The patient's words are not the book's words

A patient says her husband is bossy at home. The rubric is MIND > DICTATORIAL. A mother says her son was late to teethe; the book files it under DENTITION. The books say coryza where the patient says cold, loquacity where the family says he won't stop talking, and Kent indexes a craving under desires — search the corpus for "craving" and you find a handful of rubrics; search "desires" and you find hundreds, because that is the word the authors used.

None of this is a defect in the books. They are precise instruments written in the clinical English of their century, and their structure is why repertorisation works at all. But it means every case involves a translation: from the words the patient actually said — which are the evidence — into the words the index was built from. That translation is learned over years, and it is one of the genuinely hard-won skills of the profession.

The arithmetic, in practitioners' own words

Indian practitioners have argued this out in public, and the thread is worth reading in full — it is a rare thing, senior homoeopaths disagreeing on the record about their own workflow. Dr Gaurang Dave puts the load plainly:

"Suppose we see 100pt per day & repertrisation require 10 minutes. Then 1000 minutes."

Dr Pritesh Patel, in the same thread, on what the manual version cost a generation of prescribers:

"In the earlier days we had to Repertorise manually on a Repertory Sheet … and it was really tedious … hence most of us disliked it or avoided it … But, now with the laptops and Hom Softwares … we have a very versatile tool."

And the counter-argument, which any honest article has to print, from Dr Sameer Upadhayay:

"I see almost 100 cases in a day and repertories almost all the cases, that doesn't take much time if you know what to find and from where."

He is right, and the point deserves to be conceded rather than argued around: for a practitioner who knows the materia medica cold, the translation is fast because it happens in their head. No tool supplies that knowledge, and a tool sold as a substitute for it deserves the scepticism it gets.

The clerical half and the doctor's half

So split the work honestly. Repertorisation has two halves that are usually talked about as one:

• The doctor's half: deciding what in this case is characteristic, weighing which symptoms deserve to stand in the totality, judging the result against the materia medica and the patient in the chair. This is the intellectual work of the method. No software does it, and none should claim to.
• The clerical half: finding the rubric whose wording matches what the patient said, turning pages across several books because each indexes differently, listing what was taken, and carrying the chart forward to the next visit. This is transcription and lookup — necessary, skill-flavoured, but clerical.

The honest job of software is the second half only: do the finding, listing and carrying-forward, so that the doctor's knowledge is spent on the part only a doctor can do. Not because the doctor is slow — but because ten minutes of page-turning per case is ten minutes of a trained mind doing index work.

What a translation layer looks like from the inside

We build this layer in Homoeoclinics, so here is what the problem looks like in practice, from our own engineering rather than anyone's marketing:

• The vocabulary gap is real and one-directional. "Bossy" shares no letters with DICTATORIAL; "late to teethe" shares none with DENTITION. A search engine that only matches the letters typed will simply never find these — so the layer keeps a curated glossary that maps the doctor's phrasing to the book's, entry by entry, checked against the corpus before it is added.
• Even the books disagree with each other. Kent, Boger and the Boger–Boenninghausen repertory abbreviate aggravation and amelioration as agg. and amel.; Knerr, the Therapeutic Pocket Book and Boericke spell them out. A search that insists on either form silently loses the other books, so direction words have to be handled structurally, not typed into the query.
• Indian case notes flatter the problem slightly. Dr Ajit Kulkarni observes that "Indians narrate their complaints as they are" — concrete, objective description rather than elaborate metaphor. Concrete language translates better. But "better" is not "solved": the gap between everyday speech and a nineteenth-century index does not close by itself.

In the product, this means you write the case the way you already write it — c/o, k/c/o, < and >, one finding per line — and what you wrote is matched to rubrics across a rich set of classical repertories. The matching happens in our own software on our own servers; case text is never sent to a third-party AI service.

The list is checked, never obeyed

One boundary holds all of this together. What comes back from the translation layer is a list of rubrics — shown above the remedies, searchable, editable, each one droppable — and what comes back below them is clinical reference for the practitioner to verify before prescribing. The software proposes wordings; the doctor decides the totality. A hard case is hard because of what it demands from the doctor's half of the work, and nothing about a better index changes that. What a better index changes is how much of the clerical half is left lying on the doctor's desk.