-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathinterests.html
149 lines (144 loc) · 13.1 KB
/
interests.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="stylesheet" href="interests.css">
<link rel="stylesheet" href="index-files/index.css">
<title>Interests | Sunny Ananthanarayan</title>
</head>
<body>
<h1>Interests</h1>
<!-- dropdowns for research and leisure interests -->
<details>
<summary class="collapsible">General interests</summary>
<div>
<p id="gen-interests">
<ul>
<li>Language documentation</li>
<ul>
<li>Community material building (dictionaries, storybooks, archives)</li>
</ul>
<li>Sociolinguistics</li>
<li>Language contact</li>
<li>Multilingualism</li>
<li>Computational linguistics (largely technology for language documentation)</li>
<li>Grammatical and phonological change</li>
</ul>
</p>
</div>
</details>
<details>
<summary class="collapsible">Ongoing projects</summary>
<div>
<p id="panara">
As of 2023, I am doing my PhD at the University of Washington, and will be woking on sociolinguistic documentation of the Panãra language. It's still 2023 as of this being written so it's a bit early to report on anything! Stay tuned, though!
</p>
</div>
<div>
<p id="flibl">
I am working on a <strong>software suite that enables smoother transfer of language materials from ELAN
to FLEx, and back out to ELAN</strong>. This was started by a need from child language acquisition
work in documentation contexts that would enable maintenance of metadata and data between the transfers.
There is still a dependence on FLEx for its morphological parsing power, but quite a bit gets lost in
import/export, including important information necessary to keep at all stages. This software, which
we're calling flibl, creates an interchange format between the very different XML structures of
FLExTexts and EAFs (the respective file formats of FLEx and ELAN texts) in JSON, allowing for more
flexibility in the data's usability.
</p>
<p id="chirila">
A major part of language work for me is making sure language is not separated from its users. This
includes including sociolinguistic methods in analysis, but also <strong>archiving and
outreach</strong>. At the <a href="http://pamanyungan.net" target="_blank"
rel="noopener noreferrer">Yale HistLing lab</a>, I am working on refactoring the Chirila
archive/database to be friendly and available to language community members who would like access to
materials, and creating interfaces to see things like wordlists and maps more easily. Along these lines,
we also have ongoing conversations and research about the nature of archiving and data sovereignty,
including the disparate natures of various language materials and communities.
</p>
</div>
</details>
<details>
<summary class="collapsible">Selected past projects</summary>
<div>
<p id="naduhup">
My driving research interest is the documentation of minoritized languages. I started on this track when I
began undergrad at The University of Texas at Austin, working with <a
href="https://liberalarts.utexas.edu/linguistics/faculty/ple92" target="_blank"
rel="noopener noreferrer">Pattie Epps</a> on <a href="http://en.wikipedia.org/wiki/Naduhup_languages"
target="_blank" rel="noopener noreferrer">Naduhup languages</a>. It was also in that lab where I began
working with technology to make reclamation-related processes more effective (where "reclamation" includes
documentation, pedagogy, description, material creation, and other such sets of tasks).</p>
<p id="bats">
Another thing I got involved with early on while at UT was the Bilingual Annotation Tasks Force (BATs),
working under <a href="https://liberalarts.utexas.edu/caaas/faculty/ajt95" target="_blank"
rel="noopener noreferrer">Jacqueline Toribio</a> and <a
href="https://liberalarts.utexas.edu/frenchitalian/faculty/bb25848" target="_blank"
rel="noopener noreferrer">Barbara Bullock</a>. The lab was focussed on computational accounts of
code-switching, but, being led by sociolinguists, we were very keen to keep an eye to language users as the
key component for any kind of analysis. This was a major point of exposure to interdisciplinarity between
sociolinguistics, multilingualism, and computation for me, all of which have continued to be attractive
topics to me.
</p>
<p id="matukar">
One of the most influential projects I have been on was one I got involved in while studying abroad in
Canberra, through the <a href="https://dynamicsoflanguage.edu.au/" target="_blank"
rel="noopener noreferrer">Centre of Excellence for the Dynamics of Language</a>, working with <a
href="https://barthlab.wordpress.com/" target="_blank" rel="noopener noreferrer">Danielle Barth</a>, <a
href="https://www.jamesgrama.com/" target="_blank" rel="noopener noreferrer">James Grama</a>, <a
href="https://researchers.anu.edu.au/researchers/gonzalez-ochoa-s" target="_blank"
rel="noopener noreferrer">Simón González</a>, and <a
href="https://researchers.anu.edu.au/researchers/travis-ce" target="_blank"
rel="noopener noreferrer">Catherine Travis</a>. It was a project driven by Barth as a sociolinguist and
documentary linguist. We expanded work on <a href="https://en.wikipedia.org/wiki/Matukar_Panau"
target="_blank" rel="noopener noreferrer">Matukar Panau</a>, applying a sociophonetic analysis to audio
materials and their transcriptions from fieldwork. I worked with the <a
href="https://montreal-forced-aligner.readthedocs.io/en/latest/" target="_blank"
rel="noopener noreferrer">Montreal Forced Aligner</a> to train a language model from the Matukar
material itself, and we designed a recursive model creation pipeline to align the transcriptions maximally
precisely to take vowel measurements. We took a generally variationist approach to the quantitative part of
the project, to find likely places of social stratification.
