I'm falling out of love with Duolingo.
As I'm sure you know, Duolingo is going through a "rough patch" in its development.
Due to this, I find that I'm coming on here less and less.
A few months ago, I was on here every single day, for at least an hour or two. Immersion was still in full swing, everything seemed to be running fine, and the community, in general, was happy.
Now things have changed. With Immersion gone, Activity gone, the Forums running slower than ever, and spammers taking over, Duolingo is entering The Dark Ages.
Hopefully this will all be fixed soon, but for now I think I've fallen out of love with Duolingo.
What do you guys think?
This is probably an unpopular opinion but personally I don't mind the changes. I never used immersion as it didn't exist in my languages, and anyway from what I heard it was full of trolls and cheaters. I'm not denying that some people did use it properly and do seem to miss it a lot, and I feel bad for them, but given the amount of people that used to complain about it all the time in the forums I can't help feeling that people are remembering it through a rose-tinted lens. I also never used the activity stream.
The reason do I like duo hasn't changed, and that's its actual learning mechanism of playing you a sentence that you translate or type out or whatever. It's so easy to use and in my experience it works pretty well. So Duo still offers an effective way to learn the basics in a wide range of languages, and people can still use the comments under the sentences to ask about any problems they have, which is a much more efficient way of getting targeted help than posting in the general forum anyway.
And if removing the activity stream and immersion makes duo faster, I'm fully supportive of that because I do find it slow sometimes. Also if it makes it cheaper and easier to run, in the long run that might mean they are able to offer more languages, or simply keep going for longer.
It's worth bearing in mind that the people who will miss the activity stream the most are pretty much by definition the ones who are most engaged in the community aspect - but when you look at the numbers of people who actually use Duo, they are definitely in the minority. I think most users are probably like me and are only really interested in the lessons rather than the community stuff, and that seems to be backed up by the Duo team's own metrics and decisions. If that's true then removing the activity stream to make the website faster is actually beneficial to more people than it's harmful to. It's just that the people benefiting from it are by definition less likely to post about it.
Anybody who complained about "cheating" in Immersion in my view primarily revealed themselves to be more concerned about made-up internet points (and not even their own made-up internet points, but rather some other random person in cyberspace's imaginary internet points) than about keeping themselves focused on their own language learning. So sure, there are undoubtedly rose-colored glasses, but there have been legions of dystopia-tinted ones, too. I couldn't help but wonder how all these translation tier 1 Immersion users had managed to stumble across ever so much malfeasance when I only happened upon an isolated incident or two. Hard not to conclude that they saw negative comments and embraced them on little of their own evidence.
I can't speak to your languages, but the Activity Stream was also very important to a lot of the people who make the sentence discussions useful. It seems that most courses I've done have a small group of dedicated native speakers or very advanced learners who answer the bulk of the questions. I know it depends on language, but in the ones I'm most familiar with, these people were not moderators, but they do communicate with each other, and they even look at other users' streams to find as-yet unanswered questions. I, too, think that speeding up the site will be quite beneficial, but it is also a change that will have big negative effects many or most won't even realize.
In the "Immersion" case, some people earned XP by replacing accurately translated sentences with inaccurate versions. The part of the system that gave you XP for translating a sentence and the part that checked whether or not your translation made sense were not the same part, after all.
No doubt that behavior slowed down the process of moving a document from 0% translated and checked to 100% translated and checked.
No doubt slowing down the process of moving a document from 0% translated and checked to 100% translated and checked made it harder for Duolingo to attract paying customers to this crowdsourced-translation service, and contributed to Duolingo giving up on selling crowdsourced translations.
No wonder some people who care about language learning disapproved of this "cheating," no matter how imaginary and superficial the rewards for the cheaters were...
This is an excellent explanation for why immersion was unsustainable, and one I haven't heard before. Have a lingot. I would also add that even without the cheaters, it wasn't a viable long-term business plan. Crowd-sourced translation, when the crowd are not actually qualified to translate, could never really compete with the work of an experienced, qualified linguist, and I'm sure most companies Duo was trying to sell to could see that. As they say, a horse designed by a committee becomes a camel, and in the case of immersion, the committee didn't even speak horse, as it were. And then the cheating and the complete beginners posting garbled nonsense were the straw that broke this particular camel's back.
