I am making a few e-texts of Sanskrit plays, and I am in need of a good reference system, especially for linking the pratīkas in the commentaries to the base text. I have noticed that there are a few systems in use. The numbering of aṅkas and other higher-level divisions is not so much of an issue, so the following remarks concern the numbering of verses, prose passages, and stage directions within an aṅka:

1a. Only the verses are numbered. The prose sections and stage directions are not numbered. (e.g. M. R. Kale's student editions, and the e-text of the Veṇīsaṃhāra prepared by Yves Codet at http://indology.info/etexts/archive/texts/veni/venisamhara.xml). The disadvantage is that it's impossible to refer precisely to passages that are not in verse.

1b. The verses provide the principal numbering system, and the prose sections and stage directions are numbered with reference to the preceding verse. This is the practice of the Würzburg Bhāsa project, which assigns every prose sentence and every stage direction a number (see, e.g., http://www.bhasa.indologie.uni-wuerzburg.de/AM_1.html). As an example, the first sentence (or stage direction) after verse 1 will be 1.1, the second will be 1.2, etc.

1c. The verses provide the principal numbering system, and the prose sections and stage directions are numbered, not in reference to the /preceding/ verse, but in reference to the verse that they "belong" with. This takes some interpretation, but an example can be seen in Muneo Tokunaga's e-text of Śākuntala (http://gretil.sub.uni-goettingen.de/gretil/1_sanskr/5_poetry/3_drama/ksakunxu.htm), where the first prose sentence /before/ verse 2 is marked as 2<1, the second is marked as 2<2, and so on, and the first prose sentence /after/ verse 2 is marked as 2>1, the second is 2>2, etc.

1d. Only the verses are numbered, but pratīkas in the commentary are referred to line-numbers of an edition (this is the case with François Grimal's edition of Harihara's commentary on the Mālatīmādhava, and presumably with other editions prepared using LaTeX). The disadvantage is that this reference system is based on the lineation of the printed edition.

2. The verses, stage directions, and continuous prose sections are all assigned a number. This was the practice of the Clay Sanskrit Library.

The texts on GRETIL use (1a), (1b), or (1c) depending on the source. As far as I can tell, (1a), (1b), (1c) and (1d) provide rough compatability across editions, since referring to verse-numbers is a relatively standardard practice, whereas (2) breaks this compatability.

I am asking whether the scholars on this list have any preferences or suggestions for a numbering system that is stable, roughly compatible with (de facto) standard reference practices, and media-independent. So far, I am inclined to (1b).