Prepare for the RDR by training decisions, not just strokes. Every situation on the machine requires a choice—brief form, phrase, fingerspelling, or full strokes—and every session includes professional-practice calls, such as how to respond when counsel asks for a transcript change. The actionable core of this guide: run one dictionary audit per week on a practice take, classify every untranslate by cause, rebuild only the categories that cluster, and rehearse two- or three-sentence scripts for the ethics situations most likely to arrive casually and under time pressure.
Writing for the record and writing for realtime are different skills
Realtime writing demands strokes that translate cleanly on the first pass: defined word boundaries, built-in punctuation, and entries your dictionary resolves instantly, whereas record-only writing can tolerate cleanup after the fact.
Realtime writing is judged by what the dictionary does with your strokes immediately, not by how much you could fix later. A stroke sequence with no dictionary entry, a conflict, or a boundary error becomes visible to anyone watching the feed. That shifts the emphasis from raw speed toward building a dictionary and a writing style in which every common expression has a defined, collision-free path. Cleanup-dependent habits—leaning on scoping, audio, or context to repair mistranslates—leave errors on screen during exactly the sessions where a live feed is running.
Worked scenario: in a realtime deposition the witness says "the defendant's insurance carrier." A record-only writer might dash through it and rely on scoping; the feed shows "defendants insurance carrier," missing the possessive, and counsel reading along loses confidence. The better decision is a pre-defined multi-stroke entry resolving to the full phrase with punctuation built in, created in the job dictionary before the session. This matters because realtime feeds are judged live: correcting the file afterward does not repair what opposing counsel already saw on screen.
Choosing between a brief form, a phrase, and fingerspelling
Match each writing situation to a tool: brief forms and phrases for high-frequency expressions, fingerspelling for unique names, case-specific job dictionary entries for repeated technical terms, and full separate strokes when ambiguity risk is highest.
The decision is about ambiguity and frequency, not habit. A brief form saves a stroke only if it cannot collide with another word in the same dictionary context. Phrases shine for expressions that arrive as units, such as "prior to" or "in response to." Fingerspelling is the tool for unique names and spellings because it produces the letters you intend rather than hoping an entry exists. Frequency justifies spending dictionary maintenance time; a word appearing once often does not.
Trace this example: you encounter "capias" once on a criminal calendar. Defining an elaborate brief form risks colliding with a stroke you already use under pressure, and a single appearance rarely justifies that maintenance. Writing it through a defined fingerspelling escape produces the exact word immediately. By contrast, "your honor" appears many times across a career and earns a locked-in brief form. Practice sorting words this way during practice takes so the classification becomes automatic rather than an in-session deliberation.
| Writing situation | Preferred approach | Why it holds up | Slip to watch for |
|---|---|---|---|
| High-frequency expressions ("prior to," "in response to") | Brief form or phrase | One defined sequence prevents stacking and boundary errors | Colliding with an existing dictionary entry under speed |
| Unique proper names | Fingerspelling with a defined escape | Produces the exact spelling on the first pass | Fingerspelling resolving into an unintended word |
| Technical terms repeated within one case | Case-specific job dictionary entry | Defined once, translates consistently for the whole job | Leaving the case entry in your personal dictionary afterward |
| Ordinary low-frequency words | Separate full strokes | Least ambiguity and easiest to scope accurately | Chopping strokes for speed and stacking the outlines |
Numbers, dates, and addresses that mistranslate under time pressure
Numbers fail because number-bar combinations mistranslate under time pressure. Build habits for figure versus written-out style, distinguish confusable pairs such as fourteen and forty, and verify high-stakes figures against the live feed.
Court reporting style distinguishes numbers that should appear as figures from those written out, and the machine choices differ accordingly. Date formats, street addresses, and currency amounts each carry their own conventions, and the reporter's job is to choose the format appropriate to the context rather than translate every number uniformly. Confusable pairs—fourteen versus forty, six versus sixteen—are where deliberate single-stroke habits earn their keep, because a mistranslated figure inside a contract term or a sentence is very difficult for a reader to detect.
Worked scenario: a witness states "one hundred forty-two thousand dollars." A rushed number-bar combination translates as "one hundred for two thousand," which is wrong but reads plausibly, so nobody flags it. The better decision is to slow one beat, check the figure as it renders in the feed, or write it in a deliberately unambiguous way, confirming before the next question. This matters because financial figures drive outcomes in depositions, and a plausible-looking mistranslation is far more dangerous than an obvious untranslate that a proofreader will catch.
