1. Forum
  2. >
  3. Topic: Duolingo
  4. >
  5. [WIU] Weekly Incubator Update…


[WIU] Weekly Incubator Update: 2017, Week 28.

Note 1: Data retrieved on 2017/07/16 at 20:35 (UTC).

Note 2: If you're about to comment about High Valyrian being released while other courses are still not in the incubator, please DON'T do it here.
This WIU is to summarize the progess of courses currently in the Incubator, not to debate about what courses are or are not there. Moreover, the WIU is not done (and maybe even read) by Duo staff.

I. Phase 1 Courses

I.A. Transfers

  • Exits (Courses reaching BETA Phase): High Valyrian from English.
  • Entries (Courses entering the Incubator): None.

I.B. Completion Summary

To enlarge the image: Right-click on it then select "Open in a new tab".

(A) This week's value automatically generated from Duolingo's statistics.
(S) This week's value self-reported by the team of volunteers working on the course.

II. Latest News from Course Volunteers (All Phases)

Course "Korean from English" (Phase 1)

By Soedori on 2017/07/15


6 skills left to go

  • 1620/1754 Words

  • 107/107 Word Images

92% completed

Course "Hindi from English" (Phase 1)

By KananG1 on 2017/07/14 (full text here)

The game is afoot!

We are a team of 4 contributors now and have started working on the course. A good news: there's ...

Course "High Valyrian from English" (Phase 2)

By Dedalvs on 2017/07/13 (full text here)

Valyrio Māzitas

The High Valyrian for English Speakers course is here! A big thank you to our contributors Amoren...

Course "Esperanto from Portuguese" (Phase 1)

By librulino on 2017/07/07

Novos colaboradores

Nossa equipe cresceu!

WesleyAlcoforado e Tulioflores, sejam muito bem-vindos!

Agora, trabalharemos até o curso atingir a marca de 40%, para então chamarmos mais colaboradores. :)

III. Previous Weekly Incubator Updates

See this index of WIUs.

A few words about HV's apparent extremely fast completion from 44% to 100% in one week: as explained in our first WIU, we enforce the minimum "total words" target of 1500 in our calculation to reduce the impact of partial trees on it. The 44% was based on the 1500 lexemes target, while the "final" tree contains (far) fewer lexemes than that, which explains the apparent progress of 56%. Based on the (now known) final number of lexemes, the progress in fact went from 93% to 100% in one week, making it more realistic (about 50 lexemes completed in one week). And no, the teams typically do not share their targets with us (or even know them in advance, it seems).

July 16, 2017



Who else is excited for Korean and Czech?

I'm also happy team Hindi is starting up again. Namaste, all four of you guys. :)

And congrats to High Valyrian for entering Beta and being the second conlang Duolingo users can learn.

Keep on going, everyone!


@Hindi from English team:


there's a good chance that this course might be able to get access to the new character type challenges (similar to the Japanese course)

mean that, like for Japanese from English course, the Hindi from English course will be first (and only) available on iOS (then later in Android) but not on web?

"Not on web" of course until those new exercises will be implemented into the website (for which there is no ETA from staff...).


We can't really say anything for sure regarding that, for we do not know. But I think that once these exercises are implemented on the web for the Japanese course, it shouldn't be much difficult to do the same for our course.


But I think that once these exercises are implemented on the web for the Japanese course, it shouldn't be much difficult to do the same for our course.

This is for sure, but no one knows (probably not even staff) when it'll be implemented into website (could be next week like next year AFAIK). Hence my question. ;)

You may want to ask staff about it when you'll discuss with them the fact for Hindi from English course to use (or not) "the new character type challenges".

This in order for your team to be able to notify the community ASAP about it and, hence, to minimize the potential disappointment from web users who would think the course will be released for them but will not. Such disappointment existed for we users waiting for the JA course who saw it announced as "released" but, contrarily to all previous courses wasn't released first on website.


Sure, we will share whatever we can find out.


please have it on the web, I like the web better since it has discussions and people use it more.


Team of volunteers (to which KananG1 belongs) can't do anything about it.


Did you guys add a whole slew of new words to the course? [I've read enough of these to know that even things that look like drops are actually indicative of progress.]


You are right. We added some new words and removed some old words (in order to reorganize things). That is why there's an apparent drop in progress.


More languages leaving, no new languages entering.... It'll be cool if sometime soon we get surprised by a handful of languages entering incubator at once.


I wouldn't be surprised if that happened. The incubator is almost suspiciously quiet, which makes me think that there is stuff happening behind the scenes. (Sort of like how Japanese went from appearing in the incubator to being available at warp speed.)


