4 ms·

This has the potential to become a cool app but I think it should differentiate from other offerings in the features it provides to the user and not in the way

by tiplus 11y ago

This has the potential to become a cool app but I think it should differentiate from other offerings in the features it provides to the user and not in the way the backend is implemented. You might want to tell people s.th. along the lines of 'look, my deluxe new AI gives you annotated solutions for hard math problem X which no other app can help you with...' I am sure you are aware of http://math-42.com http://math-42.com which is quite similar (developed by high school kids, afaik). They claim to use their own math AI, too. It might be very tough to compete with them without offering additional features. Could you modify your AI to cover math problems which I cannot practice with math-42? Or maybe add features such as formula recognition with your smart phone, etc.

S4M 11y ago
Hey, thanks for the comment! I wasn't aware of math-42, thanks for pointing it out. From looking at their websites, they seem to be mainly focused on solving exercises for students while I correct students work (they have a beta training that might do that, but that's not what they are currently focusing on). > Could you modify your AI to cover math problems which I cannot practice with math-42? Or maybe add features such as formula recognition with your smart phone, etc. Yes, that's exactly what's on my pipeline. I plan to cover everything in high school, ideally helping the students to prepare for exams at the end of high school. > They claim to use their own math AI, too. It might be very tough to compete with them without offering additional features. Honestly, education is a huge space and a competitor is the last of my worries (I am more concerned about getting users first), and they are many apps or website that already that are here to help students in math.
- brudgers 11y ago
  Is there academic research underpinning the pedagogical approach? Does the underlying technology use a domain specific language in the vein of Mathmatica/Wolfram?
  - S4M 11y ago
    > Is there academic research underpinning the pedagogical approach? I can't quote direct articles but I have seen some trends in trying to make education in general more interactive. > Does the underlying technology use a domain specific language in the vein of Mathmatica/Wolfram? Yes, I wrote (and am still writing, in fact) my own symbolic algebra software. I tried to use some open source CAS (Maxima and Sympy) but I found that the way theirs codes are structured doesn't allow to offer hints on how to solve an exercise, without putting lots of work in it.
    - brudgers 11y ago
      Why did you choose middle school mathematics in particular or education in general as "the killer app" for the symbolic algebra software?
      - S4M 11y ago
        Two reasons: - I originally wanted to build a symbolic algebra software to compete with Wolfram's Mathematica but realized it would require too much work for a solo founder. - Math had always been my favorite subject and I saw studies about the decline of math level in first world countries so I thought I could try to help.
        brudgers 11y ago
        I believe education is a commodity market. Sales happen in bulk and large sales via public bids, government allocations, and lobbying. I suspect that in the first world, textbook companies are likely to be harder competitors than Wolfram Research: the market is a finite pie and the businesses are relatively low margin. I don't think the technical challenge of building a symbolic algebra problem changes much with two cofounders, i.e. if there was a business person one person would still be writing all the code. Which is to say, developing a symbolic algebra package is audacious no matter how it gets sliced. The solo founder challenges are not so much technical as social and psychological. The educational angle means piling the study of pedagogical research and analysis of pedagogical methods on top of all the business stuff. Because middle school age children are minors it also means it's going to be hard to get direct access to end users, comply with privacy laws, and separate willing use from mandated use.

by tiplus 11y ago

S4M 11y ago

Hey, thanks for the comment! I wasn't aware of math-42, thanks for pointing it out. From looking at their websites, they seem to be mainly focused on solving exercises for students while I correct students work (they have a beta training that might do that, but that's not what they are currently focusing on). > Could you modify your AI to cover math problems which I cannot practice with math-42? Or maybe add features such as formula recognition with your smart phone, etc. Yes, that's exactly what's on my pipeline. I plan to cover everything in high school, ideally helping the students to prepare for exams at the end of high school. > They claim to use their own math AI, too. It might be very tough to compete with them without offering additional features. Honestly, education is a huge space and a competitor is the last of my worries (I am more concerned about getting users first), and they are many apps or website that already that are here to help students in math.

brudgers 11y ago
Is there academic research underpinning the pedagogical approach? Does the underlying technology use a domain specific language in the vein of Mathmatica/Wolfram?
- S4M 11y ago
  > Is there academic research underpinning the pedagogical approach? I can't quote direct articles but I have seen some trends in trying to make education in general more interactive. > Does the underlying technology use a domain specific language in the vein of Mathmatica/Wolfram? Yes, I wrote (and am still writing, in fact) my own symbolic algebra software. I tried to use some open source CAS (Maxima and Sympy) but I found that the way theirs codes are structured doesn't allow to offer hints on how to solve an exercise, without putting lots of work in it.
  - brudgers 11y ago
    Why did you choose middle school mathematics in particular or education in general as "the killer app" for the symbolic algebra software?
    - S4M 11y ago
      Two reasons: - I originally wanted to build a symbolic algebra software to compete with Wolfram's Mathematica but realized it would require too much work for a solo founder. - Math had always been my favorite subject and I saw studies about the decline of math level in first world countries so I thought I could try to help.
      - brudgers 11y ago
        I believe education is a commodity market. Sales happen in bulk and large sales via public bids, government allocations, and lobbying. I suspect that in the first world, textbook companies are likely to be harder competitors than Wolfram Research: the market is a finite pie and the businesses are relatively low margin. I don't think the technical challenge of building a symbolic algebra problem changes much with two cofounders, i.e. if there was a business person one person would still be writing all the code. Which is to say, developing a symbolic algebra package is audacious no matter how it gets sliced. The solo founder challenges are not so much technical as social and psychological. The educational angle means piling the study of pedagogical research and analysis of pedagogical methods on top of all the business stuff. Because middle school age children are minors it also means it's going to be hard to get direct access to end users, comply with privacy laws, and separate willing use from mandated use.