Whether crowd-sourced translation could sustain itself as a business model was a question undoubtedly answered in the negative very quickly (if it was every anything more than a marketing gimmick - a website not supported by ads, now that's worth a write-up in the kinds of publications your future user base might just happen to be reading!). A question of substantially greater interest was whether it could generate sufficient pageviews to sustain itself as a learning modality. Obviously, this was also answered in the negative, the fault for which lies in many corners.
piguy3 is absolutely right in every point (s)he makes. As far as questioning the authenticity and accuracy of translated articles from Immersion, a wonderful tool I was fond of, one has to first understand what articles were being translated for profit and, most importantly, in what languages. Only articles in Spanish remained under this crowd-sourced umbrella of making a profit throughout the life span of Immersion. During that time, French and, to a lesser degree unless I'm mistaken, German. I personally know of some of the most active translators in Spanish Immersion and we are talking about top notch linguists. Aside from some who were professional interpreters for some of the foremost agencies and organizations in this field others were established polyglots. On top of that, they kept a very tight group of people with access to the articles they translated and this worked very well. The rest of it was left to others who, admittedly were not of the same caliber, but these folks tended to stick with Wikipedia and other irrelevant to the business aspect of this enterprise, so it really didn't matter how that went. Lastly, as someone who went up to level 10 in Turkish, one of the highest among any of active translators in Turkish and found it very difficult to keep going given how small the community around this language was, I still loved the Immersion tool and worked on it regardless of xp points, though they were a cool, little incentive.
I loved Immersion, and I miss it a great deal, but it was clear to anyone paying attention that it was on its way out. When Luis had his AMA he said there was a new website coming but that the only change users would notice was greater speed. Then in sticky post about the new website, the programmer mentioned in one of her comments that they always knew the activity stream was going so they just didn't implement any replacement for the new website. It sure seems like we were straight out lied to, and it leaves a horrible taste in the mouth. My activity stream was made up of people I'd actually communicated with, and frequently, and over quite a long period of time. It made it feel like language learning was something I was doing around a vast, global, friendly fireplace. That's what's gone now, and it doesn't look like it will be replaced.
Yes. Although I don't have health/gems, it seems like a bad idea. Heaven forbids it being pushed to all DL users.
The reason for removing activity wasn't a silly excuse if the alternative was no Duolingo at all, as stated. I understand the pessimism (out immersion, out activity, out conjugations, out flashcards... In clubs, bots, obnoxious owls, classrooms and the concomitant spamming)- in continuing the car metaphor, Duo seems a little like a rusted jalopy right now, but if you look over there (no, there, at the horizon), it looks like maybe a new auto plant is being built. It will either be built or not.
Decluttering is good for the soul. There are some things, if they are more trouble than they are worth, that if you don't free yourself from them they will hold you back from doing what you want.
That's what immersion and the activity stream were to Duolingo. The weight of maintenance and support required to carry them did not match their usefulness, and now that they have been axed, they are once again free to focus on experimenting on more useful things, with the benefit of learning from the experience of what doesn't work.
I've been around for a long time as well. I felt this way after the earliest changes were made. However, two things: 1. Duolingo is in the middle of "Reconstruction"....wait. 2. There are any number of similar web sites now....vote with your feet. And after the "Dark Ages" came the "Renaissance"!! :)
I agree so much!
I never got to try out Immersion (but always wanted to), but I miss the Activity stream and the Progress Quiz, which I liked to use, because I trusted it more than the skill percentage or the lesson status.
Adding to that I got moved to the new site and none of my userscripts work. I didn't realize how much I depended on them until they were gone... I realize that they are made by users and have to be fixed by users but I would habe hoped they'd have included some of the functionality in the rewrite (like a course switcher for different base languages or a tree trimmer, not to mention a strength viewer, etc.)
So instead of more functionality I feel like I'm getting less and less. It's sad and my motivation isn't what I want it to be.
Well, there's a new course switcher script out: https://www.duolingo.com/comment/22412321 (although I don't really understand the demand for it; I have courses from 7 base languages I think, and they're easy to navigate between by clicking the flag at the top of the current tree and then pressing the first letter of the relevant string; like "j" for courses from French or "q" for Spanish).
The new course switcher has all the languages. I'm using it for now, but it's not as comfortable as the old one...
The problem with the flag is that it doesn't work to switch to my French tree that I started from German (my native language) but I get to the tree for French from English (which I did start as well)... So yeah, problem for me...