Ask HN: My education startup idea, tell me why it sucks

2 points by havoc2005 ↗ HN
So I have many ideas and have had many ideas throughout the past week. I thought one way to finally stick to one would be to ask HN and get input on why it sucks in order to understand how to better it.

I have wasted a lot of time reading books that were on top reader lists for business, programming, design, etc. Only to find the same information in another highly rated book rewritten or condensed.

I want people to be able to minimize the amount of books they need to read and to maximize the amount of knowledge they obtain by suggesting books to a user based on what it is they are trying to learn and then giving them a sequence of other books to read in an order to further their knowledge in said subject.

What does HN think of this and are there any similar sites/competitors?

15 comments

[ 2.8 ms ] story [ 40.0 ms ] thread
Is this feasible? Is it something you can control, or is this largely controlled by the content authors choose to put into their books?

Also, just a bit of advice - not having competitors can be bad. It could indicate that you're trying to build a company in a market which can't exist because it's just not viable.

Competitors seem to be sites that simply suggest similar books

whatshouldireadnet.com reader2.com shelfari.com and of course Amazon

What they do not do is give you an orderly list of books and an order to read them in to maximize knowledge and avoid excessive repetition.

I was thinking of having it controlled through votes, up and down. So a user who just obtained say an IT certification such as the CISSP can list the books he read and the order for the search term "cissp". Then if one of the books is heavily downvoted it would be removed from the list. This way dated material will be removed and replaced with up to date books.

My question to is... how will this be monetized?

You aren't creating any content here (at least in the way you have positioned it). So, you'll be making a very small slice per sale you provide via your website. But what is to stop people from simply going to your website, seeing the order in which you've outlined the books and then not going through your site to buy (or clicking through links to buy, however you approach it)?

I don't think the problem you are trying to solve is bad. However, I'm not sure if what you have posed is the best solution.

My initial thought was to monetize through affiliate sales but you're right. I have to find another way to monetize.
I see 2 problems with this. One: for many people, books will not be the best (or cheapest) resource. How will your metric rank items like websites, college lectures series, things like Khan, and other non-linear content?

Two: How specific will you make the "goals"? If an example goal is "To be a doctor"...the route there is immense and incalculable. If you do something like "brain surgeon"...there are many subgoals even to that specific profession.

OK, three: i don't think you have the expertise to say that there's one best set of books or learning material. everyone learns differently. That's part of the reason why you found so many books in the first place

Education is a terrible market to be in. You'll be fighting the tide at every step along the way. It'll be difficult to monetize and even if you do, the cost of education is skyrocketing which leaves less and less money each year for discretionary spending. So you're essentially in a shrinking market to begin with.
Book lists on Amazon by those with good domain knowledge might be somewhat similar to this? (or for programming maybe the recommended short reading lists often posted on HN). And I think that was the objective of many of the any subject Bible type books that attempted to be comprehensive / condensed references in a single book.

Just a thought, maybe you tackle the problem instead at the recommendation level - rank books based on an aggregate of recommendation lists, sales rank, etc, to identify 'good' core books for any subject, as well identify those not often recommended that might be too 'fluffy' or what folks consider 'core' recommendations for a subject.

If it actually worked, it would be very useful. But the problem is it will take a while to test if it works or not. Can you make it work on a shorter time scale, say for something which could be done faster than reading a set of books?
This is a good question, I could do a shorter time scale by pulling specific chapters of every book and then combining them into a slimmed down reading list. Say, from book one read chapter 1-3, book 2 chapter 5 and 6, book 3, chapter 9-10 and that will give you x knowledge.
Why would I choose your site for suggestions over Amazon.com which has a vast catalog, decent search, user reviews, and a review rating system?
Our suggestions will come from users who have already read the books and have determined the bare minimum you will need to read to maximize knowledge.

Amazon suggestions simply basis off of ratings and copies sold so you might buy two highly suggested books off of amazon only to find out that it's the same materials covered, written by two different authors.

>"Our suggestions will come from users who have already read the books and have determined the bare minimum you will need to read to maximize knowledge."

1. How will you overcome the chicken and egg problem and attract users who have read the books?

2. How will you screen users to determine that they have the necessary domain expertise to determine "knowledge maximization?"

3. How can a generic list of materials be assumed to apply to individuals with diverse backgrounds?

Finally, as soon as you see reading two books on the same topic as a problem, you have somewhat divorced yourself from the kind of people who typically read books - particularly readers in pursuit of technical expertise.

1. How will you overcome the chicken and egg problem and attract users who have read the books?

A. I was going to start out in a niche that I have familiarity with, IT certifications. I would like to reach out to three primary IT certification forums where users ask for advice about which books they should read to obtain a specific certification. I could use them as a test and pivot as needed depending on the outcome.

2. How will you screen users to determine that they have the necessary domain expertise to determine "knowledge maximization?"

A. Screening is going to be done through voting on the list. So there would be one list per subject, then users can vote up or down on any book in the list to change it's ranking.

3. How can a generic list of materials be assumed to apply to individuals with diverse backgrounds?

A. They can't. The hope is that the list will give the user a better idea of what to buy/read.

The reason I want to do this is because it is a problem I face all the time. When I want to learn a new subject such as Malware analysis. I search Amazon and get a fairly large list of books to choose from. Now I try to choose them by basing my decision on the amount of customer reviews, ratings and comments. What if we could simplify that and allow those same users to rank the books so you wont have to make a determination based on looking at three meterics for every book you want to take into consideration?

4. Finally, as soon as you see reading two books on the same topic as a problem, you have somewhat divorced yourself from the kind of people who typically read books - particularly readers in pursuit of technical expertise.

A. I don't see reading of two books on the same topic as a problem but the problem lies in the materials within the book. For example chapter 1-3 in one book and chapters 4-6 in another book might be the same material, so I want to avoid that. So I want to minimize the material you read within the books.

How is that better than a test centered Q and A format - i.e. a Stack Exchange clone?