How is Learn Natively (1 guy) outpacing WK (A bunch of people) implementing so much in much less time?

@mods I know that community placeholder has been there for almost a year now, do you guys have a team? Who did my money go to? Just Koichi?

Brandon doesn’t even get paid and he’s done more in the last week then you guys have done all year.

26 Likes

Everytime I blink Bunpro has improved their site majorly in some way…
They also do this weird thing called “communicating with the users” where they just tell people what they want to do and respond to feedback.

50 Likes

While I agree with your general point that WaniKani is too slow, it’s often much easier for a startup / fresh codebase to move much faster than a legacy codebase. Speaking as someone who works at a company with 20+ years of tech debt, I understand how slow and painful the process can be. (That said, WaniKani does seem to be unusually slow here, and the lack of transparency in particular is a big issue.)

66 Likes

WK doesn’t need to improve as long as they have loyalists who fiercely defend them for free on the forums. In these people’s eyes, WK can never do anything wrong.

21 Likes

Yeah, I’ve seen Web devs being able to get things running in days. I myself managed to get a Web UI with a Python backend set up in 2 working days somehow. It was wild :sweat_smile:.

7 Likes

When Bunpro released their vocab decks a few months ago I said on the forum that I wanted the possibility to manually set the SRS level of individual entries, this way if I encountered a word that I already knew from outside bunpro but felt like I still wanted to SRS, I could manually set it at a high tier to avoid unnecessary repetitions.

I think in less than a week they had the feature implemented in a batch of other fixes and improvements.

A few weeks ago I beta-tested their new dashboard and disliked that an option was moved into a menu instead of being directly accessible through a button. I stated my rationale and a few days later the button was back.

The contrast with Wanikani is so stark it’s almost comical.

53 Likes

I imagine that they used to have a mid-sized dev team in the beginning back in 2012. Once they had the core product shipped, and got the levels defined more or less, they reduced the staff and kept only the support team + content team and probably a few devs. After all, when you have a working product and you’re getting stable revenue, there’s no reason to make major changes. They focused on the blog and SEO funnel to convert leads to WK users (that’s the case for me actually).

But now the picture is a lot different and there’s competition. They decided that they need to step up their game in terms of features, before the competition overtakes them completely.

Those were my 2 cents on the topic. Could be complete wrong lol.

9 Likes

So I would guess the same thing, but what I genuinely can’t reconcile is:

  • WK decided to work on a big overhaul of their frontend code (what resulted in the scriptocalypse). This is a costly dev effort with few short term benefits, I don’t think that many people decided to pass on WK because the HTML was a bit old school. This is the type of decision I expect from a company that is actively iterating on the website, pushing new features and experiments at a fast pace. This overhaul is a stepping stone to greater things, a preparatory work to make other changes easier later. This is not a move made by a website in “maintenance” mode.

  • Meanwhile all we’ve had since then are a bunch of new kana vocab entries (surely that could have been worked in the old frontend code easily?) and then basic features like a kana toggle or a summary page are promised and then nowhere to be seen for months. Also they broke the review API while they were at it. It almost feels as if WK has zero support left on the technical side.

As a software dev that makes zero sense to me. How could they not have a kana vocab toggle working within a couple of days? How could they not have a quick-and-dirty placeholder summary page working within a week? Third party script writers do so much more with so much less.

This is baffling to me.

43 Likes

Yeah, updating the frontend and breaking all scripts was an epic fail, combined with the burial of the summary page.

But I think the script game is flawed by design. After they saw that the community was eager to enhance the product with scripts, WK should’ve implemented a plugin system early on to prevent the hacky tampermonkey scripts. Having a native “Web store” with community plugins would’ve been way better both in terms of UI/UX. They could’ve tapped on the free community dev work to enhance their product (kudos to the people that keep supporting their scripts to this day). But since scripts cannot be integrated into WK, they need to ship new features by themselves.

Now it’s too late and they’re dealing with tech debt. I could never image that adding toggle can take this long. Completely unfathomable.

16 Likes

Maybe everyone spent 5+ years working on etoeto, and seeing all that work fall apart left the team demotivated.
Maybe after covid, everyone moved to work from home and the change in culture affected the team negatively.

Honest question, is Koichi still at the helm? When was the last time he said or did anything publicly? Has there been a change in leadership?

14 Likes

How dare you say that. I can’t believe it. After all they’ve done for you

13 Likes

My only guess at this point is that one way or an other they simply don’t have a single dev in-house working on WK at the moment. There are many reasons this could happen, I’m sure Tofugu is not a huge operation so I don’t expect that they ever had more than one dev at any given moment, and all sorts of issues could crop up that could put a wrench in the development.

Many projects have stalled because one key person fell sick or had to stop working for one reason or an other.

If that’s what’s happened here timing is extremely unfortunate since it happened just a big code change and before all the kinks were ironed out (and after years of the site being mostly in stable maintenance mode).

4 Likes

Yes, 2. 1 junior dev and 1 regular dev. At least according to the official team comp.

Which almost already happened. What WaniKani is riding on right now is the third-party apps which fix the lion share of WaniKani issues, and all sorts of integrations with other platforms.

Same. I would even say those are pretty bad decisions overall and executed in a rather strange order. Shutting down an important QoL feature (summary page) should be done as the last step before rolling out a new version of the app and not first and then leaving users in the dark regarding their progress.

None of these things are visible to the average user, I’m afraid. His account still exists on the forums, but none of the old crew was active on any topic for the last couple of years.

My suspicion is that WaniKani was always more or less a side project which grew in popularity and required extra man power to keep it running.

16 Likes

I’ve noticed that at least two WK team members are active on these forums and actively making changes to the content of the course (moving things around, fixing mnemonics etc…). It’s really the software dev side that seems to be AWOL for a few months.

I could guess what’s going on behind the scenes but I don’t think it would be helpful or charitable given that I have no way to know, what’s clear is that from an end user point of view it’s not particularly pleasant or reassuring for the future of the platform.

15 Likes

In light of this,
@koichi and @viet , do you guys still actively work on wanikani/at tofugu? I really don’t mean to be rude or snarky, I just want to know who/what wanikani is these days. Hope the best for you both!

11 Likes

I’ll be very surprised if they answer

11 Likes

Honestly that’s probably a perfectly fine way to run a website like WK. It’s really not that “big”, nor does it require a full dev team 24/7 over the course of decades. Not everything needs to grow to Netflix-size, regardless of what Venture Capitalists will tell you.

I mean if they had just kept the website running the way it was one year ago, I don’t think many would be complaining here. “If it ain’t broke” and all that.

What triggered this whole thing is the big breaking changes followed by months of nothing.

20 Likes

I don’t think that’s the right attitude for any user-facing website, especially if it’s meant to be a product. Things should be designed at scale. It goes all the way from using lists/maps/etc. instead of single values for things we expect to be in quantity of 1+ to making services scale horizontally instead of packing them on a single VM, and leaving room in a UI for potential future features.

Otherwise, this happens :frowning: .

4 Likes

Also, according to the website there are currently no open positions. So it’s not like they’re trying to increase the development staff.

9 Likes

Yes, but what have the Romans WaniKani ever done for us lately? :stuck_out_tongue:

14 Likes