https://www.duolingo.com/JoshGreenhalgh

A great suggestion for the Incubator!

  • 25
  • 6
  • 6
  • 4
  • 3
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 25

I just thought of a great idea for the courses in the incubator today!

The suggestion is, that rather than releasing the entire course at once, the idea would be to release it in parts, for example, the team would work on the first 12-20 skills in the tree and once that is ready, it gets released, then the contributors start to work on the next quarter or so of the course and so on. This would mean that firstly, the community gets to try parts of the course sooner, something we all want, secondly people can learn the tree as it gets released, and thirdly, it means if there are any stalled courses, like Romanian or Russian, we would still have part of the tree to learn.

If you like this idea, please up-vote it and if an Admin and or someone involved in the Incubator sees this, please tell me what you think. Thanks!

March 7, 2015

29 Comments


https://www.duolingo.com/Windrammer
  • 14
  • 10
  • 8
  • 7
  • 3
  • 3

I believe that's partially similar to what the Esperanto-teamo is going for right now.

March 7, 2015

https://www.duolingo.com/Yoruji
  • 11
  • 11
  • 10
  • 8
  • 7
  • 7
  • 6
  • 6
  • 4
  • 4
  • 2
  • 2

Yes, and Hungarian too.

March 7, 2015

https://www.duolingo.com/Windrammer
  • 14
  • 10
  • 8
  • 7
  • 3
  • 3

Oh, that's right!

March 7, 2015

https://www.duolingo.com/amuzulo
  • 15
  • 13
  • 12
  • 10
  • 9
  • 6
  • 6
  • 5
  • 3
  • 2

Yes, for Esperanto we had this epic tree, which you can see here: http://learnlangs.com/duolingo_esperanto but then Luis talked with us and we decided to launch with it up to Communication and Abstract Objects 2.

March 7, 2015

https://www.duolingo.com/flootzavut
Plus
  • 25
  • 19
  • 18
  • 17
  • 14
  • 13
  • 13
  • 13
  • 12
  • 12
  • 12
  • 12
  • 9
  • 8
  • 8
  • 6
  • 5
  • 5
  • 5
  • 5
  • 5
  • 4
  • 3
  • 3
  • 2
  • 2
  • 2
  • 2

I was going to say the same.

I mean, with Esperanto and Hungarian in particular, I would say yes please release asap but there are other considerations, like finding a TTS and stuff.

(Random: has anyone seen the screenshot of the Esperanto tree? looks really good :D cannot wait.)

March 7, 2015

https://www.duolingo.com/vikungen

Or maybe you would say: Eldonu la kurson tuj ;) And yes, it looks awesome!

March 7, 2015

https://www.duolingo.com/flootzavut
Plus
  • 25
  • 19
  • 18
  • 17
  • 14
  • 13
  • 13
  • 13
  • 12
  • 12
  • 12
  • 12
  • 9
  • 8
  • 8
  • 6
  • 5
  • 5
  • 5
  • 5
  • 5
  • 4
  • 3
  • 3
  • 2
  • 2
  • 2
  • 2

:D point!

I'm glad it's not just me looking at that screenshot and going "gimme!" I guess it helps a lot that with Esperanto, the grammar teaching isn't taking so much time and space up on the tree, so there's loads of room for practising stuff and learning vocab and fine tuning and so on. It just looks like it's going to be a lot of fun.

I can't remember where I read it, but as someone said, in the space where you can usually get a fairly decent grasp to lower intermediate level, for Esperanto with its pared down grammar and lack of exceptions, there's a lot of room to expand on stuff. It will be really interesting to see how competent people are at the other end of the tree...

