Language Incubator Frequent Questions
A few hours ago, we released a new version of our Android app with six courses created in the Incubator: French for Spanish speakers, and English for Polish, Russian, Hungarian, Dutch, and Turkish speakers. This was a major accomplishment for the Incubator contributors of these courses and our mobile and incubator teams. Congratulations!
Because of the occasion, I thought it was time for a brief update, so I'll answer some frequent questions:
When will these courses be available on iOS? In a few days.
What kind of Arabic is the English for Arabic course in? MSA.
When will the reverse courses (from English) be started? In about 4 weeks.
Why is that taking so long? Because in order to do this, we need to build a whole infrastructure to let the mods fully edit their courses. That means giving them the tools to decide which words are in which lessons, which skills come before others, and which words are related to each other. For example, after a certain point in the tree Duolingo needs to know that you understand verb conjugation well enough that when it teaches you the first person form of a regular verb you can also understand the second person. In order to do this, our system needs to know that two different words are both forms of the same verb for any language that we teach. We also need speech synthesizers and voice recognizers. Our Incubator team is working late nights to be able to deliver this as soon as possible.
When are you adding the language I want to learn? (1) See previous question. (2) If you want the process to go faster, encourage anybody you know to apply to moderate courses. They can apply here: http://incubator.duolingo.com/apply
Are you working on a Windows Phone App? This is not related to the Incubator :)
I just wish he and the rest of the team had been more realistic about the new english to (insert language courses here) time frame. So accounting for that and adding in a month for the first course to be ready official eta should be April 2014 probably around Easter.
EDIT: Why all the downvotes? Luis told us by the end of January we would have another English to another language course. This was back in December.
Having had to manage multi-faceted projects which necessitate having many different contributors, and coordinating all of that to make a final product, I learned that actually hitting a time projection is the exception, not the rule. All things considered I think the speed at which this is getting done is really impressive, especially if the quality matches their existing offerings, and therefore this effort is deserving of praise. Your comment comes across as a complaint. Perhaps it isn't meant that way?
Yes, i would think no earlier than that. If we're given the opportunity to change the skills and the sentences we will invest some time in doing that. And believe me, it is for your benefit, because if we just revert the current course, it would be extremely hard to learn Russian from what we have there.
hey luis, firstly thanks for the update, I think you and the duo team are awesome. One thing which is particularly relevant to me is the tree structure, I've posted elsewhere on the forum that Ive hit a huge wall about halfway down the spanish tree and I've been having to literally guess answers or done a lesson so many times I've just memorised the wrong answers lol. Ive found I've been doing lessons and passing them but not understanding them, I feel its because the tree structure suddenly introduced a whole bunch of complicated concepts all at once, I felt this was a structure issue of the tree and as you mentioned mods will decide structure is there a tool or a way they are going to be guided to introduce new concepts smoothly? (I am not asking this because of complaining, but just because I felt there may be others with similar struggles) p.s I find the duo project really really inspiring, so thanks to you all. :)
I had the same problem of progressing too quickly in the tree and only having the lessons go to my short term memory. The best way to fix this is repeat the lessons and constantly do the strengthen skills lessons.
Don't rush down the tree without understanding things. You'll get points that way but you won't retain as much info.
Agreed! Besides maintenance in French, I am learning Spanish on Duolingo. To overcome the long-term failures of blitzing down the tree, I first redid my past Spanish lessons. Now I limit myself: 1 new lesson requires two strengthening drills. I think I am much more solid in the material I have covered to date in Spanish than I was before I began this tack.
I think I'm at about the same place on my tree, and here is what I have done to keep progressing
- started reading simple texts/watching disney movies in italian
- keep all my skills golden, and use the general practice option
- I bought a grammar book and started figuring things out. Clitics, reflexsive verbs and the use of avere/essere as helping verbs are subjects that need additional explanation.
- at some point in all learning you will hit a wall, then you can either give up, or push trough and improve.
I kinda saw it coming since day one, so I decided I was going to write every new concept on a notebook, and that's working. I think that language learning is about learning the culture that comes with it, so if you try too hard to memorize grammar and vocabulary without emotion, human contact or fun, you'll forget everything. I face Dualingo as a support, I know that the team makes their best to improve this website, but I believe that it is already ok (it is not that they should stop working, lol). I guess that the main part in the learning process is to be curious and explore, so try to have more fun! :D
Check out this site: http://www.hungarianreference.com/
I've been learning Hungarian for about 3 years now, and I used this website for a lot of it (especially the grammar). I wrote stuff down, indicating which part of the sentence went where in Hungarian, and this really helped. (For instance "Szükségem van egy tollra" = "I need a pen", literally "I have need to a pen"). Don't worry if it gets tough, just practice and you'll get there! Azt remélem, hogy ez neked segít, és sok szerencsét! I hope this helps you, and good luck! :)
start basic hungarian courses on Memrise in the meantime. Or a native speaker... well, I could teach my girlfriend basic Hungarian in 2 months, so getting to this level shouldn't be so complex, if you learn every day a little bit.
And another thing... we make dubbing to almost all of the movies shown in cinema and TV, so if you can access those, they can help you in learning too. Poles don't do dubbing sadly, just voice-over, which I despise, so the only thing I can do is to watch movies with Polish subs.