Speaker identification and colloquy formatting that scoping cannot fix
Colloquy, question-and-answer format, speaker identification, and parentheticals each play distinct roles. Set up speaker entries before proceedings begin, and decide in advance how you will handle cross-talk and unintelligible passages.
Question-and-answer format works when the exchange is one examining attorney and one witness; colloquy—entries such as THE COURT or named counsel—takes over when proceedings become multi-party. Parentheticals record non-verbal events: exhibits marked, objections, recesses, who is present. These are structural choices rather than formatting polish, because the reader reconstructs the proceeding from them. Converting a mis-formatted colloquy after the fact is slow and error-prone, which is why the choice belongs at the moment of writing.
A practical application: before the session, build the appearance list and speaker entries so counsel names translate with correct capitalization and titles from the first mention. When two people speak at once, stop the flow, ask for the answer to be repeated, and note the interruption on the record rather than guessing at attribution. An untranslate is recoverable; a wrong speaker attribution can misassign testimony, and the reader has no way of knowing it happened.
Ethics traps in transcript requests and conflicts of interest
Ethics questions turn on impartiality, confidentiality, and the integrity of the record. You may clarify logistics and note uncertainty, but you do not change testimony meaning, give legal advice, or accept assignments with unresolved conflicts.
The reporter is a neutral officer of the record. That status creates specific duties: keep proceedings and transcripts confidential, decline engagements where a conflict exists, refuse requests that would alter testimony, and avoid even the appearance of favoring one side. Corrections belong to the witness and counsel through established errata procedure, not to informal edits by the reporter. Knowing the boundary is one skill; rehearsing the response is another, because these requests tend to arrive casually, mid-job, and under time pressure.
Worked scenario: near the end of a session, counsel says the witness's answer was "obviously" a misstatement and asks you to fix it in the transcript. The tempting mistake is to agree, because it sounds like ordinary cleanup. The better decision is to decline to change meaning, offer to reflect the exchange exactly as it occurred, and remind counsel that corrections run through the witness's errata process. This matters because your certificate attests that the transcript is true and complete, and agreeing would make you a party to the alteration.
Transcript pages built wrong: certificate, exhibits, and appearances
The certificate, appearance page, and exhibit record are structural transcript components with requirements of their own. Consistency problems—names spelled differently across pages—are the failures to check for before finalizing any transcript.
An appearance page records who was present, how each person is identified, and caption details; the certificate page is the reporter's signed attestation; the index and exhibit list make the transcript navigable. Each has its own conventions, and each is a place where small inconsistencies surface: a name hyphenated on the caption, spelled without it in the body, and titled differently on the appearance page. Treat these pages as a separate proofing pass with its own checklist rather than an afterthought to the testimony.
Build a cross-check habit: pull every distinct spelling of each proper name and party from a practice transcript and confirm the caption, appearance page, body, and exhibit descriptions all match. Do the same for dates and for exhibit numbers referenced in testimony. The task is mechanical, but it trains the eye for the exact class of defect that survives normal proofreading, because proofreaders read for language and tend to skip structural pages written in formula language.
A preparation sequence and a dictionary audit you can repeat
Alternate transcription-quality weeks with professional-practice weeks, and run a repeatable dictionary audit after practice takes. Track untranslate categories over time and treat improvements as learning milestones, not predictions of any exam result.
A realistic, adaptable sequence: begin with baseline practice takes in your weakest material—multi-party colloquy, technical testimony, rapid numbers—and log results without judgment. Spend the next block rebuilding the specific dictionary and writing decisions those takes exposed, then add realtime feed sessions where someone else reads your output. Rotate in professional-practice study: work through ethics scenarios on paper and compare your scripted responses with the principles of impartiality, confidentiality, and record integrity. Adjust block lengths to your schedule; the rotation matters more than any calendar.
The dictionary audit exercise: write a ten-minute practice take, then list every untranslate and classify it as a missing dictionary entry, a misstroke, a conflict with an existing entry, or a fingerspelling error. Expected observation: errors cluster in one or two categories rather than spreading evenly, which tells you exactly what to rebuild. Self-check rubric: category counts trend downward across successive takes; a fixed untranslate does not recur; your defined phrases and names translate with correct boundaries on the first pass. These are learning milestones only, not passing predictions.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.