Every business starts with an idea. Not every idea turns into a business.
The purpose is to filter, refine and develop ideas into a fundable business. Apparently, many people have spent time thinking about this. Evan Williams article "Will it Fly" is a useful one to start. The following is the methodology I will embark for that purpose.
1. Passion
It's going to be a tough journey. You may as well do something that you care deeply about. I believe that education enables and enriches people; thus education changes lives. People change the world; thus education changes the world. I am going to stick to that.
2. Market
The biggest market, or rather the biggest potential impact in education is where the most people are. That is clearly China. Just English language training itself is projected to reach $13.6 Billion dollars in 2014.
3. Making money
Sustained impact comes from sustainable growth. Sustainable growth requires valid, proven business model, where added value per users is larger than average acquisition cost per user. Unless incoming traffic is astronomically high, advertising is not going to be the solution. It needs to be one of transaction, subscription, or lead generation.
4. Validation
The idea needs to be validated, i.e. there is some dude, who you can name and locate, actually willing to reach the wallet to get some version of your product offering. Find the dude and offer him "a solution". A great example, which was used 10 years ago for Zappos was, to just take pictures of shoes from other people's stores and offer them online.
5. Secret sause
Is it possible to build one aspect of the solution exceptionally well so that nobody can compete with. It must be a dimension that real customers actually care about.
6. Scalability
How difficult to scale the operation, hardware cost, salesforce build up, customer support team?
The ideas will be going through this methodology, filtered, refined, and developed. The eventual results will be published in a later post.
Read more...
Friday, July 01, 2011
Develop An Idea Into Fundable Business
Thursday, June 30, 2011
Founder Institute Learning: Team Work and life
I was excited to enroll the Founder Institute Silicon Valley Summer semester. Only a couple of days into the program, I already feel learning a ton.
Everyone dies if the team die
Everyone knows the importance of a team, complementary skills and all that, but only when you experience a live or die situation do you realize what it truly means for a team to band together.
I was a few minutes late to show up at the second session, and looking forward to the next few hours of adventures, only to be greeted, "Why do you show up here? You dropped out!". I was stunned, "What the ...?" I realized later that, only 30 minutes earlier, some sophisticated issues triggered an email to Adeo indicating that a member on my founding team is dropping out, and that caused me to drop out as well immediately after. I hassled in the next hour or so trying reach my team with emails, phone calls, whatever communication channel I could get my hands on, and eventually we reached an agreement. This program is important for us and for the company, and we need to do whatever it takes together to stay in.
The environment is brutal out there. Unless everyone holds up, the whole team sinks. You may still die even if you had done everything right yourself.
Paying a price for screwing up
Life is unfair sometimes, but it's brutally fair in business. Whatever you screw up will come back and you have to make up for it. No short cuts. No second chance. Learn from mistakes but never repeat a mistake.
We were hoping to negociate our way back in, but it seems clear that we have to give a one-minute pitch on the spot. If the average score is 2.5 or above, we stay, otherwise, we're out. We'd better put our act together.
Live or die in one minute
You have to deliver under huge pressure. Just one minute, any mistake in execution and you die. That's life.
I delivered. Even nurvous as hell, I managed to cover the essential elements of a good business idea and convinced the three mentors that it's a 3+ pitch. We stay.
It turned out that a little bit of luck does strike, only when you expect it the least. At the Zuckerburg Bar after the session, my team member struck a genius idea, i.e. asking Adeo to credit the 3-score pitch to the team, even though I was not "officially in" the program when I delivered the pitch. Adeo decided to flip a coin, and I called Head, and it was Head. Cheers, team!
Only a beginning. More adventures to come.
Read more...
Tuesday, February 22, 2011
A Hypothesis of Brain Learning Based on Scale-Free Neural Networks
![]() |
| A simple neural net model I wrote sometime back |
The basic functioning of the brain is based on Hawkins' memory prediction framework [see book "On Intelligence"]. However, I propose that the neural cells in the neocortex are connected as a scale-free network, rather than a static hierarchical structured outlined by Hawkins. Under this assumption, neural cells are not connected at birth through a "wiring diagram" found in our DNA. Rather, the connections are formed dynamically throughout our human lives, especially during infancy.
Information is stored in the brain as patterns of connections among neuron cells. Some neurons are more capable in attracting new links than others. With these two properties, growth and preferential attachment, I believe that neurons form a scale-free network, i.e. the distribution of links per node follow a power law [see book "Linked"]. Thus the neural network inherits all the fundamental properties of scale-free networks, such as robustness, and vulnerability. (Evidences of scale-free cortical network have been found recently by researchers.)
