Design a less confusing input method for different writing systems

[danielinux7]

Describe the bug In the "Type what you hear" on an Android phone, the keyboard pops up when choosing a letter from the proposed letters.

To Reproduce Steps to reproduce the behavior:

1. Go to 'librelingo.app and choose start learning Spanish'
2. Click on 'Basics and skip until you get to "Type what you hear"'
3. Choose a letter from the proposed letters.
4. See keyboard popping up.

Expected behavior In the reproduce section, the keyboard shouldn't be popping up in step 4.

Screenshots

Smartphone:

• Device: Xiaomi RedMi Note 8T
• OS: Android v10
• Browser: Chrome v90

[kantord]

Hi, @danielinux7 :wave:, welcome to LibreLingo 😁!

Thank you for reporting this issue.

My question is, what should really be the expected behavior here? To me it seems like maybe the keyboard should have popped up right away and should have stayed there? So like this:

1. Go to 'librelingo.app and choose start learning Spanish'
2. Click on 'Basics and skip until you get to "Type what you hear"'
3. See keyboard popping up.
4. Choose a letter from the proposed letters.

[kantord]

@allcontributors please add @danielinux7 for bugs

[allcontributors[bot]]

@kantord

I've put up a pull request to add @danielinux7! :tada:

[danielinux7]

@kantord I am closing this issue, because I misunderstood the task in hand. I thought the task was to pick the letters in the right order to form the words in the audio.

[kantord]

Thanks @danielinux7.

I think if it was confusing, then that in itself is an issue. Do you have any suggestions as to how to make it less confusing?

[danielinux7]

@kantord I think providing few letters specific to Spanish, then relaying that the user will have a latin keyboard is a partial solution. Probably a generic solution would be to provide all the letters of the uttered words in the task, and disable the keyboard, but this is more of an enhancement than a bug.

[kantord]

hmmm interesting! For me, those special letters are just there for convenience, just in case the user cannot use the proper keyboard layout. (For example on desktop computers it's not as convenient to switch, and requires more setting-up as well)

[danielinux7]

It seems you are referring to a specific region or place, what if the person in question uses Russian, Chinese, indian keyboard? That solution won't work for them, it would be confusing. Also coding wise, it would be easier to just split those words into letters and provide those letters in random to pick from. The solution should be universal to avoid specificity in languages.

[kantord]

I think it's a good idea to have something like what you suggest, however in that case I think I wouldn't spit them into characters necessarily as it would be too annoying.

I think letting the user use their preferred input method to practice, though, is essential.

Maybe the special characters could be hidden behind a popover to make them less confusing?

[danielinux7]

There could be two kinds of tasks; one that splits into letters and another with longer sentences that splits into words. Two different tasks of 'Type what you hear' The keyboard could pop up when the user clicks on the text field, that is the only way that should allow him/her to use their keyboard, because it means they actually want to enter the characters themselves.

Those special characters will add complexity to the application, are you willing to handle all the special characters in every language?

[kantord]

Those special characters will add complexity to the application, are you willing to handle all the special characters in every language?

What added complexity will be added for other languages?

[danielinux7]

A universal solution would be to split words into characters, randomize and provide them for selection. The solution is language independent.

Handling edge cases by providing partially few characters is language dependent, i.e Japanese has 2000 characters, Chinese 50000.

This is an opinion but I could be definitely wrong.

[kantord]

@danielinux7 "splitting into characters" might not necessarily be language-independent. Or at least, it's not always clear how you'd do the split.

• I'm not an expert on this but it seems to me that for a language like Chinese, there's little difference between splitting them into words, so perhaps the solution would not work the same way for each language
• In several languages, the same word could be encoded in different ways. Sometimes there are multiple equivalent ways to encode the same symbol in unicode, and splitting them into characters could result in different splits depending on which type of encoding is used. The inconsistencies could be solved by normalization, however the question is still valid: do you want to split accents from base characters or not? Several languages use separate keys for their accents. For example, for some accented characters on a Czech keyboard, you have to type the accent with a separate key. This question is even more acute for a language like Hindi, where basically the same system is used to type vowels, as generally vowels don't have separate characters and are expressed with modifying characters. Thus I'm not sure there's a universal way to split up any string into characters in a way that will make sense for any language
• The above issue is made worse by the fact that in some languages it might be possible that you even put multiple accents on the same character
• I think the UX might be worse, especially for long sentences. It's better to be used to a keyboard layout that always stays the same

Handling edge cases by providing partially few characters is language dependent, i.e Japanese has 2000 characters, Chinese 50000.

If I understand your point correctly. It's definitely a fair point that the current system is not a really universal solution either, it's just something that kinda works for some languages as a special tool to extend your existing keyboard layout on your phone. It could probably be hidden behind a button or something. The way I see it, choosing characters from the screen is just not going to be a good option, so it seems like we definitely need to let courses to customize this to a certain level.

So for me it boils down to choosing the minimum number of configuration that work well for the maximum number of languages. The way I see it:

• For language pairs where they use the same alphabet (but maybe with small differences) the current solution should work, and using the correct native keyboard layout on your phone should be encouraged.
• For languages where it's 2 different writing systems, we could probably let the course authors manually slice up sentences into relatively small chunks that could be used as "chips" or the chips challenges. That being said even in that case we could give the option to the user to start using the keyboard available on their phone for example, so that they learn can start learning the skills they'll need to start texting people and such

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.