Approach
Three principles that run through my translation and digital curation work
I’m a translator and digital curator who works with Korea’s cultural heritage. On the translation side, I translate heritage site signage and museum exhibition text into English. On the curation side, I organize knowledge about Korean classics and cultural heritage as structured data and publish it on the web. I studied cultural informatics in my master’s and doctoral programs, during which time I also worked as a heritage translator. Over the past ten years, I’ve handled more than 6,000 heritage sign texts, and I’ve contributed, together with colleagues at the Academy of Korean Studies, to the Korea Heritage Service’s guidelines for heritage signage.
Translation and digital curation might look like two different jobs, but to me they’re not so different. In everything I do, I’m working as a translator across three divides: between Korean heritage and global audiences, between expert knowledge and public understanding, and between meaning made by humans and curated data that’s ready for AI to use. On every axis, the task is the same: to take what exists on one side and hand it to the other in a form the other side can actually use, whether that’s an international visitor standing in front of a sign, a fellow researcher entering records into a database, or a search engine and an AI model looking for reliable information. That’s why I approach both kinds of work with the same three principles in mind: accessibility, transparency, and sincerity.
Text Is Data: Two Sides of the Same Coin
The clearest place to see how translation and digital curation come together is in the importance I place on consistency: the same heritage site, the same historical figure, the same place, and the same term should be identified by the same English name everywhere they appear.
A text printed on a gallery wall doesn’t end there. Visitors search for the names they see on signs, and they ask AI about them. Institutions post the same text on their websites and enter it into databases. When the same heritage site is given a different English name at each institution, visitors can’t tell that it’s the same place, a search can’t connect the two, and AI’s answers about Korean heritage become that much less accurate. Asking what the most appropriate translation is for a given term is a natural part of any translator’s work, and I ask it too. But I give equal weight to a second question: is there a precedent for how this has been translated, and are we being consistent with it? The goal isn’t that everyone use exactly the same word in every context, but that we keep the number of competing versions to a minimum. That’s because I see a text not only as a text but as data, and to a computer, a name that differs by one letter is a different name.
This is why I follow the government’s standards first. In Korea, these cover the Romanization of Korean; the translation of public-facing terms such as place names, tourism content, and signage; and the official English names of designated heritage. They’re well made, and they’re a single standard everyone can follow together. Yet I’ve often found English texts, from designated heritage names to museum labels to promotional content, that don’t follow them, and this includes institutions not following the standards they themselves wrote. I don’t believe this comes from anyone’s carelessness. Staff at public institutions rotate between posts, translation is outsourced, English texts rarely get the kind of review that Korean texts do, and large projects are divided among many translators over many years. The result is that no one, at any point in the process, is responsible for checking the text against the standards. To take one example, the Korea Heritage Service’s own rules state that the name of an inscribed object takes the form “with Inscription of,” yet a bronze gong it designated in 2024 is registered under a name that doesn’t follow that form. If there had been a step where each new name was compared against the rules before registration, it would have been caught.
I can be that step, the person who still remembers the standards after the staff have moved on and who checks the text against them. But the larger point is this: we are well into the digital age, and now into the age of AI, yet the translation of Korean cultural heritage, and public translation in general, is still largely approached from an analog point of view, one text at a time, with each sign or label treated as a job in itself. Some texts really are only ever read on a wall. But the names they use shape how people think about what they’ve seen, what they search for afterward, and what the databases and AI models they turn to will find. Whether a text is printed or published online, it’s part of the same picture. That calls for a new point of view, and it means consistency matters more today than it ever has before.
Accessibility: To the People Who Need It, in a Form They Can Use
Knowledge should be delivered according to the needs of the person receiving it, not the convenience of the person who has it. What this means for translation is set out in detail in my Working Standards. In short, I start by asking who will read the text, I write in words that reader already knows, and I arrange the text so that the main point reaches them within the time they actually have.
Digital curation starts from the same concern, but with a bigger question behind it. The knowledge I work with is humanities knowledge, much of it produced with public funding, and the public should be able to reach it easily. Digital curation is often described in terms of preservation, but the other half of it is access. Digital resources open up new ways not only of reading what humanities research has found, but of taking part in producing it. That’s what draws me to digital curation: the question of how to share humanities findings in ways that are new, interesting, and actually useful. So for each project I ask who will create this data, who will use it, and whether those people can work with its structure.