Remembering
Memory is stored as relations of sequential pattens. We remember by making associations of pattens that appear together spacially and temporally. When a child first see a triangular object, she is aware that this is an actual object because the pixels of the triangle (or rather pattens of neuron excitations) always move together in our retina each time our eyes samples the world. These pixels are associated with each other. At the same time, the triangle is also associated with the background, the present time, its color, the sound it makes, the lighting, the people around, and everything else that are happening around her at that moment. All these associations become connections between the neurons representing these excitation patterns.
The association mechanism is based on Hebbian learning, which can be simplistically summarized as cells that fire together, wire together.
Learning
Going back to the triangle analogy. The child is aware of the triangular object, yet she has no idea what object it is, until someone tells her that this particular triangular-shaped object is called a triangle. That is when the neurons representing the triangle are associated with name "triangle", which is by itself a set of neurons representing the actually symbol of the word. These neurons are also connected to other neurons that represent the sounding of the word "triangle", and any concrete facts associated with it. Learning happens when any new association is made.
Hawkins argues that during learning, some cells "learn" to fire when lower-level memories learn a sequence of patterns. These cells are passed to higher-level as abstract "names" of the detailed pattens. However he did not layout how these cells are selected in the first place. I think the naming is a natural result of association rather than explicit cell selection.
This explains why some cells are able to attract more links than others. The cells that represent abstract shapes, forms and concepts have a much larger probability to be associated with other cells. Whenever something happens, whether an image appears, a melody of music rings, or a train of thought proceeds, the general conceptual cells will fire if these concrete events fall into their category. Whenever this firing occurs, these conceptual cells make new associations.
Classifying
According to many studies, the brain learns to clasify information into categories. As information is passed from lower to higher levers of the memory hierarchy, more and more details are filtered out, forming abstraction. I think that this classification is a natural result of the threshold logic built into each and every neuronal connection. Each synapse collects electronic signal from all connections along its extension. Whenever the collective strength of the electronic signal become larger than its built-in threshold, it fires. When it fires, it propagates the its own firing power to all connections along its synapses.
Connectionists
This is in some sense similar to the "connectionists" theory of brain learning. However there is a fundamental difference. Rather than dividing the brain into functional modules, or sub-networks, each responsible for one particular domain, such as name networks and image networks, I believe all neurons are able to connect to any other neuron within their physical reach. I believe that neurons make connections, or associations, dynamically according to the information provided by all sensory input.
To be continued, on Hyppocampus. Read more...
Thursday, February 17, 2011
Digging Into Google Public Data Explorer
Google public data explorer seems to be a very useful tool. I generated this graph below, showing which cities in the Santa Clara county of California has seen increase of Asian population, as sampled from the ethnicity of enrolled students.
No surprise in seeing Asian population increase in "good-school" districts like Cupertino and Fremont, but interesting enough, what is Orchard Elementary, which has shot from 20% to more than 40% and emerge as one of the heavily populated Asian communities in the past 5 years? Guessed right, that's the North Valley, San Jose area, which saw the appearance of a new Costco, Lowes, tons of new town-houses, and a new shopping center currently under construction.
It would be even more interesting to use the google API (Dataset Publishing Language) to upload my various junk data and see how they visualize. :)
Read more...
Saturday, February 12, 2011
Idea 50: On Demand Boolean Satisfiability Cloud Solver
The Boolean satisfiability problem, though barely encountered in every lives, has wide applications in software engineering and mathematical optimizations, in particular the problem of formally verifying the correctness of software and hardware. See the wikipedia page for an introduction, and paper written by Joao Marques-Silva on "Practical applications of boolean satisfiability".
The idea is to provide a web service, such that anyone can input a Boolean expression in terms of conjunctive normal form (CNF), and instantaneousness obtain the result of whether the expression is satisfiable. For example, input the following expression:
(x1 + x5 + ~x3) (~x6 + ~x2) (x2 + x3 + x4)where '+' represents Boolean OR, and '~' represents Boolean complement; the result will be a satisfying assignment (one of many) to the 6 variables involved.
In addition to HTML web interface with manual input, a developer's API service will be provided, where software programs can make remote requests to the server and supply an encrypted string of Boolean expressions, while the server respond by solving the problem instance, a satisfying assignment in the satisfiable case and a simple "no" in the unsatisfiable case. In some cases, delay may be encountered if the problem instance is large and requires large computation time.
