Ask HN: input on the idea please? | Hacker News Reader

Ask HN: input on the idea please?

21 pointssnitko16 comments

Hi HN. I wouldn't ask you, but I had a pitch about this idea with some tech and non-tech people and all of them said they really like it, so I'm really confused. And also I'm excited as this might be finally the thing I'll be passionately working on. I'm not emotionally invested into it yet, so please crush it if you feel like.

Summary of the idea: a web-service that makes it easy to plan an education program on any discipline and share it with the world. People may search for specific keyword, for example "Ruby On Rails", then they'll be handed a list of plans created by others. A plan is a document that describes what a person should learn (subjects, concepts, books, articles, other plans) in order to become a "novice", then a "mid-level" and finally a "senior" specialist. A plan by default is editable by others (with author's supervision) and others may, thus, contribute and rearrange it.

Full description and profit models: http://docs.google.com/View?id=ddr4mr7n_30d52n2hgw

I think this is definitely an area worthy of taking on, but like another comment said, be prepared for it to take years to mature.

I've thought about this problem a bit as well mainly from the position of "you don't know what you don't know". This has always bothered me in terms of self-motivated learning. Google has changed the world by enabling people to learn almost anything online. There still exists a problem though in not knowing what search terms to enter into google? If I don't know anything about sql and database normalization how am I supposed to use google to find what it is I don't know?

My thinking led to a service somewhat like yours. A kind of collaboratively curated "basecamp" for certain topics. I would suggest concentrating on one specific area of knowledge like "computer science". So for example there would be vague things like "Overview of the internet" which would branch off into what a web app is, html, css, browsers, protocols, what a database is, what is mysql , and eventually "database normalization" etc. I think its useful to create properly formed structures of information. That way when you navigate the information you can navigate back up the tree if you need more prerequisite knowledge and then go further down into specifics once you understand your node. I also like the idea of a an overview "glossary". Don't you think that every discipline has its own language? It's own lexicon of slang and sayings. Even here in hacker news: PG, ASK HN, ror, a/b, seo, CS, machine learning, GAE. I have to be honest here and say that when I first came to HN I didn't think much of it at all because the front page looked like a foreign language to me. "I don't get it" I would think. That's because I was unfamiliar with the language and terms. So I think something as simple as a glossary would make things a lot easier for some people. Database: mysql, normalization, key value store, index, foreign key, tables, rows, locking, read, write, etc. If nothing else it would act as a starting point to start a google search party.

Yes a lofty goal indeed but I think it would be very impactful.

edit:

A lot of people would point to Wikipedia to solve this problem but as I'm sure you've figured out, wiki functions as a database of nuggets of information that possibly link to one another. I'd argue that hierarchal structured information (or what you might call a guided lesson plan) would be a lot better in terms of actually learning the material. Also I imagine a world in which these lesson plans were curated by people who have "been there before". I can't tell you how many times I have studied something in a book, looked it up online, tested it out for myself, spent 2 more weeks thinking about it and perfecting it, AND THEN sometime down the line later, some guy somewhere mentions the same lesson i learned in a 5 minute conversation. He literally tells me everything i know in 5 minutes, because he's been there before. And I could probably do the same for other people regarding my own domain knowledge. Need to know how to print t shirts? I guarantee you I can give you a crash course in an hour that will save you months of trial and error.

I like the idea. I've often thought about this within more narrow verticals (e.g. programming).

I like to learn using iTunes U, but it would be nice to have those video lectures included in a broader curriculum recommendation from somebody whose credentials I trust.

<Off-topic rant for your consideration>: Personally, it would be nice to force me into a certain timeline. Perhaps using some form of calendar style metaphor where I can check things off as I complete them and it lets me know if I'm getting behind (or even how I'm doing compared to the average). Maybe it's just me, but sometimes it is nice to have some enforced discipline when learning. Further to this idea, imagine if an author could create a curriculum and then constrain it to what they considered an optimal timeline; that way a group of learners could be working through the curriculum simultaneously - asking questions in the forum about "this week's task" etc. Or meeting up with other local learners. Then again if you had enough learners participating, there would always be someone at the same stage in the material as you anyway. Maybe I just miss my undergrad years :(

Anyway, I really like the idea; would love to know how you will recruit curriculum authors with the necessary credibility.

