4 ms·

You say this snarkily, but I don’t see what is wrong with someone growing up and wanting to try to use coding to solve problems relevant to their stage of life.

You say this snarkily, but I don’t see what is wrong with someone growing up and wanting to try to use coding to solve problems relevant to their stage of life. It turns out some problems are complex, but so what? I’ve been working on “programming interfaces for beginners” for 12 years and still haven’t succeeded. That’s not because I’m dumb, nor is everyone who thought it would be easy and failed. It’s because there are building blocks that don’t exist yet, so when you try to make the thing you end up getting distracted making a building block—or eight—and then your schedule gets shot and this kills the project. But there’s no shame in that. If we’re lucky, you made a new building block which will eventually mature, in one form or another, maybe just in conversations with other devs who are closer to cracking that low level problem. And then guess what... time is ripe for a hubristic person, young or old, to step up and try their hand at your original problem, now with one more building block. There’s now a 5% higher chance of success. Eventually someone will hit a threshold. Is your post meant to be more than classic HN stop energy?

djb_hackernews 9y ago
Not meant to be snarky at all. Though I do wish my comment wasn't the top voted one. People, please upvote the other on topic comments or have the mods bury it. FWIW this is basically my path. I've done pretty much all of them to some degree and a few in earnest. In fact I'm working on my own language learning "tool" while I wait around for #10.
- avn2109 9y ago
  Any discussion of computerized language learning is woefully incomplete without a mention of [0], which has an interesting combination of properties. It is: * A) run by people who suck at SEO and marketing * B) technologically and graphic-design wise really janky * C) harder and less fun than e.g. Duolingo * and D) drastically, overwhelmingly more effective than every other system I've ever seen at actually teaching you a language (for people with almost all common linguistic-cognition neurotypes aka nearly everyone) The first 3 attributes mean that almost nobody has heard of it/uses it. The downside is that for most languages they didn't build out enough curriculum to get you to fluency. It'll take you roughly a third of the way there. I'm building an open source clone (with longer curriculum) for the other two thirds of the way. Also, L101 should be paired with Yabla, which will train the skill of listening-and-understanding in your target language. I'm not affiliated with [0] in any way other than by being a happy customer and by the fact that I'm writing an open source clone of their stuff. [0] https://language101.com/ https://language101.com/
- dang 9y ago
  I agree with you that topcommentness can subtly but immensely change a comment's meaning, through no fault of the commenter or even any individual upvoter.
funkaster 9y ago
The problem is the attitude "this problem is unsolved and no-one seems to be working on it. I'll fix it, even though I'm not an expert in the domain... because coding fix everything". Don't get me wrong, this is a very common problem for engineers (I'm an engineer and I fall into that trap every now and then). But from the post, what's lacking is actually talking and working head-head with teachers and linguists and people doing language learning research, because, you know... these people have actually spent their whole life trying to teach other people languages. So they might know a thing or two, that an engineer trying to teach himself a language might be missing.
- schacon 9y ago
  I'm not sure where you get this. I studied linguistics in college and one of my cofounders got her degree in linguistics from Harvard. Another of our employees is a computational linguist from Cambridge. Additionally, all of the curriculum and much of the course methodology itself was designed by teachers who have been teaching languages both online and in the classroom for years. We have additionally consulted active linguists and educational psychology researchers to see if there is anything else we should be looking at in their fields that we can apply. This post was about how my understanding of the product I felt I wanted as a customer was formed. Linguists and teachers built the curriculum and work with the engineers here to see how we can apply the ways they want to teach.
  - funkaster 9y ago
    If that's the case, then I'm sorry I got the wrong impression, but it was the tone on your post that led me to that. Before your reply I took a look at the about page on your site and saw the diverse team. You should've give them more credit to them and how you searched for that type of support when building the app. Otherwise it's very easy to reach a similar conclusion to the one I got.
- panglott 9y ago
  I think the problem with this view is that academic linguistics has a limited amount of useful information to tell people who are trying to learn a language. Absolutely people learning a language need informed linguistic advice. And our understanding of language learning is much less rudimentary than it used to be. But linguistics is a multidisciplinary, multiparadigmatic field in which there is serious disagreement over some very basic questions. Dost of the good language apps definitely include that. But a lot of language learning is about keeping up motivation and finding ways to access and practice the target language more often, to bring down the barriers to using and communicating in the language. Good software tools can help there, because we can't all afford personal tutors at $30/hour.