Slightly unrelated, but in addition to this, will sentence discussions be added to iOS in the next update? The app is great for practicing, but it can be difficult for learning without discussions (which usually provide excellent information and answers to a lot of questions).
Can't the beta be split even more granular? I.e. the whole tree doesn't have to be released in beta. It has taken me a long time to progress down the tree, so that releasing simply the first half of the tree in beta (with no second half) for new languages would add a massive amount of utility to people. It will also encourage moderators more than having much bigger goals for release with less feedback. A partial solution would also encourage more potential contributors to get involved... or at least I can see an argument for that.
The downside I could see to that is that contributers/moderators might get overwhelmed with feedback from the users in the first half of the course whilst simultaneously trying to write the second - can see a distinct advantage to separating into the distinct phases we have at the minute.
I'd imagine that the structure of the second half is strongly dependent on the structure of the first half, and due to that there would be changes to the first half going on as problems were discovered further downstream, right up until release. Which would make it confusing and counterproductive to do a partial release.
I have applied to moderate and / or contribute to multiple courses since I am native in Romanian and fluent in French and English (and Portuguese is starting to catch up thanks to Duolingo!) but all my applications were left unanswered. I had applied for contributing to English for Romanian in the first week of the incubator, and was ignored even after the course began developing. It has such a slow pace that I think I could have helped a lot if given the opportunity.
I used to be a frequent user of Duolingo but I grew a bit bored of it. First the good things in Duolingo. It is a good application that one can use freely. For me as a native speaker of Finnish there is not a problem to learn languages through other languages. I have several profiles everyone of which has a different teaching language. What I have experienced in several courses, not just in courses in which the teaching language has been English, has been that the lessons don't progress progressively. I mean that when there comes a new grammatical thing, it is concerned only in the particular lessons which teach that skill. The problem for me is that in many lessons that try to teach vocabulary the sentences are too simple. It would be nice to be able to practise skills progressively so that the more one learns grammatical things, the more one would also use those skills in the following lessons. To put it more clearly: If I am already an advanced learner, I would like to have also lessons that also include advanced sentence structures concerning grammar, not just vocabulary.
I know that the lessons won't most probably be changed in anyway. But why can't there be a possibility to jump over those dull "vocabulary lessons" in which one mostly has to translate too simple sentences, at least when one is in a more advanced level. I would be interested to go through every lesson which governs a new grammar issue but I don't have the motivation to learn each and everyone of those dull "vocabulary" lessons.
There aren't too many languages out currently, but I am expecting the language incubation process to speed up, especially when we can convert from ex. (from english to foreign convertion). Then we can expect something like a new course or so each week. That would be incredibly helpful for our community. I am really looking forward for the new courses to come out.
I love what Duolingo has been doing and I've been waiting with unbridled anticipation for the reverse courses!
But I do have an enquiry about how this will work. Are you going to get native experts of new languages work out a tree for each language that will then be used by each incubator team to create their reverse courses? Or will each tree be totally different based on which language you are learning from? Reason I ask is because I wonder if it's not a bad idea to try and come up with some sort of consistency between courses.
It's based on the source language. Since you're learning Portuguese and German from English, those are grouped together. If another language, such as German, were available for Dutch speakers, and you switched to the Dutch interface and chose to learn both, it would show that you're learning both English and German, same with learning from Spanish, if you choose to learn Portuguese, French and English from Spanish, these three will all show up with the Spanish interface because they're all from the same source language.
Thank you Luis! This will open a whole world of language learning opportunities. I really hope Duo will stay awesome and free. I'm learning German now, and I want to finish both trees before I start a new language; this will definitely take me some time so I hope by the time I'm done Duo will still be around to help me learn yet another language :-)
Japanese to English has been in stage 2 for months with no updates from the mods on its status. I would not hold your breath. My guess, is that Korean will come out first because the writing system is easier. Obviously the Japanese course is too buggy for them to even give us a little ray of hope in the form of an update. If it is too buggy, then it is too buggy. It was bound to be buggy since it has 3 sets of characters. Also English is a really hard language to teach a native Japanese speaker. They are not talking so I am guessing that there is not much progress to talk about.
Check this site for all sorts of information about Duo. The site was created by Duolingo users for just such a purpose :http://duolingo.wikia.com/wiki/Duolingo_Wiki#Getting_Help
When will speech be available for Turkish for English speakers? The tree was completed, you can write, listen to audio and write what you hear but there's no speech. I learn French and I can write, listen to audio and write what I hear and speak. Also Turkish has less features like Immersion and other things.
I don't believe it works this way. There are already lots of courses in development so it's challenging to add any new one. Even if the course is completed, there is still a lot of support required, especially in the first months. But in any case, the creation of the new course depends on many factors, some of them being:
- a presence of a team to create such course, or at least availability of mentors from other courses who can work on building such team (gathering and training new volunteers is not an easy task);
- Duoteam's interest in such course, maybe it's an interesting market, maybe there are some new marketing opportunities, maybe the team has someone who's good with a given language and interested in supporting it, and other internal reasons.
- What other plans the company have, say investing resources in redeveloping some of the old courses.
- How hard this new language is technically - does it require new alphabet added, new teaching methods, are voices readily available, and so on.
Overall, I would think that the more languages Duolingo has, the less likely it is that any new language will be added, though some combination of already existing languages becomes slightly more possible.