I've been kicking around a similar idea, but with a stronger emphasis on the user model as a mentor/apprentice experience. It would be certainly harder to implement than what you are describing, which, and correct me if I'm way off here, would be a set of interactive lesson plans presented in a fashion similar to a wiki where the end-user gains experience and credibility given their progression and rank (or karma, points accumulated) for a set of lessons belonging to a particular category or domain of interest. As they gain more credibility for domain they are granted additional permissions for editing or adding content for that domain.

Is this like a wiki for education plans? Sort of like ehow except for education plans?

That's very simplified. It doesn't look like wiki in my mind at all, though it shares some concepts of wiki, like collaborative editing. The major difference here would be that in wiki anyone could edit and commit, while in this project the author of the plan first reviews the changes made and then accepts them. Alternatively you can clone the whole document and build your own based on it.

Just do it. Like Seth says, there's no Idea Approval Board. Where you run into trouble is when you fall in love with your solution to the exclusion of seeing the actual problem.

I once had a boss/mentor who believed that teachers should have a nation-wide lesson plan repository. He lamented that there was no way to get teachers to use even the most "simple" version control system (e.g. TortoiseSVN). I realize you're doing more than lesson plans, but you may want to consider teachers as a good first market.

That brings me to my next point: make sure you find out who your market is _soon_. This advice is old-hat for HN folks but its worth repeating:

1) Read up on mvp [1]. Create a "smoke test" to see if you're even in the right universe. We could psyche you up and tell you this the best idea ever, but until you verify that with The Internet you don't really learn anything

2) Read Steve Blank's book [2]. Get out of your bubble of encouraging friends and contacts and, say, find some teachers and ask them if they would _pay_ for that sort of service. (Find people who have a problem, know they have a problem, have hacked together a solution, and who are willing to _pay_ for a better solution.)

It's these visionary customers who are really the only ones who can answer this question for you.

[1] http://www.startuplessonslearned.com/2009/08/minimum-viable-...

[2] http://www.cafepress.com/kandsranch

I like the idea. I think key to success of this idea - getting people already know the subject to participate in the service. Their input might be more valuable. So it also brings down to point to ranking various lists available on the same topic. Digg/reddit like voting might be useful here and in turn might improve engagement also.

If I imagine this as the best possible way to develop a curriculum (which it may or may not be), there is still a big problem: learning is driven by curiosity. What is curious to one person is boring to the next so the specific path from "novice" to "expert" is very personal.

It's a curriculum, not a compulsory work load. Also, the curriculum would be updateable. I think it's often the case that somebody wants to go farther into a subject and would like to know what the standard knowledgebase is for that subject.

Contact me. I had another idea for a similar goal but vastly different implementation. Maybe we can create a hybrid or tear down each other's ideas?

Thanks everyone for the replies. I had no chance of replying right away, because I couldn't login to HN (LiveJournal's openid was down).

Sorry to be a broken record, but I always plug David Gelernter's 'tracks and clusters' concept when similar ideas come up:

http://edge.org/q2009/q09_9.html#gelernter

Roger Schank's ideas about just-in-time, story-centric curricula may also be interesting:

http://www.edge.org/q2009/q09_print.html#schank

http://www.rogerschank.com/

I like this idea too. Mind if I steal it? ;)

Also, a thought, if you do start this, be ready for at least 5 years of hard work before you either quit or are a success. This kind of radical new stuff takes time. So that's your 20s right there. (or 30s. or 40s).

Actually I do mind. But it's not like I can sue you or something. Maybe just punch :)

It's a rehash of other established ideas: Get a bunch of people to do it, call yourself a college. Use the credentials of your teaching staff to raise the rates you charge 'students', you have a university.

Who's to say who is the 'expert'? As they say, those that can't do, teach.

You can expect a ton of poorly created curriculum, with the assumption that the expert _might_ fix them at a later date, but without ownership, why bother? And without the front payment, why make the effort?

Reply on news.ycombinator.com