Do you wish Tinderbox were localized in French? German? Chinese?
If there were community support for this, together it could happen.
3 Likes
Also In Russian please.
I‘d rather not - not least because this will create confusion by complicating discussions on this forum. Menus and options will be named differently in country versions and re-translating them into English for the debate of issues on the forum will be error prone. That is at least the experience with other applications that have chosen this path.
Theoretically this may, of course, broaden the audience for Tinderbox to non-English speakers. But since it has a demanding learning curve, such users will need support from the forum community where business is conducted in English - which I assume will not be able to be changed due to lack of resources.
So I would conclude that the required resources will produce more utility for the Tinderbox community if they are put into further development of the app!
4 Likes
My concern is chiefly for users whose English is focused on reading proficiency. That may be solipsistic; I’ve recently had to learn sufficient French to read research sources I need. My hearing is not up to the task of acquiring a speaking knowledge, and I don’t need to speak as urgently as I needed to understand computing in postwar France.
On the whole, I’ve found that support in a foreign language via deepl.com is surprisingly practical. I haven’t done that with Tinderbox, but https://pizzaForUkraine.com is entirely dependent on automatic translation, and it uses a particularly difficult language pair.
In meeting academics at conferences and such over the years, I’ve certainly encountered researchers who managed to write good papers in English but who were greatly strained by dinner conversation. (I can think of several examples in French and in Chinese, though none in German.). This is mostly in computer science; early conferences in digital humanities assumed that, naturally, everyone spoke excellent French and English and could manage a talk or two in German if necessary. But all this is anecdotal and based on insufficient sampling.
1 Like
The points above make sense. I used to include a Google translate widget on all aTbRef pages, but dropped it when it was apparent that auto-translate struggles with the technical writing. AI, might improve this (how, magic?) but I’d not bet on that.
As @abusch points out, correctly IMO, the menus alone don’t—for the effort involved-actually move the needle much. Not because the request isn’t well meant, but even after lots of effort it might not help . A lot semantics are packed into the names of system attributes and I don’t think it is proposed to change those.
Perhaps what the community can work at is to both tighten use of (terminology (panels/panels/tabs/drawers) to improve consistency for same-type things and, separately, work at (human†) translating and contextualising key terminology.
If might also help to (human) translate some key Tinderbox concepts like inheritance, prototypes, etc.
†. There is often a misplaced assumption that software translation captures meaning. But true translation is not about merely doing vocabulary exchange.
3 Likes
I like the idea.
But I also understand Abusch’s concern. Originally coming from international IT, my working language was English. After my studies, my job changed and today I am on the state that I have a good reading and listening comprehension. At the Tinderbox meetups, I also understand most of the discussion about the context. Except on vacation, I rarely speak this language. For writing I use DeppL and thus the knowledge atrophies faster and faster.
For applications, I distinguish whether I use them in English or German. For example, with 3D printing I only use English, because the German translation hardly makes sense, since the English terms are established.
For applications that I use in German but the forum is in English, I have made a habit of setting the software into English beforehand in order to really use the correct names for contributions.
A German version of the help pages would be cool
1 Like
For context, the ‘Help pages’ are created from c.390 Tinderbox notes with c.126k words. IOW, this is the size of the translation task, per language. The Help files change, albeit slightly/not at all for minor releases. But, certainly each translation would need review for each release of vX or vX.y.
FWIW, aTbRef is more dynamic in terms of change as I edit it near daily (I’ve changed 2 notes already this morning). Most such changes are typo corrections, re-writing for clarity (e.g. following a user being confused about how a feature works) as well as adding cross-reference links. In context it is c.3,150 notes and c.443,600 words.
One practical starting point might be a translation of key terms, e.g. ‘note’, ‘container’, ‘agent’. Though an (English) monoglot when working in text of another language I find I can quickly learn the ‘shape’ of words in another language. If (assumption!) this holds in reverse having a reverse look-up, e.g. [term in English] → [word in my language] might help give non-English speakers better access to the writing available. It might also assist in reading wider resources (e.g. the PDFs in the Help menu).
I’m also mindful that a lot of important terms that occur in writing about Tinderbox, e.g. ‘note’ or ‘expression’ are common English words with a range of meanings. Regardless, they have important and specific meaning in the context of Tinderbox. Translating these seems useful, not least as a stepping stone for anyone doing wider translation, noting that technical writing seems less well auto-translated, e.g. via AI, than general writing (smaller training set?).