I would be surprised. In my mind the most likely reasons are (in reverse order):

  • They have started to prepare the tools to teach Mandarin
  • They spend all their time trying to get Japanese to web
  • They aren't interested


Well that link was truly painful to read......

To be fair... As someone who really wants Arabic and has studied it a lot... If all they really care about is usage metrics then it is actually perfectly rational to not not do Arabic because when they do add Arabic TONS of people will sign up initially out of interest and I'd be willing to bet that a disproportionately high number of them will quit quickly because beyond the pretty script learning Arabic won't be what people expect it to be like.

Duo knows a lot of people are asking for Arabic now, but I predict that a majority of those people would quit and the Arabic usage will be lower than people expect, and Duo may be on to that.

But some people would really get so much out of studying Arabic on Duo and there would be a strong small core community of people that would benefit and engage so much and make the Arabic course worthwhile and very excellent indeed. And that's so much more important that Usage Metrics! So let's ditch the false idol of Usage Metrics and add Arabic Puhleeeeaaaase!!!!

Oh and if it's so hard to adapt duo to teach the Arabic script, then here's an idea: Don't bother teaching the script! Memrise is free and PERFECT for learning the script, and then it is terrible for everything else after that! So if teaching the script requires too much resources, slap a link to Memrise Arabic script on the beginning of the course and then let them come back and start the course as a course that starts from the beginning but under the assumption that you already know the script! That's basically how the Hebrew course is, they don't drill you on the letters but you can learn the letters on Memrise and come to Duo for words, sentences, reading, writing, and listening practice! If adapting duo to teach the Arabic alphabet is the real obstacle then don't bother trying to reinvent the wheel, that's already the only aspect of Arabic that is easy to learn WITHOUT Duolingo, it's the rest that people actually need Duolingo for. There is no quality free Arabic course anywhere else online, so the people who are actually serious about Arabic will have no choice but to flock here.

Because online Arabic resources are so rare beyond entry level, I think that the biggest opportunity for finding people who would use Duolingo Arabic would be Arabic university students seeking a supplement to their classes. I would have really benefited from that in my Uni days. I bet those students would flock to the Duo Arabic course and they certainly wouldn't need the alphabet to be taught on Duo.

And if Duo can make a course for Hebrew, then they can make one for Arabic. Grammar, script difference, usage metrics, none of those factors can deny that. We've had enough silly excuses on that regard...


So let's ditch the false idol of Usage Metrics

Yeah, I don't see that happening at Duolingo.


Haha of course, but we can dream, right? =D


I honestly wouldn't be surprised if keeping people engaged with Duolingo long enough to do the Arabic alphabet skills of a hoped-for Arabic course could actually yield more lessons completed / ad impressions registered and billed than the average engagement for many other courses.


Good point!

Either way... with or without teaching the alphabet, with or without high and consistent usage metrics, I really hope that Arabic enters the incubator soon...


Thx for this link @DonFiore.

One (light) hope, however, is that despite this message 6 months ago saying that adding Japanese to Duolingo (whatever the platform) wasn't a priority, it nevertheless has been added to Duo system(*). Hence the same could happen (or not...) to Chinese, who knows... ;)

(*) Yes, only web but that's not the point here as "only web" is due to some exercises type not yet existing on web while the "difficulty" to add a ne language is (mainly) about managing the script which ha been tackled.


Is there something about the system for Japanese (which is used for kanji) that would be insufficient for Chinese? I'm under the impression Chinese would actually be more straightforward since characters usually have only one reading.


Woah! I just noticed Jitengore isn't even a contributor anymore. They must have gotten serious given the previous lack of progress. Well I hope Jitengore is doing okay and I wish him well in his endeavors. I look forward to what this new Hindi team can do! I am still excited to see if the Korean course is able to tackle characters within the course the same way as Japanese does (Or similar). I asked about it before but the answer was a little bit cryptic lol. Regardless I am immensely excited for Korean all the same.


I just noticed Jitengore isn't even a contributor anymore.

They still are a contributor to the English from Hindi course (but indeed aren't anymore to the reverse one, for at least 2 weeks AFAIR).


My cousin's one of the contributors for that course wishing them luck!


Progress in Swedish from Russian AND Yiddish??? Somebody might have to go for the smelling salts!


Haha, I hadn't even realized Swedish for Russian speakers had made progress! I must have stopped reading the graph a while back, since there is often so (depressingly) little change there, although I always look for these updates just to have an overview.


Thank you for making this! =)


I can't believe I missed High Valyrian's hatching! Thanks for letting me know :D


whistle I'll be so swamped by languages that I don't know how to react. If Klingon takes a spurt anytime soon, I may actually just die of exhaustion. ^^


So Klingon is actually happening?

Learn a language in just 5 minutes a day. For free.