Wiktionary talk:Votes/pl-2018-12/Allowing attested romanizations of Sanskrit

Rationale
Attested romanizations in romanized text are in the spirit of WT:CFI: "A term should be included if it's likely that someone would run across it and want to know what it means." Furthermore, romanizations are often found in English text, and its readers may have a lookup need; many readers of English texts are unable to use Devanagari, and even with willingness to try, they would not know how to input that, and they would not find Devanagari in the source text anyway. Let us create the best experience for our readers that we know how. --Dan Polansky (talk) 11:10, 31 December 2018 (UTC)

Why attest?
There is no inherent value in ascertaining if a romanization has been used. I would understand it if one created just mechanically, that is by bot, all romanized forms, but attesting romanizations I do not understand, this siphons off the limited attention of editors. Fay Freak (talk) 01:04, 3 January 2019 (UTC)
 * Presumably due to the many different romanization schemes and ad-hoc systems in use. DTLHS (talk) 01:36, 3 January 2019 (UTC)

Claimed uses
Though I still oppose even adding romanizations automatically
 * saying you should learn write Devanāgarī (not true that even with willingness to try, users would not know how to input that, it is a contradiction)
 * or if you are a casual user you can still use the search function to find Sanskrit words via transcriptions
 * I do not believe anyhow much in the existence of users who do not learn Sanskrit but still search Sanskrit words, and I presume that even having the clutter of having the transcriptions charging the servers is not worth it. It is prodigal. Fay Freak (talk) 01:04, 3 January 2019 (UTC)


 * As for existence of certain users mentioned above, the vote page provides a link to a book that contains a lot of romanized Sanskrit and no Devanagari Sanskrit; readers of that book are those who existence the above post seems to deny. --Dan Polansky (talk) 07:57, 5 January 2019 (UTC)
 * Fascinating, no one's ever not believed in me before. I was reading about Hinduism recently, and did just that. A lot of romanized words used/mentioned in running English text on the subject are clearly not intended to be used as English (as hinted at by italicization, retension of macrons, etc.), and if they are not allowed in Wiktionary as Sanskrit, they sort of fall through the cracks. Andrew Sheedy (talk) 02:22, 7 January 2019 (UTC)
 * Why, Ugaritic texts are more often published in transcriptions than not, and the transcriptions are still not included in Wiktionary but only Ugaritic script entries. When is that magical point reached where a script employed is not only a transcription but a use? Ugaritic somehow being more dead and its users having never heard of Latin script won’t be the reason since the living of a language is somehow independent of its being used and the “living” of Sanskrit is also of dubious nature like the living of Latin. Fay Freak (talk) 02:34, 7 January 2019 (UTC)

Credit
The wording is a minor modification of Votes/pl-2014-07/Allowing well-attested romanizations of Sanskrit. --Dan Polansky (talk) 10:56, 31 December 2018 (UTC)

Status quo
As far as I know, we have no policy that forbids attested romanizations, nor do I know of consensus to forbid romanizations. I created the vote as "Allowing" instead of "Forbidding" to follow the tradition of a previous vote, but that choice should not indicate or imply any statement about the status quo. --Dan Polansky (talk) 10:58, 31 December 2018 (UTC)

Number of votes running at the same time
I think that's going to be too many votes at once. Per utramque cavernam 11:22, 31 December 2018 (UTC)
 * I don't think so. My heuristic is no more than 10 votes running concurrently, but that was not approved anywhere. Another magical number is no more than 7 votes at a time. Please check our voting history to see what our practice is, e.g. in the revision history of Votes/Active. --Dan Polansky (talk) 11:29, 31 December 2018 (UTC)
 * As it stands, when this vote starts, there will be 3 votes running in total, all created by me. --Dan Polansky (talk) 11:45, 31 December 2018 (UTC)

Test case: mahā
Let's assume a reader sees "mahā" in Sanskrit text that uses Latin script or in English text and wants to find out about it in Wiktionary. They enter mahā into the searchbox and press "Go", which takes them to mahā. Let's assume that mahā has no English section (there is one now, in RFD). Then, they only find Pali section in mahā. What now?

They may enter "mahā" again in the search box and this time press "Search" instead of "Go". They obtain a list of found hits; among them, the one they need to find is महा, at 15th place in my search results. How will they even know महा is the sought entry?

Another user may try to depend on the See also items at the top, sparing themselves entering the diacritics. They enter "maha", and then click on "mahā" at the top. Assuming again there is no English section (there is one now, in RFD), they are either stuck or they try their luck by pressing "Search" after typing "maha" into the search box, without the macron. That was a less fortunate idea: the sought महा entry does not appear on the first page of the results.

In both scenarios, having in entry mahā a Sanskrit section pointing to महा provides for vastly superior user experience, aligned with "A term should be included if it's likely that someone would run across it and want to know what it means", italics on "run across" mine.

--Dan Polansky (talk) 22:37, 2 January 2019 (UTC)

3 months?
Why is this vote 3 months long? I find that excessive. -- 08:33, 25 January 2019 (UTC)
 * It is because it took Votes/pl-2014-07/Allowing well-attested romanizations of Sanskrit over 5 months to collect enough input. I created the vote to last 3 months since there is opposition to vote extensions. I created and presented an extension mechanism that is result-agnostic, but I have not seen much support for it. --Dan Polansky (talk) 07:18, 26 January 2019 (UTC)
 * I consider unprecedentedly and unreasonably long voting periods, through extensions or otherwise, a manipulation of the voting process. -- 22:14, 5 February 2019 (UTC)