revision of EGC 5.3.2

[arlogriffiths]

Dear @michaelnmmeyer !

@zakariyaaminullah is editing a text on the basis of 3 manuscripts (J1 J2 J3), plus a previous edition (EdLC) which is known to have relied on one of the mss. (J1) that Zaki is himself collating. It often happens that EdLC makes an emendation (or conjecture, normalization) that @zakariyaaminullah wishes to adopt. Even if EdLC does not say that a given lemma is due to editorial intervention, @zakariyaaminullah can see that it is because he has accesss to J1. In cases like that, he wants to assign both @wit="#EdLC" and @type="emn" (or @type="conj", @type="norm") to the given <lem>. This is currently not allowed by our schema. See the discussion in the EGC gdoc at 5.3.2 and the list of errors at https://dharman.in/texts/tfd-nusantara-philology/e9b85c09f8e811a9e2ef115757b1731d32a23351/DHARMA_CritEdCandrakirana.

I prefer allowing the combination of @wit and @type on <lem>, rather than the solution with @resp or @source suggested in the discussion at EGC 5.3.2. Do you see any objection (possible ripple effects) from the point of view of the encoding model? If we do decide to allow this combination, then I will need to propose a rewrite of or an addition to the sentence "If the adopted reading is not directly supported by any of the witnesses, then you must apply to the <lem> an attribute @type." Some of what I wrote in the last paragraphs of the Preliminaries to the EGC may also need to be updated.

@aditiagunawan's Siksa Kandang Karesian edition confronted a similar situation, but dealt with it entire as per the current instructions in EGC.

[michaelnmmeyer]

I see no problem with this, so OK for relaxing the current rules.

[arlogriffiths]

I have to revise my point of view again.

I have just come across

^2875. pūruṣān] em. EdLC J2 J3, puruṣa J1

How can the lemma be an emendation it is is the actual reading of J2 and J3? I think @zakariyaaminullah may have meant that it was an emendation in EdLC's case because his single witness J1 had a different reading.

So for the example given in the comment in EGC 5.3.2, I now suggest

<lem wit="#edlc">pariśrānto ’pi</lem><witDetail wit="#edlc" type="emn"/><rdg>...</rdg>

to get a display like this

pariśrānto 'pi] EdLC (em.), ...

And in the example of app2875, I suggest the following encoding

<app><lem wit="#edlc #msjb #msjc">pūruṣān</lem><witDetail wit="#edlc" type="emn"/><rdg wit="#msja">puruṣa</rdg></app>

to get a display like this

^2875. pūruṣān] EdLC (em.) J2 J3, puruṣa J1

Do you think it is valid TEI, @michaelnmmeyer ? Can you easily implement the required changes? If not "easily", when do you think you could do so and can @zakariyaaminullah already now start revising his code as per my suggestion? If you approve of the encoding proposal, should I try to draft the necessary changes in EGC or can you take this upon yourself?

[arlogriffiths]

If you agree, @michaelnmmeyer, then as least I assume you could do some batch replacements for cases where a <lem> bears @wit and @type, like these:

1. <app><lem wit="#edlc" type="emn">nr̥̄ṅś</lem><rdg wit="#msja #msjb #msjc">naś</rdg></app>
   becomes
   <app><lem wit="#edlc">nr̥̄ṅś</lem><witDetail wit="#edlc" type="emn"/><rdg wit="#msja #msjb #msjc">naś</rdg></app>

2. <app><lem wit="#edlc" type="norm">smared</lem><rdg wit="#msja #msjc">smaren</rdg><rdg wit="#msjb">saren</rdg></app>
   becomes <app><lem wit="#edlc">smared</lem><witDetail wit="#edlc" type="norm"/><rdg wit="#msja #msjc">smaren</rdg><rdg wit="#msjb">saren</rdg></app>

[arlogriffiths]

@michaelnmmeyer : when you return to work, could you give at least a quick reaction to my recent messages in this thread? Thanks!