Make sure you get accurate captions for your talk by following these instructions!
- Find your directory, if your directory is missing feel free to create one by copying another.
- Inside your directory is a
caption_wordlist.csvfile. - Edit
caption_wordlist.csvwith terms and their pronunciation, as well as any extra details that might help the captioners.- You can use libreoffice to edit this.
- Include words that the average person wouldn’t know. Anything spell-check wouldn’t understand by default is a good idea.
- Names of author’s and anybody who will be cited
- Names of tools
- Names of programming languages
- Technical terminology
- Feel free to add your slides or paper, we can pass these off to the captioner as well
- Submit a pull request to this repository with your updated CSV file
- If you don’t know how to submit a PR, feel free to make an issue on github
More detailed notes below :).
Real-time captioning for ICFP will be provided by human captioners. To ensure that your talk is captioned correctly it is important to provide a list of any uncommon words and technical terms that will be used in your talk ahead of time, as well as the pronunciation of these terms — this will allow the captioners to prepare ahead of time and add the terms to their personal dictionary, ensuring that they can quickly and accurately caption your talk.
A good rule of thumb would be anything that the average person wouldn’t know about and anything that spell-check wouldn’t know by default:
- Names of author’s and anybody who will be cited
- Names of tools
- Names of programming languages
- Technical terminology
The captioners are skilled professionals, but may not be aware of terms in our field, and may mishear them unless given context ahead of time.
Each talk should have a directory, and within that directory there is a CSV file. The most important fields in this CSV file are the term and pronunciation columns, please fill these in and submit a pull request to the repository.
The CSV file has the following format:
| Term | Pronunciation | Context / Example Usage | Notes | Tags |
|---|---|---|---|---|
| itree | eye-tree | shorthand for “interaction tree” | semantics | |
| interaction tree | interaction tree | semantics | ||
| semantics | semantics, theory | |||
| compiler | com-pile-er | A program which translates code in one programming language into another, often performing optimizations to make the code faster | compilers | |
| frontend | front end | compilers | ||
| backend | back end | compilers | ||
| vellvm | vel-um | Name of the project, it’s a formal semantics of LLVM IR | compilers, semantics, projects | |
| IR | I.R. | Shorthand for “intermediate representation”, a programming language | compilers | |
| LLVM | L.L.V.M. | A compiler infrastructure project that provides | compilers | |
| LLVM IR | L.L.V.M. I.R. | compilers | ||
| clang | clang, sometimes “see-lang” | compilers, projects | ||
| Coq | Like in peacock | Coq is a proof assistant / programming language that is used to implement vellvm, “we have implemented and proved this in Coq” | proof assistants, programming languages | |
| monad | moe-nad | “because itrees are a monad…” | theory | |
| monadic | moe-nad-ick | “we have a monadic interface…” | theory | |
| lambda | lam-duh | Used to refer to anonymous functions “we have a lambda function here” | theory | |
| lambda calculus | lam-duh calculus | theory |
The most important fields are the term and pronunciation columns, please fill these in!
In addition, it may be helpful to provide some example uses and context can also be helpful for understanding how the term might be used, and you can leave any other notes you have about the term in the notes column.
For pronunciation feel free to use IPA, or whatever works for you for conveying pronunciation.
The tags column is also optional, but adding some tags about the usage of the term can help us organize a larger glossary to provide to captioners for future conferences.
Uploading your slides / paper into your directory can also be helpful. This will give captioners access to your talk and they can have a backup copy to follow along to, and we may skim through the talks ahead of time for any extra keywords that might be worth adding.