A useful feature that I think would make LingQ even more awesome

Basically it’s an “Identifier Extension” used on other websites like Youtube; that displays unknown words and learning words within the contents thumbnail (like it does with a lesson on lingQ).

It would work like this;

  1. LingQ calculates if the displayed content has CC.
  2. If it does, it will show the word stats in the thumbnail. If not it will show “no cc” or be blurred/greyed-out.
  3. The import button can also be incorporated; and can be clicked on the top right and works like it does now.

Why I think this is useful;

  1. Saves time and effort finding suitable material.
  2. Allows for more convenient navigation into finding comprehensible material.
  3. Utilizes general feed on websites more effectively.
  4. Would work better with the import function since the video displayed will show whether or not it has CC. and can import from the thumbnail itself.

Here’s what it might look like;

People may ask; “Why not just use youtube better”, these are my issues;

  1. A video has to be searched in order to show the filter list to apply CC. (+ other search issues)
  2. Content is displayed as an ongoing list and may take a while to find what i’m looking for (unless it’s very specific)
  3. It is unclear if that content is suitable for learning or not (too many/too little lingQ’s)

And also for Caption pop;

  1. A video has to be searched in order to show the content with CC.
  2. SAME AS POINT 2 ABOVE.
  3. SAME AS POINT 3 ABOVE
  4. There is no feed home page like youtube has. (seems to be fixed)

Also, I don’t see how I can use Zero to Hero at all, it returns with very little results.

CONCLUSION

Perhaps people think this is lazy and that all needs to do is “look for what they want to import”. Personally I’m struggling to do that as a beginner to be honest.

In summary, the main idea from this is to fully utilize the feed (calculated from the things I am interesting in/search for anyway) instead of keep having to search, and determine if something has CC to allow content to easily be identified as useful and imported into lingQ.

That would be great. Finding something that is just right really isn’t easy, also if you’re already at an advanced level.

As a LingQ user, I think this would be a great feature, but as a software developer, this idea makes me feel a bit queasy. I’m not familiar with the specifics of how the LingQ importer works and web scraping is not my area of expertise. However, as a general idea, my best guess for how a feature like this would work is as follows:

  1. You would need to generate a list of all the videos displayed on the page you are visiting. They are all included in the HTML file for the page, so this should be easy, except for the fact that YouTube uses infinite scroll for both the homepage and search results. This means that you would have to cap the number of videos in the list (first 10? 50?) and the site would need to be completely reloaded once you scroll past the last video in that list.
  2. For each video in the list, you would need to check if the video has captions available. If so, you would then need to check to make sure the language of the captions matches before scraping the video’s page for the captions.
  3. Once you have the captions, you would need to cross-reference the captions with your LingQ vocabulary (thousands or tens of thousands of words for most people!) to check for the number of blue and yellow words. This is doable, but does take some time, especially when this action is performed several times in a row. (For example, the LingQ homepage tends to lag from performing this calculation on the lessons in your feed, and unlike YouTube, LingQ generally is not loading high-resolution images at the same time.)
  4. After all of this, now your browser can load YouTube (or whatever site) for you to see it. *

Again, scraping is not my area of expertise, so I could be wrong and this could be a trivial process. However, I imagine that at best, YouTube would be annoyingly slow to load, and at worst, it would just not load at all.

I don’t think you are lazy for suggesting this. On the contrary, I’m all for searching for content as efficiently as possible, especially if you have detailed criteria! And I’m not trying to shut you down, because I would also find this really helpful and probably use it just about every day. However, I think you are more likely to meet with success on this by appealing directly to YouTube to include a search filter for captions in a specific language, for example, rather than waiting for the already overworked LingQ team to build a thicc feature like this, which may or may not actually work. I’m sure that YouTube already has a way of filtering content this way for people with screen readers at least, but I’m not aware of a way to do that without cracking into the source code.

In the mean time, good luck with your searching! I hope you find the perfect content soon!

*As an aside, I have no idea whether or not YouTube will allow modifications in the appearance of the site. If not, this definitely wouldn’t be possible :frowning:

As another aside, I have generally had much better success importing captions from Netflix than YouTube, because Netflix allows you to filter by subtitle language and the subtitles will always be high quality. If you don’t use Netflix, I would suggest trying another platform with longer-form content to look for things to import.

1 Like