mouseover for <supplied reason="undefined"> in edition

[manufrancis]

The current mouseover for <supplied reason="undefined"> in edition is currently:
"Lost or omitted text".

EGD §5.5.1 currently says:

–	instead of "lost", you may use the value "undefined" for @reason if and only if [[**I add numbers**]]
–	(1) you are encoding your digital edition (or an apparatus reading) from a printed edition without access to the original inscription or a visual representation of it
–	(2) and the previous editor gives no indication whether the supplied text was omitted (for which see §6.2.4) or lost
–	(3) and you cannot make a reasonable guess as to which of these was the case when the previous editor did their work

I have used <supplied reason="undefined"> when encoding from a printed edition (e.g.: dhar[ma] in a SII edition), where conditions (1) and (2) are met. As for condition (3), my reasonable guess is that it is just "unclear".

@danbalogh : I guess I just then should use <unclear>, right?

Anyway, should not the mouseover for <supplied reason="undefined"> be something like
"lost, omitted or illegible text"
or
"text supplied for undefined reason (lost, omitted, illegible)"?

And should this be to
https://dharmalekha.info/editorial-conventions

[danbalogh]

Hmm, good questions, food for thought. I can't give a very good answer to the main question, but I think that if you are working from a previous edition without access to visual material, and the previous edition showed something as "supplied" (using markup different from their markup for "unclear"), then even if you think it's probably semi-legible in the original, the previous edition's notation should be preserved, so stick to <supplied reason="undefined"> rather than <unclear>. (Note that this also ties in with the ideas of "encoder" vs. "curator" - mechanically creating a digital version of the printed edition is "encoding", but overruling the published statement of the previous editor is "curation".)

On the secondary question, this goes back to the EpiDoc idea of forbidding <supplied reason="illegible"> (see the footnote in EGD §5.1), which I disagree with, but which the EpiDoc community are unwilling to change. Since we don't want to transgress EpiDoc conventions, our ontology simply doesn't include the concept of "text supplied because it is illegible", so after careful consideration I think the answer to the secondary question is no, "illegible" should not be included in the tooltip for <supplied reason="undefined">. I have no objection to revising the tooltip to "Text supplied for undefined reason (lost or omitted)" if this is preferable.

I would be happy if @arlogriffiths could let us know whether he agrees with me on the above two points, and what he thinks about changing the tooltip.

[arlogriffiths]

I agree with all of Dan's response and would vote in favor of "Text supplied for undefined reason (lost or omitted)".

[danbalogh]

Thanks. @michaelnmmeyer , could you change the tooltip for <supplied reason="undefined"> to "Text supplied for undefined reason (lost or omitted)"?

[manufrancis]

@danbalogh

You write
if you are working from a previous edition without access to visual material, and the previous edition showed something as "supplied" (using markup different from their markup for "unclear"), then even if you think it's probably semi-legible in the original, the previous edition's notation should be preserved, so stick to <supplied reason="undefined"> rather than <unclear>.

The point is that in SII (and in Ep. Ind. also if I am not wrong; see EGD appendix A), there is basically only 3 markups:
[abc] which appears to stand for unclear, most of the time, but could be also supplied reason=lost, illegible
[abc*] for supplied reason=omitted
(abc) for surplus

[danbalogh]

I repeat: there is no such thing as <supplied reason="illegible">. If text is supplied by the editor, and the original is not altogether gone, then it can only be encoded as <unclear>. This means that for SII's [abc], you would generally use <unclear>, unless you are quite sure that the piece of stone/copper is actually missing, in which case you encode it as <supplied reason="lost">. When in doubt, stick to unclear.

It seems to me that in practice, some authors use(d) [abc*] (or sometimes [*abc] as distinct from the former) for <supplied reason="lost">, and because of this, we may sometimes need <supplied reason="undefined"> - but if you are sure that SII does not, then you do not need to use <supplied reason="undefined">.

In EI generally, I'm pretty sure I've seen all the kinds of markup usage listed in EGD Appendix A.

[michaelnmmeyer]

@danbalogh OK for <supplied reason="undefined">