Problem with Eo tree decaying too fast
I really think there is a problem with the decay rate of the tree in Esperanto. I've done both Esperanto and French on Duo, and my Esperanto tree decays much faster. It's almost impossible to keep it gold. Today I strengthened a subject, and less than two hours later it had decayed again!
It is impossible to assess without hard data but I feel that, since a few months, lessons are decaying much faster than they used to. Before gilded lessons stayed so for a while whereas now it is often whack-a-mole.
The more Duo languages you learn the more the dumb decay algorithm wants to transform repetition into a part-time job.
Yes, I definitely feel there was a change to the algorithm a few months ago, that's made it much harder to keep your tree gilded.
I once finished a skill only to have it immediately decay and never actually become gold.
I have a gold Esperanto tree, and it stayed almost all gold over the ten day period I completely neglected it while I was doing the Russian tree. I had about ten skills decayed by one notch. I regilded them all, and then the last few days I've been back to Russian. I went to check just now, and - since about Friday, I think? - I have had one skill decay by one notch.
I am guessing you are either using hints/making a lot of mistakes when you regild lessons, or else you have a bug specific to you. My experience suggests that once the tree is solidly gold, it'll stay that way. I don't think it's an issue with the tree itself, sorry, otherwise there would be no way my tree could stay that way...
I'd say it was just me, but I've heard multiple people make the same complaint. I almost never use hints, and I'd say I average about one mistake or less per lesson when re-gilding. I definitely used many more hints and made many more mistakes in French. So I'm leaning towards a bug.
It might be worth posting about in the troubleshooting forum?
Like I say, I don't see how it can be an Esperanto-wide issue, else my tree would currently be a sorry mess! But that's not to say there isn't a bug in the machine somewhere.
PS: I assume you've been doing French longer? Which may be part of it, if you've reviewed things more often? Though your EO level is higher so it's not an explanation by itself, since that implies you've done more with EO...
I did French for four months before switching to Esperanto, and I haven't touched French since. So I'm basing it off my previous experience with French in Jan-April and how hard it was to keep my French tree gilded then, as compared to Esperanto now. There may have been a change to the algorithm since then?
I wonder if finishing your tree has an effect on the algorithm? Because I hadn't finished my Eo tree when I posted this, and I have now. I will see if that alters the decay rate.
I haven't kept a close eye on finished versus unfinished trees, so I couldn't say for sure.
I did work pretty hard making my EO tree 'solid' gold, if you will; prior to Russian making an appearance, I was seeing maybe one or two skills per day maximum decay, and I usually did at least one general strengthen anyway.
I am currently attempting to get my Russian tree to a similar state, so Esperanto has been horribly neglected... 8-o
There's a possibility that they're running some kind of A/B test and you're in the group where skills decay faster to see how that affects site usage. That's only my personal theory though. To get a more definitive answer, I'd recommend going to the Troubleshooting forum like flootzavut recommended. :)
I don't think its an A/B/test simply because my EO tree has been decaying rapidly but not my other languages. I have more decayed skills in Esperanto than in Irish, French, and German COMBINED every single day. I can't keep up with it. I have a R/L job that's isn't reviewing Esperanto. At this rate I'm gonna have to give the language up because I don't have the time.
Aren't you confusing the goal of learning the language with that of keeping your tree golden? It's no reason to quit. You're still learning Esperanto. Enjoy that and don't get too worried about the Candy Crush style points aspect of Duolingo. That's a helpful incentive, but not your ultimate goal.
I'm not confusing anything. You need to have some measure of progress in order to know if you are learning anything. Just because my measure of progress is different from yours does not mean that mine is automatically defunt. I am not at all worried about the "candy crush style points" aspect and quite frankly I'm insulted that you have implied that I am. Having to pour hours into falsely decaying skills is not enjoyable its frustrating. My ultimate goal (since you just assumed again) is to ENJOY learning Esperanto. This flaw is making that difficult for me, so I am considering leaving Esperanto.
l was trying to encourage you, not have a go. I'm sorry you've taken it that way.
I don't quite get how you're "having" to put time into your skills score (especially when you think it false) to quantify your progress when you still have your level.
I was trying to keep my skills up a while back, but realised it was keeping my progress static. So l decided to push on with the tree. Once I've finished that l'm going to work on the skill rating as a daily practice and if there is a glitch with it they'll hopefully have it fixed by then.
Remember, lots of language learning systems don't even have a skill rating and people get through them OK. So I say put a bit of imaginary tape over that part of the screen rather than throw away all your hard work
Once again, sorry my post read the way it did to you I was going for "Hey, don't sweat that particular number" as opposed to "You're doing it wrong!" (especially as you're ahead of me!).
The gildening should help to stay in a state of a Czentmihaly FLOW by rewarding and marking progress. At the current setting, the decay rate is set too high for me. Too often it is not rewarding my efforts but almost mocking it (by either not moving the needle at all or gildening the lesson for a very short time period). Instead of motivating, it is actually demotivating as despite efforts, little progress is made.
What is lacking is a concept of user input into the training plan. When you train your body in a fitness study, you commit to a number of repetitions (sets) that you are comfortable with and which are doable by your current level of fitness. I wish Duo would offer an option of setting the decay according to some point commitment level. Perhaps I am not willing to go for gold but only want to train at a bronze pace.
The decay rate should also differentiate between different stages of a learner's progress (opening, middle, endgame). My Spanish "Duo fluency" has been reduced from 57 to 43 percent within a three month period. This is not an accurate measure of my Spanish skills but just shows that I have shifted my Spanish learning away from Duo. There should be some kind of "maintenance/endgame" mode once you have outgrown Duo's level.