For PoetryTalks.org, a multilingual linked open data archive of classical Korean poetry criticism (sihwa), I kept the ontology deliberately simple and used Page Forms for data entry. This means that a contributor who knows nothing about semantic data or the underlying ontology can still enter records just by filling in a form, as long as they know whether they’re dealing with a person, a place, or a poem. The structure takes care of the complexity of the data, so the people involved can concentrate on the content.
For my doctoral dissertation, I built a prototype of an English-language site on Korean cultural heritage along the same lines. The idea behind it was to link historical and cultural context to the everyday ways we encounter culture, through books, events, tours, and merchandise, and to make that context available in English to anyone who wants to introduce Korean heritage to visitors from abroad.
Transparency: Publishing the Standards, Showing the Construction
Transparency means two things in my work. The first is making it clear what standards I work by, and who did the work. The second is letting other people see how the result was built.
In Translation
My Working Standards are public. An institution can see in advance what I base my judgments on, and if we disagree about something during a project, we have a page we can look at together.
I also believe that publicly funded translation projects should make it known who translated the text, and which staff were responsible for reviewing it. These texts are made with public money, and many people trust and rely on them. A name attached to the work is what keeps its quality accountable, and it lets everyone see how carefully the review was done. This likely won’t be a popular position with translators or public servants, but I think it’s what citizens and taxpayers deserve, and I’m willing to hold myself to it first.
In Digital Curation
In digital curation, transparency means not hiding how the result was made. Poetry Talks and the dissertation prototype were both built not from scratch but on MediaWiki, Semantic MediaWiki, and existing extensions, with data that follows the general standards for linked open data. I use open-source tools wherever I can and keep proprietary components to a minimum. Some customization is always needed to fit a project’s particular needs, and I’ve written my own additions where I had to, but the more of a project rests on open tools, the lower the barrier for anyone who wants to emulate it. Nothing is locked behind a paywall, the tools come with a wealth of existing resources, and building on them rather than from the ground up makes it far easier to set up an infrastructure for sharing data. And because the structure is visible, others can take what I’ve built and adapt it to projects of their own.
There’s one more reason I use MediaWiki: every page keeps a record of who changed what, and when. The knowledge doesn’t sit behind a wall, and the process of making it is there for anyone to see. This is the spirit of openness that the digital humanities have held to from the beginning, and it’s the one I try to follow. I can’t claim that my own accessibility and transparency are perfect, but they’re the standard I judge my work by.
Sincerity: From Sincerity to Transformation
There’s a passage in the Doctrine of the Mean (Zhongyong) that has stayed with me since I first came across it:
曲能有誠，誠則形，形則著，著則明，明則動，動則變，變則化。唯天下至誠爲能化。
Even in small things one can be sincere. Sincerity takes form; form becomes manifest; what is manifest becomes bright; brightness moves; movement brings change; and change brings transformation. Only the utmost sincerity under heaven can transform.
What speaks to me in this passage is how attainable it is. The starting point is sincerity (seong, 誠), being wholly genuine in what you do, with nothing false between what you intend and what you actually do, and that is within anyone’s reach, in work of any size. From there the passage traces a chain, and each link in it is about something becoming visible: sincerity takes form, what takes form becomes manifest, what is manifest becomes bright. Sincere work shows itself, and because it shows itself, it can reach other people and move them. In that way, sincerity is already bound up with accessibility and transparency. The end point is transformation (hwa, 化), the world becoming a little different than it was.
I find this an accessible way to approach work, and life. You can’t control the outcome, but you can control your own sincerity. Rather than reaching for a perfect result, you hold to the starting point and trust that a good, even transformative, outcome will follow.
That’s the sincerity I try to bring to my work. It doesn’t mean I get everything right. Like anyone, I make mistakes, and sometimes I only learn a better way of doing something after the fact. What it does mean is that I bring the same care to a one-sentence sign as to a database of thousands of records, and that I align myself with the purpose of the work: the reader it’s for, and the knowledge it’s meant to open up, and for whom. The rest I leave to the chain the passage describes, in the belief that sincere work, accumulated over time, brings about a little transformation somewhere.
If this page resonates with you, or if you’ve been thinking about similar questions in your own work, please feel free to reach out through my contact page.
September 2026