March 7, 2015

https://www.duolingo.com/amuzulo
  • 15
  • 13
  • 12
  • 10
  • 9
  • 6
  • 6
  • 5
  • 3
  • 2

I wrote that somewhere, but I don't remember where either. I still believe it! :)

Needless to say I'm also super curious what level people will be at when they finish the Esperanto tree. :)

March 7, 2015

https://www.duolingo.com/flootzavut
Plus
  • 25
  • 19
  • 18
  • 17
  • 14
  • 13
  • 13
  • 13
  • 12
  • 12
  • 12
  • 12
  • 9
  • 8
  • 8
  • 6
  • 5
  • 5
  • 5
  • 5
  • 5
  • 4
  • 3
  • 3
  • 2
  • 2
  • 2
  • 2

Hah! Well at least I know who I was quoting now, I guess ;p :D

Yes, I imagine the team's curiosity will far outstrip mine ;)

I wonder if there are any free tests out there - like the ones for Euro languages on transparent.com - to give people an idea of where they're at (if they have any EO pre tree) and then after they've finished it? I know there's a limit to accuracy to those kinds of tests, but it could be interesting.

I noticed it's back up to 83% now - good to see!

ETA: spoke too soon... 84% Nice!

March 8, 2015

https://www.duolingo.com/peter.kristof.hu
  • 23
  • 19
  • 17
  • 16
  • 15
  • 14
  • 9
  • 8
  • 4
  • 3
  • 2
  • 2
  • 1817

Yes, the PUB (percentage until beta) and PUC (percentage until completion) but these are only two levels. It's a good idea.

March 7, 2015

https://www.duolingo.com/Dessamator
  • 25
  • 14
  • 13
  • 13

You seem to be suggesting someting like the RAD approach [1]. Unfortunately generally this doesn't work for volunteer led projects because this generally requires committed people who stick to deadlines.

They'll be stuck with maintaining an old tree in addition to creating an new one. This currently works for reverse courses because the original tree is finished and doesn't need to be changed as often.

Duolingo also probably doesn't allow courses to be released before they reach a certain threshold, maybe 40 skills and 1000 words. In those cases the tree is already good and can be slowly developed thereafter or even left as is and wouldn't matter much. This is unless Duolingo finds some way of helping contributors fix mistakes in the old tree without as much effort as they currently describe.

March 7, 2015

https://www.duolingo.com/amuzulo
  • 15
  • 13
  • 12
  • 10
  • 9
  • 6
  • 6
  • 5
  • 3
  • 2

I'm under the impression that the Incubator doesn't make it very difficult to do this. All the sentences are just sorta saved somewhere, not specific to any lesson, so if you move words around the tree, the sentences just automatically move to the place where all the words in that sentence have been taught. So, considering that sentences are separate from the tree itself, it seems like it would be relatively easy to work on Tree 2.0, because you probably won't be making too many big structural changes anymore in Tree 1.0.

In any case, we should find out in a few months. ;)

March 7, 2015

https://www.duolingo.com/Dessamator
  • 25
  • 14
  • 13
  • 13

Well, I was highlighting two issues.

  1. With popular courses such as Spanish and French (and others to come), there are probably more than 10000 reports per week. Going through these reports is probably time consuming, and will prevent course developers from simply focusing on creating a new tree.

  2. A French contributor and moderator indicated that at least for Staff courses, creating a new version was a lot of work. Not something as simple as one would expect.

Personally, as someone trained in Software Development, I'm all for divide and conquer. But, Duolingo's incubator doesn't seem well-suited for this.

It seems more sensible to release few discrete units (skills) , and debug them heavily rather than releasing a big one and and awaiting for the eventual fallout (tons of reports and bugs).

The problem here is simply human resources , and efficacious filtering algorithms for the reports.

P.S. You could also sitesurf and Dxli , the French moderators about their experience creating a new language tree.

March 7, 2015

https://www.duolingo.com/Dessamator
  • 25
  • 14
  • 13
  • 13

I just realized that another problem is that extremely short trees will invalidate Duolingo's assertion that its language courses teach about as much as a semester of college [1]. Which is probably still the case for most language trees.

March 8, 2015

https://www.duolingo.com/ConorHoughton
  • 25
  • 25
  • 25
  • 18
  • 10
  • 10
  • 9
  • 8
  • 8
  • 6
  • 6
  • 5
  • 3
  • 2
  • 1599

Or to have a separate category of languages that only do a small tree; minority languages, endangered languages, historic language, dialects, for example; releasing a small tree, eg one you could finished by level 9, would allow the many people who might be mainly interested in getting a taster, a feel for the language to have a go and would allow a community to build up around the language on duolingo which might, later, lead to a full tree.