djb_hackernews 9y ago

Not meant to be snarky at all. Though I do wish my comment wasn't the top voted one. People, please upvote the other on topic comments or have the mods bury it. FWIW this is basically my path. I've done pretty much all of them to some degree and a few in earnest. In fact I'm working on my own language learning "tool" while I wait around for #10.

avn2109 9y ago
Any discussion of computerized language learning is woefully incomplete without a mention of [0], which has an interesting combination of properties. It is: * A) run by people who suck at SEO and marketing * B) technologically and graphic-design wise really janky * C) harder and less fun than e.g. Duolingo * and D) drastically, overwhelmingly more effective than every other system I've ever seen at actually teaching you a language (for people with almost all common linguistic-cognition neurotypes aka nearly everyone) The first 3 attributes mean that almost nobody has heard of it/uses it. The downside is that for most languages they didn't build out enough curriculum to get you to fluency. It'll take you roughly a third of the way there. I'm building an open source clone (with longer curriculum) for the other two thirds of the way. Also, L101 should be paired with Yabla, which will train the skill of listening-and-understanding in your target language. I'm not affiliated with [0] in any way other than by being a happy customer and by the fact that I'm writing an open source clone of their stuff. [0] https://language101.com/ https://language101.com/
dang 9y ago
I agree with you that topcommentness can subtly but immensely change a comment's meaning, through no fault of the commenter or even any individual upvoter.

funkaster 9y ago

The problem is the attitude "this problem is unsolved and no-one seems to be working on it. I'll fix it, even though I'm not an expert in the domain... because coding fix everything". Don't get me wrong, this is a very common problem for engineers (I'm an engineer and I fall into that trap every now and then). But from the post, what's lacking is actually talking and working head-head with teachers and linguists and people doing language learning research, because, you know... these people have actually spent their whole life trying to teach other people languages. So they might know a thing or two, that an engineer trying to teach himself a language might be missing.

schacon 9y ago
I'm not sure where you get this. I studied linguistics in college and one of my cofounders got her degree in linguistics from Harvard. Another of our employees is a computational linguist from Cambridge. Additionally, all of the curriculum and much of the course methodology itself was designed by teachers who have been teaching languages both online and in the classroom for years. We have additionally consulted active linguists and educational psychology researchers to see if there is anything else we should be looking at in their fields that we can apply. This post was about how my understanding of the product I felt I wanted as a customer was formed. Linguists and teachers built the curriculum and work with the engineers here to see how we can apply the ways they want to teach.
- funkaster 9y ago
  If that's the case, then I'm sorry I got the wrong impression, but it was the tone on your post that led me to that. Before your reply I took a look at the about page on your site and saw the diverse team. You should've give them more credit to them and how you searched for that type of support when building the app. Otherwise it's very easy to reach a similar conclusion to the one I got.
panglott 9y ago
I think the problem with this view is that academic linguistics has a limited amount of useful information to tell people who are trying to learn a language. Absolutely people learning a language need informed linguistic advice. And our understanding of language learning is much less rudimentary than it used to be. But linguistics is a multidisciplinary, multiparadigmatic field in which there is serious disagreement over some very basic questions. Dost of the good language apps definitely include that. But a lot of language learning is about keeping up motivation and finding ways to access and practice the target language more often, to bring down the barriers to using and communicating in the language. Good software tools can help there, because we can't all afford personal tutors at $30/hour.