The business model is to charge based on the bandwidth of API usage, or the amount of CPU cycles involved.
Update: the SAT solver itself could be implemented using current programming language Erlang and deployed on a large set of computer nodes. Read more...
Tuesday, February 08, 2011
Idea 49: Live Web Interactions With Video Chat
The popularity of Chatroulette shows that plenty of people on the Internet are eager and ready for live video interactions with another stranger. Why not? People are social animals.
Tuesday, January 25, 2011
Idea 48: Keyword Group Chat - Real Time Twitter / Quora
Please let me know if you are interested in implementing such an app.
Tuesday, January 11, 2011
North-America Chinese Professional Business Plan Competition
The UCAHP Chinese professional business plan competition was held on Jan. 9th. 20 startup teams, selected among more than 100 submissions, presented their ideas and business plan, in front of judge panel consisting of top-tier VC's and angel investors, such as Northern Light, GSR Ventures, West Summit Capital, NEA, Sierra Ventures, etc.
- Zebra - multi-media storage and bandwidth optimization
- PalMap - indoor map for shopping complex and convention centers
- BCBM - converting low-value coal into natural gas
- Clean Solar - Titanium dioxide solar power panel
- HealO - wound care chamber treatment
- Deep technology accumulation in solving a bottle-neck problem in an established market;
- Creative user experience, or business model, that acquires user base rapidly in a fast growing new market;
Thursday, December 16, 2010
China Fast Development Represents Huge Opportunity
After two months in China, the country where I grew up until 12 years ago, I feel anti-culture shock. It has changed so much, the bullet trains, the highways, the constructions of new cities everywhere, high rises, etc., and the way people live and talk, you don't feel it unless you really stay there for extended periods of time. For this time, 10 cities across the country was good enough for me.
Saturday, September 26, 2009
The Perfect Pitch - Biz Idea In 140 Characters
The business idea competition on Twitter hosted by @perfectbusiness completed. I covered this in an earlier post about what to do with your startup ideas.
Thousands of idea was Twitted in the short period of time, many of which are extraordinarily bright ideas. The final winner was indeed an excellent one, quoted here:
@perfectbusiness #micropitch Power from water: turbines safe for fish, drop-in w/no dams, 1M+ sites, helps 1B+ people www.hydrovolts.com
My idea 28 on mobile route sharing and smart calpooling was featured in the top 10 runner up. The full list is here.
Thank you @perfectbusiness! This gives me enough confidence to pursue the idea further and continue this blog with even better ideas.
image credit Sir Richard Branson & PerfectBusiness
Read more...
Saturday, September 05, 2009
Entrepreneurs: Be An Enabler

Having a brilliant idea is still far from real entrepreneurship; striving to provide real value to real customers is.
One efficient way to do that is to be an enabler, an enabler in technology, relationship management, customer service, supply chain, or whatever area that is critical for your customers to do their business.
1. Develop a process, or as people in academia like to call, methodology.
It's better to know a whole lot about one thing, than a little bit of everything. Any Ph. D. graduate will tell you that a methodology with good theory can save a lot of hard work. Be an expert. Use the expertise to invent new ways of doing things, selling staff, or managing operations, not just a little better, but 10 times or 100 times better than the ordinary.
Sometimes, it requires a few genius minds to work things out. Just check the TED talks for some mind blowing new inventions, like the wireless power transmission; or the processor with order of magnitude less power consumption from PA Semi (accquired by Apple).
Other times, it's just a bag of accumulated tips and tricks. John Chow, like other probloggers, has developed a system of how to make a living through blogging, and enjoys a good life style teaching others doing the same.
2. Develop an algorithm to automate the process.
Don't stop at the methodology. Automate it. Turn it on and forget it, if you can.
Google is a great example of process automation. You hardly need to talk to a real person when dealing with Google services. It uses algorithms to figure out where the hotspots on the web-sphere are; uses algorithms to match buyers and sellers through keywords; uses algorithms to pull and push information to just about every corner of the web.
The electronic design automation (EDA) industry itself is enabling semiconductor design companies to tape out their chips faster, cheaper, with better quality. It emerged from simple drawing tools like magic and zero revenue in the early 80's, to now a $4 billion industry (estimate).
3. Duplicate the process.
With a methodology and automated process, all that's left is to enable everyone to do the same, so the entire world runs more efficiently. There goes your entrepreneurship dream and global impact.
Just some thoughts. If you liked it, good, subscribe the newsletter for more exciting updates; otherwise, let me know and let's start a conversation.
image credit estherase
Read more...
