What a Slugline Is Made Of

A slugline usually carries three pieces of information on one line: whether the scene is interior or exterior, the location, and the time of day. The difficulty is that those three ideas do not have one universal literal form. INT., INTERIOR, and a Chinese term such as “内景” can serve the same semantic role while having different lengths, punctuation, and separators. For a parser, semantic equivalence is not literal equivalence.

What is relatively stable in a slugline is where information tends to sit, not one mandatory vocabulary.

The three parts are usually positional

A common shape puts interior/exterior at the beginning, location in the middle, and time of day at the end. INT. OFFICE - DAY can therefore be read as three segments. A Chinese heading may use a form such as 内景 控制室 - 日. English headings may also spell out INTERIOR or EXTERIOR rather than use abbreviations.

A person can understand those forms as equivalent in purpose. Software receives characters. INT. is short, INTERIOR is longer, a Chinese term does not use the same punctuation, and the boundary before time of day may be written with different dash characters or a colon. Once the literal shape changes, the same splitting rule can produce a different result.

When a script is imported, SlateX keeps the original scene-heading text as stored, and separately attempts to derive interior/exterior and time-of-day fields from that text. Writing a heading back into a script file re-composes it, so a round trip is not guaranteed to be character-identical. The stored slugline and the fields split from it are therefore different things; deriving fields does not mean the original line has been rewritten into a normalized form.

Why a fixed vocabulary is not enough

A slugline is human-authored free text. A fixed list of interior/exterior or time-of-day terms will encounter abbreviations, full words, other languages, combined forms, and production-specific habits that were not on the list.

SlateX does not use a keyword whitelist for this step. Its behavior is narrower: it splits by separators and length. The time portion is taken from a separator near the end of the line, while the opening portion is tested around spaces or periods and is subject to a length limit. The parser is responding to string shape rather than first proving that a word is a valid scene-heading term.

That is why 内景 控制室 - 日 and 内景：控制室 - 日 are not the same input to one splitting routine. The second form attaches the opening term with a full-width colon instead of a space. A full-word heading such as EXTERIOR. OFFICE - DAY can also exceed the opening-length condition. These are character-shape differences, not evidence that one language is inherently handled better than another.

A different separator can change the parse

Characters that look like a dash can be a plain hyphen -, an em dash —, or a full-width －. A colon may also be attached directly to the opening term. Readers can interpret these as similar visual separators, but string rules do not automatically treat them as interchangeable.

The current parser handles the shapes that are explicitly represented in its branches. A rule that covers a plain hyphen does not automatically cover an em dash or a full-width dash. A rule that takes a prefix before a space does not automatically cross a full-width colon.

“This looks like the same kind of slugline” is a human judgment; the parser still receives a specific sequence of characters.

Three extracted parts are not proof of three valid values

This is the hard boundary. SlateX can attempt to split one line into interior/exterior, location, and time of day, but successfully producing parts does not prove that those parts are valid terms. A number placed before the heading can become the opening prefix. A hyphen inside a location can cause a trailing piece to be treated like time of day.

The parsed fields should therefore be treated as an editing starting point. Interior/exterior and time of day need human correction when the split does not match the script, and the script supervisor should verify them against the original heading. Presets can make repeated entry practical, but the existence of a preset does not validate a particular parse.

The useful check is not simply “did the app fill three fields?” Ask instead whether the opening segment is really the interior/exterior designation, whether the middle still contains the complete location, and whether the trailing segment is actually time of day. When any of those disagree with the script, the script and human review remain the reference.

FAQ

Are INT., INTERIOR, and “内景” the same thing?

They can serve the same semantic role in a slugline, but they are not the same literal string. A parser that splits by separators and length is affected by character count, punctuation, and spacing, so these forms cannot be treated as one identical template.

Why not put every slugline term into a fixed dictionary?

Sluglines are free text. Abbreviations, full words, other languages, combined forms, and production-specific habits keep creating forms that a fixed vocabulary can miss. SlateX currently uses structural splitting rather than a keyword whitelist.

Why can an em dash or full-width dash change the result?

-, —, and － are different characters. A reader may see them as equivalent visual separators, but a rule that explicitly handles only certain characters does not automatically cover the others.

If the app splits out interior/exterior and time of day, are those values verified?

No. A successful split only means the string matched a parsing branch; it does not prove that the extracted terms are correct for that script. The script supervisor still needs to compare the result with the original slugline and correct it when necessary.

Axiom One LLC — SlateX. Figures current as of 24 September 2026.