How did you know that expected title is "Naga berjenggot"? There is no corresponding entry for this in id in wikidata - https://www.wikidata.org/wiki/Q743973

In T217029#4983892, @santhosh wrote:

How did you know that expected title is "Naga berjenggot"? There is no corresponding entry for this in id in wikidata - https://www.wikidata.org/wiki/Q743973

That's the title in the translated page, provided by Google translate. I don't know if it can be passed over reliably or an additional MT request may be required to get it from the Special page. Using the Wikidata label (if exists) can be considered, but that would be an extra that's probably not worth it if it adds complexity.

I would not recommend to get that title from machine translation. That is possible, but it is like scraping the html and grabing the text. Note that after MT, the page HTML is highly unpredictable since the MT engine modified it.

If there is a corresponding Wikidata label, there is a high chance that there is an article in the target language. That is one of the common ways an entity get label in different language.

I don't see an easy and reliable way to predict the target title.

In T217029#4983999, @santhosh wrote:

I don't see an easy and reliable way to predict the target title.

Is making an additional MT request to CX-cxserver not an option? The scenarios that we are covering are for languages with MT support, so MT should be available for those.

This generated confusion to several users during the user research conducted by @dchen.

@santhosh, considering my previous comment at T217029#4984068, and the recent work to find translated titles based on Wikidata and MT (T197787). Is there a simple and viable solution for this?