Thursday, September 03, 2009
Idea 47: Voice Enabled Auto Tweet

I like one thing about human nature that is bringing a lot of motivation for technology innovations -- laziness! Yes, we are lazy, and we invent things to justify being lazy, which is good.
Today's idea is for all those lazy Twitter-holics. You don't have to type anything on the keyboard in order to Tweet your status. Just speak to the application on iPhone, or other mobile devices; the voice recognition software will translate the sentences to text, making sure it's within the 140 character limit, then transmit to Twitter server.
What would make it more interesting is, if it can understand emotions, and transcribe that into Tweets as well. Make a laugh, now you have "HAHAHA", or ":D" in your Tweets. That would be fun.
Or, how about automatically discover URL links? Whenever you say "Guy Kawasaki blog", it finds the URL and brings that link into your Tweet.
Lazy is good; lazy motivates innovation.
image credit Louise Lazell
Read more...
Tuesday, September 01, 2009
How To Generate New Ideas - Part 1
In this intriguing video presentation at MIT, Prof. Patrick Winston summarizes his recent research work on how to make computer smarter by developing algorithms to learn basic reasoning and associations, which he call "common senses".
The key insight is that creative thinking process of us humans, is not a linear progression of words or ideas, but rather a complex combination of images, languages, symbols, dialogues, and interactive feedbacks.
He summarizes four things to do in order to become smarter, which I agree and think are also the key points in coming up with creative new ideas.
1. Take notes - The act of taking notes forces us to reason about the thought, and use the logic of language to formalize the thought. With an iPhone in my pocket, it is now easier than ever for me to take notes anywhere, any time.
2. Draw pictures - We think with images. Even with things we have never experienced, we use our imagination to picturize sequences of images, no matter how blurry they might be.
3. Talk and imagine - We learn by talking. There are certain logic reasoning skills we acquire in our childhood only after learning how to talk. Talking not just solicits feedback of your ideas from others, but yourself also.
4. Tell stories - Life consists of stories. Ideas are illustrated best with real-life stories.
The video is rather long, but worthwhile to go through. If you're pressed on time, then check out the conclusions section, at around 41 minutes into the presentation.
video source MIT world
Read more...
Sunday, August 30, 2009
Idea 46: Free Local Public Transportations

Today, I took the ferry between Wall Street and Staten Island. The view was simply amazing, blue waters and city landscape, under the bright sparsely decorated sky. I would say this is one of the must-do's for anyone visiting New York City.
Even more surprised to me was that the ferry is completely free, running every 30 minutes with heavy traffic, 7 days a week. Similar ferry in San Francisco, between the Pier 39 and Sausalito would have costed $22 round trip. Salute to the transportation department of NYC!
This got me thinking, why should we pay anything at all for local public transportation?
Local city buses, subways, ferries, should and can be made completely free! This would encourage every commuter to seriously consider abandoning driving, thus reducing our carbon emission, eliminating highway congestion, resulting cleaner air and better lives for all.
The online world has gotten used to free services with ads support. Why the same model can't be applied to public transportation? Make some creative use of the empty inner walls on the bus, in the subway, and in the stations. Free services will guarantee a huge spike in the amount of people taking these buses and subways, and thus the amount of eyeballs, which increases the value of these ad spaces also.
Let's learn something from the Staten Island ferry, and make public transportation free!
Read more...
Friday, August 28, 2009
Idea 45: Record And Share Your Life Style

iPhone is bloody useful. We touched merely the surface with idea on personal metrics for diet optimization, and idea on real-time route sharing for smart carpooling. Now with people sharing daily meal pictures on Twitter, it might be useful simply to record and share your entire life style.
That means recording the places you visit, thing you do, media you consume, people you befriends, what, when and how. The new iPhone app will have the following:
A push button to record the current time and place, and the reason you're here. It might takes a bit of discipline to form the habit of writing down the activity whenever you visit somewhere, but that might just prove useful. Over time, you'll leave a trace of footprints as to how you lived your life, what you've learned, what you did right, or wrong. It might help if there are pictures annotation as well.
Another button to start and stop the recording of real-time routing trace. When started, the app will intermittently query the GPS for current location, and record the complete travelling route, until the stop button is pushed. Want to share your shortest driving routes? Want to share your favorite running tracks, hiking tracks, biking tracks? This is the way to do it!
Life is short, interesting, and full of excitement. Why not record it, share it, discuss it, and pass it on?
image credit omar eduardo
Read more...




