The Polyglot Scrivener
Learn languages by writing. Japanese (mainly) & Spanish
It has been a while since my last post, but the project is still ongoing. I have diverted my focus to developing a web app. The idea behind The Polyglot Scrivener website was to serve as a launchpad for a Japanese language learning app.
As a programmer, creating the app has been more rewarding than content creation. After a few months of hard work, the beta version is already available. While there is still much content to be created and testing to be done, I am excited to share a quick summary of the app with you. You can also find it in https://polyglotscrivener.com/language-app .
The app's purpose is to facilitate the learning of Japanese vocabulary in the context of sentences. Memorizing words by going one by one in vocabulary builders has proven ineffective. To address this issue, I created an app that enables users to learn vocabulary through sentences. Here's how it works:
- Sentences are grouped by topic in "bundles" which can be added to the user's dashboard.
- Once sentences are in the dashboard, users can learn, practice, and review them using spaced repetition, fill-in-the-blanks, and guessing games.
- Learned sentences are stored in a main table with a "recall" bar indicating how well the user remembers the sentences (using a forgetting curve algorithm).
- There's a "Lessons" section with bite-sized explanations of Japanese grammar.
- The app features audio, difficulty levels, write or select for fill-in-the-blanks game, and an option to enable furigana.
I recently added a writing game as I want to keep it writing-oriented. While this feature is not yet fully integrated into the app, you can already practice writing some kanji. I found a nice machine learning model for kanji recognition that I use in this feature.
Moving forward, I need to give the app more exposure in social media and drive traffic to the website. Here are some of my plans to achieve this:
- I have requested to add the app on Betalist, but it may take some time.
- I will announce it on Indie Hackers building-in-public.
- I am planning to post in Reddit to get some feedback.
- I also need to work more on the website content as a gateway to the app.
For now, I am not planning to put much more effort into developing the app itself. I have already invested a fair amount of time, and it is now time to get it out there, see if learners are interested and get some feedback. As I have shifted my focus towards the app, any tips on what to do next would be greatly appreciated.
You can check it out from https://polyglotscrivener.com/language-app
Thank you.
I launched a new feature to see how the characters are written for the different sentences. This is available for all previous posts with sentences lists.
Before this feature, I think the value of those sentences lists was very low. Now that the user can see how to write it, I think it starts to be a bit more useful.
I am kind of proud of this new feature, it has actually helped me with my Japanese writing skills which is definitely a good sign as, at the end, I am a user of my own website too.
I have never been very "social" as in having my own YouTube channel, Twitter, Instagram, etc. Attempting to boost my SEO, I created a small channel with a couple of video lessons based on some of the posts with sentences I created on the website. I added a link to the footer of the website to access the channel.
The videos go one by one through a set of sentences, the user can listen to the pronunciation in Japanese (I recorded my Japanese girlfriend 🙇), and then see how to write it, character by character.
I don't have big hopes on this, it took me a while to create those videos so, I doubt I will continue but let's see.
As expected (or maybe not?), I am getting from very little traffic to none.
The site has very little content which means that it is not very SEO friendly. Many of the posts are actually not real posts but a list of sentences to learn and try to write. Based on GA, I had no more than 15 unique visitors in the last month.
I am also suspecting on the technology used. I am using Angular and, in theory, it is Angular Universal so it should be SEO friendly.
I run some SEO tools that pointed out some H1 issues, images, speed, etc. Overall it was not bad for desktop, weak for mobile phone. I am adding a CDN and some other performance improvements and see how it goes.
-
1
PS: Maybe you should be posting your updates in this group: https://www.indiehackers.com/group/building-in-public
-
1
I think you have to start with clearly identifying the target market, before figuring out the best way to draw them in. So SEO might make sense for you or social or content marketing or ads or something else. Point being, no point in offering advise on SEO until it's clear that's a strategy that makes most sense for you (for one thing it's a very long term strategy).
I note you have done some Youtube videos and if you're doing something educational, and your target audience uses Youtube, then that could be a source of traffic for you. Here then, you'd want to focus on keyword research first then make the videos and then make sure they are optimised for those keywords and you have a clear CTA with link to your site - for example - rather than focussing on written SEO content.
I know from experience, you can build as much as you want, but you wont attract any visitors without the marketing piece
You'll also get the most useful help, if you can be specific about what you want help with and include links.
Best of luck.
-
1
"The site has very little content which means that it is not very SEO friendly" Seems like you've identified the main problem right there, but i'm not sure if you're asking a question or for advice or just updating on your progress?
-
1
I just added my product to Indie Hackers and already got a comment! This is indeed a very active community.
Hi @SymbioticNick, initially I just wanted to track a bit my progress but any advice would be highly appreciated.
As you said, I probably have identified the main problem. I really struggle with content while I have a lot of fun coding new features (I am already thinking about some Saas for the near feature.)
