If you didn’t want baseless speculation, you should have given users an in-depth explanation for why you were implementing an annoying update :^)
20 Likes
Thank you for the update and more details! 
That new timeout logic sounds more reasonable to me.
This! I never minded the change, but the whole reason I pulled the switch on lifetime is because I felt good supporting the team, even though I was iffy about whether I really needed the product. Why did I feel good supporting the team? Because I sent a few different emails asking questions about the service and I received really nice, detailed replies within hours. Even if the situation is something absurd like “our code is so broken we either need to charge 5K per user or let it fall apart in a year”, most people aren’t going to be too resentful about it as long as there is clear communication up front. It’s the real issue with breaking add-ons, IMO: losing functionality and having to rebuild everything probably doesn’t even hurt peoples’ experience as much as the anxiety of having no idea what’s going to break when.
7 Likes
I have just been logged out again after 2 weeks despite daily activity. I’m on Firefox (Android) and checked the “remember me” box. Does the automatic extension only work if I don’t check the box or did I misunderstand something about this post saying that it would be extended through activity?
2 Likes
Genuine question, why can’t you just use the fact that users have not done a review or lesson for two weeks to mark users as inactive (rather than tying “active user” bit to “user is logged in” bit)?
3 Likes
The auto extension should work regardless of whether you checked the box.
Sorry I am not following your question
1 Like
Thanks for the quick reply! Unfortunately it currently doesn’t work for me.
If there is any further information I can provide to help debug, please let me know.
the remember me token only lasts for 2 weeks and is not automatically extended. So if you have clicked remember me and after 2 weeks your session expires (by closing your browser) then you will need to log in again. If your session cookie remains active, and you continue to do reviews then you will remain logged in.
1 Like
So, it basically kicks me out every two weeks unless I keep my browser running 24/7 and the renewing is just to circumvent the 4 hour kick? Seems I misunderstood the renewing as it keeps me logged in as long as I’m active.
To me this change is a nuisance that will make sure I get reminded of WK’s downgrades every two weeks. 
Even as a fellow dev, I don’t see how kicking people out every two weeks is beneficial when you factor in the worse UX caused by it, especially when most other applications just have “remember me” without an artificial two week limit.
I really hope this will be revised in the future.
6 Likes
Yes. I did shorten the time for remember me from 4 to 2 weeks, but that is how it has always worked.
Presumably you were reminded every 4 weeks prior to this change.
Noted
5 Likes
You mentioned that you force log people out after 4 hours to mark them inactive so you can delay some database processes for those users. Why can’t you simply use the fact that the users haven’t done review in 2 weeks to mark them as inactive instead (which I know you can do because of the ability to ensure these people don’t get logged out that you mentioned earlier)?
If you’re relying on the “logging in” step to perform those delayed processing, can’t you just move it to when users load the homepage URL. Or delay them further to when they hit the lesson/review endpoint to further delay the processing—the worst I can see here is some UI inconsistencies until after the database processing is completed, an acceptable trade off for users I think.
5 Likes
I think I already answered that above. Users who still log in but don’t do reviews are still considered active. Consider a Level 60 user (as one example) who has no reviews left to do.
4 Likes
But why though? Most websites extend the session cookie whenever the user visits. If they visit the site they are clearly active, so why shouldn’t the session cookie be refreshed for another two weeks?
9 Likes
Is this population so large that it causes an immense burden to the system to automatically consider their login to be the same as a <60 user?
1 Like
The level 60 population is about an order of magnitude larger than the level 59 population, if my memory of the charts that were published back in the day is correct. Though, the active level 60 population might be another matter. 
4 Likes
What if you just booted everyone off the system en masse every time you wanted to do an update, mid-review session or not. I mean, people gonna complain no matter what, might as well go all the way.
This may be why I’m not in a public-facing job.
13 Likes
This is what I wonder, since there are certainly lvl 60s who linger around, but I’d imagine most of you guys have moved on? I guess I just don’t understand how the population of lvl 60s who still log in would be the given example of why a two-week login time limit is necessary.
Well, if someone posts in the help section of the forums and says something like “this example or audio or whatever from a WK page is confusing” and posts a link, in the past I would check it out and see if I can help. But now it’s a near-guarantee I’d be logged out of WaniKani so I can’t be bothered.
(obviously level 60 users isn’t a major use case, but it is still annoying)
8 Likes
As another firefox user, I am experiencing the same thing. Running 131.0.3 on PC.
EDIT:
Apparently I misunderstood the refresh function. I thought the 2 weeks were extended every time I logged in. This appears not to be the case, and we are logged out 2 weeks after we last logged in.
2 Likes
There are already cheaper or even free products that do what WK does and more. WK is already disappearing off of so many recommendation lists.
5 Likes