May 5, 2019

#ask-ih: Do you learn faster by watching or reading?

Sup fella hackers!

Many times I found it time-consuming to watch a 50-60 minutes conference talk or business interview video only to get that 5-minute-long useful content I was looking for.

So I came up with this idea to provide text summaries or bullet points for long-form YT videos or podcasts for people to cut to the chase when they don't have time to consume the entire episode.

It could be a chrome extension that pops up and shows a video summary when people open a YT page (how it works on mobile would be a problem tho). Or maybe a website that people can paste any video link to search and read it's summary first before diving into the whole content.

Would love to have any feedback or general advice!

  1. 5

    It depends, but yeah, I often just want the 5 minute punchline. Not everyone is like that, but I am and it's why I prefer blogging to vlogging, etc.

    1. 2

      Thanks for sharing!

      I watch videos/ listen to podcasts about tech and biz a lot, but don't really have the time to digest them all. I'd use this to save more time for myself.

      And may I ask, if this tool costs $3-5/mo, would you still use it?

      1. 2

        No, but I'm deeply introspective about my time and a cheapskate. As soon as I have to pay for it, I'm committing to a lifestyle of overindulgence of media. I'm at this point now where I listen to podcasts and don't really listen. It's more of a sign that I need to consume less. I can only act on so much information.

  2. 3

    I personally like to read because you have an active mind vs. a passive one when you're digesting information via a video. Not to bash video consumption altogether tho. There are definitely times when all I have the energy for is watching a video.

    Also, that's a great idea OP. I'd use that. Also I like the medium of Chrom extensions. Although not expandable to mobile it scales nicely

  3. 3

    I'd rather reading.

    It makes me interact with my thought.

    1. 1

      I prefer reading as well, but many thoughtful people I'm following tend to put valuable content into videos (for them it's easier than writing to record and share things). That's why I have this idea to make learning easier. May I ask if you'd consider paying a little ($3-5/mo) to use this kind of tool?

      1. 2

        It's difficult to answer but I think I will.

        Audiobot is great to use, it read aloud the uploaded article, and it is so nice.

        So video as well. Are you thinking transcribe the video?

  4. 2

    Yes I agree with you, I learn much better by reading, cause I can pause and type up some notes. With video you have to find the pause button, and it's not easy to skip to the important parts.

  5. 2

    Reading for me, helps me scan and quickly go through a lot of stuff.

  6. 2

    I generally hate learning by watching a video (or a live presentation). I can read many times faster than I can listen, and I'm free to re-read or skim, to refer to diagrams when I want to (not at the presenter's whim), etc.

    I will go to fairly extreme lengths to avoid watching YouTube videos.

    I don't (just) want a summary though, I'd want a complete transcription.

    Re payments, I love the method offered by inkl.com: I'll pay a small amount upfront, and then you can take a per-use fee from my upfront payment every time I use it. For inkl, I pay $10 or $20 upfront, and then each article i actually read is 10c. I like this model so much better than a time-based subscription.

    1. 1

      100% agree on why you hate video learning (videos are good for learning certain subjects tho). I'm planning to provide transcriptions as well because summaries could be subjective when it's generated either by AI or a human.

      The inkl pricing model looks friendly for infrequent users. Will take that into consideration!

    2. 1

      I would love this, in fact i have something similar in my ideas book.

      Repeatedly i find myself in need to a tool like this. And SPOT on @davidarnold Pay per use is an excellent model, i wonder why startups don't use this model as often

      1. 1

        Would love to learn more about your ideas!

        As for the pricing, I'm guessing that startups prefer predictable profits so they ask users to pay a fixed amount per month and it's beneficial for (only) frequent users.

  7. 2

    I usually watch videos first because I'm more of a visual person. Then I'd read the supporting docs afterwards.

    1. 1

      I see people have different learning styles. Also glad to know that summary docs have the potential to serve visual people in a supportive way!

  8. 2

    Generally speaking, I read first, then watch videos for supplement info. Case by case though!

  9. 2

    For myself, it's less about the content format, and more about allocating distraction-free time

    1. 1

      Good point!

      May I ask, besides entertainment, what categories of content would you allocate time for? (entrepreneurship, technology, etc.)

      1. 1

        Currently, I have a singular "Personal Development" time block as part of my weekly routine, spending 4-6 hours on Sunday mornings at my local coffee shop. I primarily use it to further my engineering skills

  10. 2

    I'd go straight to DOING rather than watching or reading. In the DIY process is where you really learn, (it implies reading/watching but 90% of the time small and useful concept pills, rather than one hour conference or a book).

    The only drawback is that you have to create a problem from the things you what to learn, so that you can learn with DIY philosophy.

    1. 1

      Agree on that we learn best by doing. That's part of the reason I wanted to build this tool -- to trim the unnecessary info (like rants or chitchat) in a video but leave only the core concepts so that people can learn faster and start doing things.

  11. 1

    Somewhat related to this idea, specifically around watching those 50-60 minute talks, it would be awesome if there was some kind of tool that could process the text (auto-generated captions?) and create a document outline or table of contents for the video. Bonus for whatever platform it is on if you could click on one of the auto-generated sections and it jumps to that point in the video for you.

    This is probably a really hard problem to solve because it's like one step beyond the language recognition and mapping a flat stream of text back into a structured outline. Picking out the important things and defining headings would be tough.

    One way to potentially get training data for that would be to take blog posts from the internet and strip out the header tags. This would give the flat stream of text as though someone was reading it. Then see if it can learn to identify the separate sections within a blog post and how accurate the naming is.

  12. 1

    This comment was deleted 5 months ago.