</p>
<!-- So it's interesting to think about "research interests," because the way I think about it is kind of funky? So there are methodologies I like (and frameworks for acquiring and using data from those, which is...methodology also). There's also features I'm interested in. There's also categories of typological intrigue I'm interested in. All of these end up being called "subfields" or whatever, but it's hard for me to justify that kind of framing. I guess you could say I'm interested in morphology and semantics and historical linguistics, but historical linguistics can deal specifically with morphological and semantic features, no? I haven't figured out how to word it yet.
I consider my chief interest in methodology to be at an intersection between language documentation, sociolinguistics, and computational linguistics. More specifically, though:
As for sociolinguistics, I am fascinated by linguistic variation, the methods developed for the analysis of variation are many but there is room to improve. The reason I consider my methodological interest to be at the intersection of documentation and sociolinguistics is because I find that the documentation of sociolinguistics, including of course variation, is severely lacking. There are many active efforts to improve this state of affairs, and I am excited to be a member of that movement.
My interest in computational linguistics is less about machine learning and more about the processes in linguistics that can be aided by use of computational methods. Almost all pipelines in linguistics involve some kind of computational methods, of course, so where do I fit? I go where I'm needed at a high level. That is, I want to work with others to build software that is easy to interface with as linguists with little experience in building computational tools. The semantic satiation of "computational" is hitting me hard right now, wow. Just as I see a poor state of affairs in sociolinguistic variation in documentation, I see an unnecessary loyalty to software that doesn't necessarily best suit us as fieldworkers and people working with data from the field. That said just as I see a pattern of improvement in documentation of sociolinguistic variation, I see more and more efforts to improve our options for tools in the field<span style="color: #f1f1f1;">, especially finding a way to remove FLEx from the picture</span>.
Features I find especially interesting these days involve TAM(EV), the labrynthine pathways of phonologization and grammaticalization (and how they come together in amazing ways!). I really like language though, is the thing. Like, dang! Language! Rules! Even if I don't nevessarily buy into all aspects of a framework doesn't mean I don't think it isn't cool. We can have some lovely hypothesis generation from just some sweet OT tableaux. Language is fun and I'm here to have a good time. -->
</div>
</details>
<details>
<summary class="collapsible">Leisure</summary>
<div>
<p id="wiki">
I love to edit Wikipedia and other sister Wikimedia Projects, and I run Edit-a-Thons every so often to
encourage this of others as well. You can find resources about getting into Wikipedia editing on the <a
href="./resources.html#wiki-header" target="_blank" rel="noopener noreferrer">Resources</a> page.
</p>
<p id="other">
Other things~
<ul>
<li>I like games! TTRPGs like DnD, card and board games like Bohnanza and Anomia and Betrayal at the House
on the Hill.</li>
<li>I like to dance! East coast swing (with all the bells and whistles of charleston, lindy hop, collegiate
shag, etc), Latin dance (when the machismo is low), and y'know what? Yeah I'll just boogie down for no
reason.</li>
<li>I like cooking! Pretty much any cuisine, and am always trying to find new recipes (but, despite being
Indian and Texan, I can't handle my cilantro, so that's a major lifestyle limitation).</li>
<li>I like cats? I don't have one at the moment but I do like them.</li>
<li>I speedcube slowly and crochet poorly but they're fun for me!</li>
</ul>
</p>
<!-- I try to do the best I can to learn what is learnable, and learn about the difficulties in learning things that are at the moment insurmountable. What the latter means is that I cannot quite approach learning about the Sentinelese language, considering that research is not possible there due to the priorities of Sentinelese people: it is important to respect that and it is not a situation under my control.
Learning is my greatest passion, and that means that I must understand things incorrectly or with less nuance than is optimal, until I can learn about something more deeply.
It is unreasonable to expect that someone new to a field learn everything with perfect nuance, after all.
I value the notion that people can change their minds. People grow and change, personality is as mutable as perspective.
The idea that what I put on this website or post about on Twitter or what have you is a perennial view or opinion...that, to me, is ludicrous. We've all made claims we no longer stand by, and if we find out that someone has indeed changed their stance, there's not much reason to attack them for having had a bad/wrong opinion before. There is perhaps reason to attack them for having had a lasting negative impact by promoting or doing something, because some things are impossible to live down because of their significance; that doesn't mean that the person in question hasn't changed their mind of their own accord, through a discussion, or through retrospect provided by any force, really. Examples can be somewhat extreme, so I don't know that I want to go in depth, but just believing that people <em>can</em> change? I think that's valuable.
And hey, I might change my mind on this opinion.
I prefer to promote actual discussion about things I find worth discussing: if something can be looked up, there is no need to argument over it! -->
</div>
</details>
<!-- <script src="interests.js"></script> -->
<script src="./index-files/index.js"></script>
</body>
</html>