1. Forum
  2. >
  3. Topic: Duolingo
  4. >
  5. English for Thai has graduateā€¦


English for Thai has graduated!

That was fast, it has barely been a month since the course became available. Congratulations to the team who now holds both the record for the longest time in phase 1 (1127 days) and the shortest time in phase 2 (40 days).

June 19, 2017



I don't mean to belittle the efforts of the contributors in any way, but it seems to me that the quick graduation is more due to the fact that this is the first entry of DL into a whole new market in a very long time.

There are still very few learners, especially considering the size of Thailand's population (plus that of Laos for that matter, as I'm sure there are at least a few Laotians taking advantage of the similarities between the two languages). And I am not so sure that most people taking the course are even aware of the fact that they can and should submit reports.


Isn't (wasn't) the criteria for graduation having less than 3 reports/100 users for a certain period of time? In that case the size of the userbase shouldn't affect the graduation speed that much.


I believe the primary criterion is being at or below 3 reports/100 active users per day for two weeks.

I haven't been around for launches of other to-English trees, but it seems conceivable to me that this event is attributable to a combination of the Thai tree entering beta in a more advanced state than others, Thai's SVO sentence structure + analytic grammar, which presumably makes things a bit more straightforward than some other language, plus general unfamiliarity with Duolingo among the learners. I've certainly seen greater numbers of any number of Duo-unrepresented native languages represented in the English-language forum than Thai over time, and it seems reasonable to me that a relatively smaller percentage of users is made up of Duo-experienced / English-fluent users, the kinds of people best positioned and most likely to make numerous sentence reports from the moment of launch. There could also be impact from the fact that Thai was available on the app at a much earlier point than previous trees (I'm sure I saw it on iOS already at least two weeks ago), which increases users while simultaneously decreasing the likelihood of reports.


My random snooping around suggests you are onto something with the under-reporting angle. On the English side, the course has a completion level corresponding to what En for Cs had around the time I joined it as a contributor in the early stages of a tough beta. Exercises tended to have one or two English translations. Our daily reports at the time, if memory serves, were about 20-30/100. We had another 5 months of hard work ahead of us before graduating.


Interesting. Thanks!

Random question: how did you wind up having the top percentage contribution to English for Czech after only joining in the beta phase?


Those percentages are based on the contribution totals lumped across both directions of the course when the reverse exists. So my tally benefits from the roughly 10000 sentences I touched in the nearly completed Cs-En course. The runner-up, while dominant in the En-Cs course, is active nearly exclusively on that side.

The percentage ratios between us (Contrib1% : Contrib2%) differ in the two courses because the old course has localization string contributions lumped in as well.

One can derive interesting stuff from the percentages displayed publicly and the internal contributor tallies shown to the teams. For example, the 6180 localization strings must have generated a total of about 21000 edits. The remaining 47000 total "contributions" look like sentence edits. But because each sentence could have been edited by everyone on the team during its lifetime, this does not help much with figuring out how many sentences we have between the courses. Informal review of the two course leads' totals (and my recollection from the time of Phase 1 launch) suggests 15000 in the old course and 10000 in the new one.


@ nueby

Fascinating, thanks! I didn't know that even course contributors didn't have an easy way of knowing how many sentences there are in a course! I hadn't noticed until inspired to do the addition by this comment that the publicly-posted percentages seem to not generally add up to 100% (and they frequently only reach about 90).


Wyqtor, do you think the fact that English is taught in Thai schools or that Thai speakers might not know that there is an English for Thai speakers class available on Duolingo might account for the fact that there are so few Thai learners? How much does the Thai population use Duolingo at all? How many languages are there even available at this point for Thai speakers, which would make them aware of the platform. It takes a lot of effort on the part of a lot of people to get these classes going.


I followed the progress closely over the years, first with a lot of excitement, then disappointment because it just wouldn't move for months... It would be so awesome to get a glimpse behind the scenes, what happened there all this time! Anyhow, hopefully we can soon see Thai for English enter the incubator :)


Wow! Three years?

Anyway, congratulations to the Thai team!

I wonder how long the reverse course will take?


Just one resource I found good so far. I thought Thai was quite popular. Why doesn't Duo have it even a lot of people have been asking for it a while ago??? Anw, the app is called Ling On Play Store: https://play.google.com/store/apps/details?id=com.simyasolutions.ling.universal On App Store: https://apps.apple.com/th/app/learn-languages-with-ling/id1403783779


when will this option be available in the Duo Lingo app? Or is there some way to use the beta version? Wanna share this with some of my Thai friends, thanks!


It's there and has been for months. The only course offering listed in the Thai alphabet ;)

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