In addition to that, I haven't really done any keyword research, marketing, social or any other thing. I also know that backlinks are really important driving traffic to a site but I don't know how to get them.
-
I knew I was going to struggle with content as I am not a content creator, I am a programmer (or at least I used to be). I decided to make the website a mix between an app and a blog. The content in the form of blog posts will be required for SEO as well as for providing additional value, with the features I was planning to differentiate myself from other Japanese/language learning sites.
Content: I started with some posts about language learning and, more specifically, how writing can help.
Features:
- Sentence by sentence posts: a list of sentences for a specific topic is displayed.
- Copy (write down) the sentence you see.
- See the translation
- Test yourself:
- For the different JSLP levels, see a word and click thumbs up if you remember or thumbs down if you don't. The user can try 20 or 80 words. A small summary is displayed at the end.
You can also try to write the word. - Guess the Kanji reading. A list of sentences is presented one by one with a Kanji character highlighted. You need to enter the reading in Hiragana.
In my next feature I will try to add a bit more value to the sentence by sentence posts by adding a small visualisation of how the Japanese characters are written for the sentence.
I always wanted to create something by myself. Nowadays, I don't get that much time for coding, but I remember always having fun doing it. Together with my interest on learning languages, particularly Japanese (currently living in Tokyo), I decided to start a website, a bit more interactive than a blog, focused on writing skills.
I have always been amazed by the Japanese script, but I could not find many websites focused on it.
3 Comments
I'm definitely interested - kind of busy right now but will plan to take a look in a day or two.
I am learning to speak, read and write Japanese.
I'm at an 'advanced beginner' level - I have passed the JLPT N5 exam, missed passing the N4 exam by 3 points, and am currently studying to (maybe) take the N3 exam in December (but that is probably too aggressive a plan, so maybe I will retake the N4 instead).
Currently using a variety of sources, including books and the Wanikani.com and Nativshark.com web apps, and have a subscription to Satori Reader but need to start using it. Also watching a lot of anime with Japanese vocals and English subtitles.
Would like to get to a spot where I've learned all of the Joyo kanji, but that will be a long time coming, as I'm not a student, rather I have a fairly demanding full-time job.
Hi,
At this moment you will find the app very easy as I have not added yet any beginner advanced stuff (like Genki 2 and beyond). Still in the process of adding content though but I want to prove the idea before spending too much time on content.
It would be great to have some feedback and ideas from someone like you, considering that you are actually learning Japanese and in N4 - N3 level. I am probably around that level too but, as the creator of the app, my perspective is a bit bias. Content aside, I am looking at what kind of features could be helpful and are not found in some other apps, including writing related ones as I am considering giving strong focus to that.
Whenever you have time, take a look and let me know!
Many thanks.
It took me much longer than expected to get back to this - I have only started to look at the site.
FWIW, I started exploring and saw the 'Write The Kanji Reading' exercise - the instructions were (too) terse, and even though it did say to enter readings in hiragana, it did not identify and reject the romaji that I entered upon entering it, and so it was not until the end that I found out that all of my input was marked as incorrect, even though the romaji that I entered for my answers would have resulted in maybe 12 correct out of 20. Some of the provided answers were 'broken', in that it reported "Correct Reading: error" for several of them.
Not sure - if I started again would my previous answers be discarded, or would it save them for future review? The latter would probably be more useful than discarding them.
But a quiz of kanji readings without any awareness of (or prompting for) the user's current knowledge level is not particularly useful - with my current level of kanji knowledge, some of the quiz questions were trivially simple while others tested kanji that I have never seen before.
At least some Japanese language study apps (such as Satori Reader and Bunpro, and probably others) have the capability of importing the user's known kanji (for example, via the WaniKani.com API).
Also note that delaying correct/incorrect feedback until all 20 questions have been answered is not a particularly sound teaching/learning exercise (as opposed to getting immediate feedback right at the point of entry). Also - not sure why the 'cheat' of providing a hidden (until the user clicks on the icon) English translation of the sentence is even part of such an exercise - unless it affects the scoring somehow, but that possibility was not even alluded to.
My standard of reference for kanji learning SRS apps is WaniKani - If you're familiar with it, I'm almost done with level 6 (out of 60 levels) - I signed up for a lifetime subscription to WaniKani because, even though I find their mnemonics mostly dumb and useless, so far the app has been effective for me - at least so far, it has been the most useful kanji learning tool that I've ever worked with - in fact, only an Anki deck created by NihongoShark (that had kanji and vocabulary embedded in context within sentences) has even come close to being as effective as WaniKani, although it was not what I would think of as being 'terrific'.
Anyway, I'm not certain whether that kind of feedback is valuable to you or not. Let me know (and if not, what would be more useful for you).
BTW, I'm not sure whether the terms of service would allow it, but the Wanikani community forum has a terrific set of kanji learners of all levels - and could be a rich source of feedback for you, if you were to handle things correctly without violating the terms or spirit of that community.