March 7, 2015

https://www.duolingo.com/flootzavut
Plus
  • 25
  • 19
  • 18
  • 17
  • 14
  • 13
  • 13
  • 13
  • 12
  • 12
  • 12
  • 12
  • 9
  • 8
  • 8
  • 6
  • 5
  • 5
  • 5
  • 5
  • 5
  • 4
  • 3
  • 3
  • 2
  • 2
  • 2
  • 2

Now that is an intriguing idea! Maybe aiming for "tourist plus" as a goal to get people to? Basics and a little more.

ETA: I think what appeals to me most with this one is the idea people will get a feel for the language, and those which prove popular will hopefully build a community and then perhaps a whole tree later.

March 7, 2015

https://www.duolingo.com/RikSha
  • 18
  • 18
  • 13
  • 13
  • 13
  • 13
  • 12
  • 11
  • 11
  • 11
  • 10
  • 9

And then maybe later even an advanced tree after the basics have been covered?

March 7, 2015

https://www.duolingo.com/deguo

Some courses are already doing this (Hungarian and Esperanto), and all teams have the option to do this if they so choose. Russian turned down the offer, and I can't say I blame them. I'm personally not a fan of the idea, but it all comes down to what the contributors want.

March 7, 2015

https://www.duolingo.com/amuzulo
  • 15
  • 13
  • 12
  • 10
  • 9
  • 6
  • 6
  • 5
  • 3
  • 2

I wonder how many want this, so we can stop getting barraged with the question, "When is your course going into beta?" :-D

March 7, 2015

https://www.duolingo.com/Pa11en

I am just appreciative that the Esperanto course is being created. I consider the wait time get ahead time.

March 7, 2015

https://www.duolingo.com/Chuggy370
  • 8
  • 6
  • 5
  • 5
  • 4
  • 3

What's the point of learning a tiny chunk of a language, but having to wait upwards of 3 weeks for the next part of the course to come out, you'll forget it.

Still a good idea, I up-voted :)

March 7, 2015

https://www.duolingo.com/WahahaDrills
  • 22
  • 18
  • 18
  • 14
  • 9
  • 7
  • 7
  • 5
  • 5
  • 4
  • 4
  • 2

I think that, if anything, it would keep you from speeding through the tree.

March 7, 2015

https://www.duolingo.com/Chuggy370
  • 8
  • 6
  • 5
  • 5
  • 4
  • 3

I speed through 6 units on a weekend, then study them 2-3 hours a day through the week. Nothing wrong with speeding through stuff as long as you revise.

March 8, 2015

https://www.duolingo.com/-NicolasSarkozy-

Good idea, Francois Approves!

March 7, 2015

https://www.duolingo.com/HappyEvilSlosh
March 7, 2015

https://www.duolingo.com/sbeecroft
  • 20
  • 12
  • 12
  • 11
  • 11
  • 11
  • 7
  • 4
  • 74

I like the idea, personally. But we already have tons of people who whine and moan every time a restructure comes to a set of language lessons. Releasing a pre-beta version means that you will certainly have such major restructurings, probably quite a bit. If there were some way of insuring that only the non-whiners could participate in the alpha stage, that would be great. But I imagine it would be a great demotivator to work hard on making a language availabe in alpha, only to be greeted by a chorus of boos every time you needed to update the nascent tree.

March 9, 2015

https://www.duolingo.com/evlamb
  • 24
  • 22
  • 19
  • 10
  • 10
  • 7

Fixing obstacles in beta for the first tree would be a serious problem when coupled to making additional skills. The resources and contributors would be spread thin and less likely to be as efficient and effective. I would love to see some of the courses out earlier but unfortunately it wouldn't be efficient enough.

March 7, 2015

https://www.duolingo.com/-Feodor-

I think this is a fantastic idea.

March 7, 2015

https://www.duolingo.com/HebrewLearner

I think this is a great idea! I would love to start learning Polish and using this I could start learning Hebrew and/or Arabic faster once they get on here! Thanks for the suggestion :D

March 8, 2015
Learn a language in just 5 minutes a day. For free.