my own work! #TISLR13

Phonological complexity can be defined by 1) markedness (frequency/economy) and 2) structure (quantity).

E.g.:
Some handshapes are easier (1). Two-handed signs with simultaneous movements have more structure (2).
#TISLR13

Sep 26, 2019 • 7 tweets • 5 min read

Now exciting presentation by @AnneliesKusters, @jordanfenlonlx & @foundinblank on convergence in international sign! #TISLR13

Very relevant to current work at @SLL_Radboud!

International Sign is sometimes used in formal, more conventionalized settings (conferences, etc), but here the authors did a longitudinal study at Frontrunners, looking at language attitude and lexical convergence over time! #TISLR13

So relevant for emergence and creolists too!

Sep 26, 2019 • 4 tweets • 3 min read

@freya_watkins_ on comprehension of signs depending on angles. Especially L2 signers have a harder task mentally rotating signs perceived from an angle (as opposed to front facing).
#TISLR13

I'm wondering if mouthing plays a role. 👄👀

In a lexical naming task, scores for different angle representations were correlated with scores on a general mental rotation task. #TISLR13

Sep 20, 2019 • 4 tweets • 3 min read

@kristianali_ presenting work on name signs and tactile sign language in the Bay Islands at @SLL_Radboud Sign Pop-Up.
Many #signlanguages in the #Caribbean! 🤟🏝️

In Bay Islands Sign Language, tactile signers make use of both their own hands and the addressee's hands, as well as both their own body and the addressee's body to produce and perceive the language. Such interactive modality! #signlanguages

Jul 25, 2019 • 7 tweets • 3 min read

Ardavan Guity & @jahochcam (with a pre-recorded intro) on ethics concerning language documentation of #signlanguages in #Deaf communities. #wfd2019

A very interesting story here is Ardavan being involved in language documentation as an informant at first, but becoming more and more involved, gaining experience and skills along the way (by interaction, collaboration, sharing), ending up a researcher! #wfd2019

Jul 24, 2019 • 5 tweets • 4 min read

@AnneliesKusters on stage at #wfd2019 presenting on the @MobileDeaf team on International Sign (IS). IS is a phenomenon that comprises a lot of strategies in various contexts.

The level of conventionalization depends on context. Formal, conventionalized IS may be used at e.g. #WFD, but vocabulary and conventionalization varies with different events and contexts. #wfd2019

Jul 19, 2019 • 6 tweets • 3 min read

OK, so some points on posters in #academia:

Posters shouldn't be seen as a 2nd grade stage presentation. They are often seen this way, unfortunately, but some topics work better in a poster format. But conference organizers need to ensure posters are dealt with adequately. 1/6

Have dedicated poster sessions at the conference. Don't cram them into regular breaks, because 1) then other participants want to have a break rather than looking at posters, and 2) it means the poster presenters don't get breaks. 2/6

Jun 9, 2019 • 7 tweets • 3 min read

Is there any work on milk carton color #iconicity? I read a discussion online with Swedes disagreeing about the use of the word "rödmjölk" 'red milk' to mean '3% milk', because not all milk producers use the same color coding...

Most of them seem to use a ❤️💚💙💛 palette, but with different meanings. What I believe are the 5 largest milk companies in Sweden have (3%, 1.5%, 0.5%, 0.1%):
Arla: ❤️💚💙💛
Norrmejerier: 💚❤️💛❎
Skånemejerier: 🧡💙❤️💛
Milko: ❤️💚💙❎
Falköpings: ❤️💚💙💛

Jun 7, 2019 • 10 tweets • 6 min read

3rd keynote of the day: Deanna Gagne from @LingdeptGU on bimodal bilingualism. #LingCologne

Code-blending is the simultaneous productions of (aspects of) an utterance in sign and speech.

This is not the same as SimCom🚫, but something that can reveal the structure of and interaction between languages. #LingCologne

Jun 7, 2019 • 8 tweets • 6 min read

Keynote 2 of day 2 at #LingCologne: @ozyurek_a (@GestureSignLab) on the integration of hand gestures in spoken language as multimodal communication.

Traditional approaches to language research have focused on the:
— spoken/written (not visual)
— arbitrary (not iconic)
— discrete/categorical
— unichannel (not multimodal)

Luckily, more recent work has broadened the perspective wrt the above points.
#LingCologne

Jun 7, 2019 • 7 tweets • 5 min read

Day 2 of #LingCologne starts with Wendy Sandler giving her keynote on the compositionality of theatrical sign language.

Wendy acknowledges the important work of Svetlana Dachkovsky ("our resident squintologist") who has done extensive work on squinting in Israeli SL, for example as a marker of shared information and subordinate clauses. #LingCologne

Share this page!

Enter URL or ID to Unroll