Startup Idea: Moonscribe - Draw user specified art on the surface of the moon using teleoperated robot.

Here is an idea to create value from dirt by moving it around ;-) on the moon.

We are almost at a stage where we can send a privately designed teleoperated vehicle to the moon. From the Lunar X prize teams we know that it nearly fits within a 30 million budget.

The awesome thing about the moon is that since it does not have an atmosphere or flowing water, a footprint on moon or a sandcastle of regolith  will stay intact for a really really long time. The only problems are meteors and human activities.

Let us say a single drawing can be completed in an hour. In a year 2500 drawings could be made assuming a little more than 100 days of operation. Over a period of 4 years, you could draw 10,000 items.

Suppose each drawing costed 5000$, that is 50 million dollars. a decent profit for those involved.

Now the price of the individual drawing need not be a fixed amount. It could be $100,000 for a large company logo like say a Google logo or a 500$ for a small one expressing a guy's love for his girlfriend. A bunch of opensource guys could crowdfund a large firefox logo on the moon using kickstarter. A billionare might pay for his child to scribble A-Z on the moon using an iPad app.

With negligible expenses, the million dollar home page managed to sell a million pixels and became part of internet history. Imagine letting people draw an image on the moon that will last for decades or centuries.

The drawings cannot be seen from the earth. But the coverage on the Web and on TV will be the actual thing that will give it maximum value. It being done for the first time will give it most of the value. Apart from cash there are a number of other barriers that will make it non trivial to clone. So the resource will be a limited resource for reasons apart from money.

Will this turn the moon into an eyesore?

The idea is not to turn the moon into a big billboard for advertisers. Man left footprints on the moon. They have become heritage. Think of it as art... not banners. IBM crafted its logo using 35 atoms of Xenon to demonstrate its technical prowess. It does not have to be huge in size... what you draw must be viral. Think of of it like Google doodles not like skywriting...

Howto fund?

An angel could fund the creation of the prototype vehicle/robot. The main expense would be the actual transportation to the moon. This could be done after the users pay up for the service. They can see a preview of their drawing done in a remote desert on earth but will have to payup before liftoff.

Will angels and VCs care?

This is not exactly a billion dollar idea "by itself". But a company that starts off with this today can keep expanding.

Imagine a grandfather who could gift his grandson an ice sculpture in a martian cave with a note that says. "Go and get it for yourself whenever u can". The beauty of the scheme is that melting and refreezing ice on mars is a negligible expense. But a sculpture is much more valuable than the water in it just as a picasso painting is worth a lot more than the paint and the canvas. Build a strong igloo at a high enough latitude and leave it untouched. No active refrigeration required.

The real challenge is creating a product for the masses. It is not enough to get the star trek fan excited about this. When a baby is born, when a guy proposes to a girl, when a music band launches, or killing time while on a commute, each person should have enticing possibilities to derive value from spending money on something space related. That is when it becomes a billion dollar market.

What is needed is "real execution"!

Think about this, a company like Zynga makes people enjoy the creation of virtual farms. Imagine creating tangible objects on a remote planet and owning them for the rest of your life... Remember this... No one owns the land on mars or moon. Moving stuff to and from earth costs a lot of money. Moving information hardly costs anything. These situations lead to some really interesting opportunities. There exists a huge spectrum of opportunities between launching billion dollar rockets like SpaceX does and selling food to vistors at a Virgin galactic launchpad. This particular idea is in the tens of millions of dollars range and can expand into billion dollar range by hiring the right people and expanding into neighboring activities. I guess I have disclosed way more than a person would usually disclose with out an NDA ;-)

I have a lot more to say. This is just the tip of the iceberg. But I guess that would need an NDA... Want to be a co-founder or investor? Ideas are dime a dozen. Execution will be the differentiator. Right now a dozen other teams would be dreaming up the same thing. The real question is who will manage to pull it off the ground. Farm town existed before farmville. Execution is what finally counts.

Write to me ibm.sudarshan <at> gmail.com

Prediction market for News and tweets

Prediction_markets are speculative markets that can predict outcomes. Can u predict whether a newsitem will turn up on Techmeme/Slashdot/Digg/Reddit/Twittertrends/Buzz/Zeitgest... Can you predict how many tweets it will get by a particular time

If u can, then buy low and sell high...You buy and sell karma points... 

This will help spot raw news much earlier than even twitter...

It could even end up being a Facebook game with real money

 

 

e^x on Venus

a set of inflated windmills like the one shown can be made of nitrogen instead of helium on venus. the plastic can be made on venus itself. 

To create a wind farm there would be a nitrogen filled windmill attached to a large CO2 balloon connected by a tether. At the 2 heights there would be a velocity differential

This would allow exponential increase of energy production

Need to make a flying car using a stack of quadcopters that lift a 2 person box like the ones in a roller coster

 

Self-assembling vehicles take flight (w/ Video)

(PhysOrg.com) -- Researchers in Switzerland are developing miniature vehicles that can self-assemble and then take off vertically and fly as a stable array.

The system, developed at the Institute of Dynamic Systems and Control at the Swiss Federal Institute of Technology (ETH) in Zürich, is known as a Distributed  Array. The system is basically a flying platform consisting of a number of autonomous wheeled vehicles, each driven by a fixed ducted propeller. Each vehicle has its own motor and flight control system and is equipped with a computer,  and wireless communication systems.

The individual vehicles can drive on the ground, dock with other vehicles, and can fly but only in an uncontrolled and erratic way. When the vehicles are joined together, however, the combination becomes a sophisticated flight platform capable of coordinated flight, with data shared rapidly between the individual vehicles and flight control distributed across the array.

A short video which introduces the Distributed Flight Array which is being developed at ETH Zurich.

Each vehicle has sensors, and the information from all the vehicles is combined to calculate the thrust required for take off and to maintain level flight. If the flight level is disturbed each vehicle in the array determines the thrust required to correct the flight of the array, taking into account the position of the vehicle in the array and its motion. Instead of landing, the array breaks up, with the individual modules dropping to the ground on their own and then driving off.

The distributed flight array is at the proof of concept stage, but in the future could be useful for applications such as lifting heavy objects, with the number of vehicles or modules in the array selected according to the lift required. The modular nature of the array also means some failure can be tolerated since if one or two vehicles fail the others can compensate.

Self-assembling vehicles take flight
Enlarge

Among ETH Zürich’s many other research projects is a Flying Machine Arena (FMA), which is a 10 m cube of indoor space designed specifically for testing and validating autonomous vehicles. The space has glass on one side and nets on the other three sides so flying vehicles can be tested safely. An optional safety net can also be installed on the bottom of the space to prevent crashes onto the hard ground.

ETH Zürich, founded in 1855, is one of the leading universities in the world for technology and the natural sciences, and aims to facilitate research and results beneficial for society as a whole.

A Dancing Quadrocopter. Synchronizing the Motion of a Quadrocopter to Music.

Spottr - Core dump... do not read

Image labeller shows humans like spotting patterns. It activates our reward centers.
 
It can be generalised to tagging in general.
 
FluidDB provides a URI centred tagging service.
 
quicktagger provides a good UI.
 
Mechanical turk provides general framework.
 
Facebook provides credit distribution system.
 
Farmville provides addictive labor pool.
 
Spotting could be
1. Animals.
2. Recipies
3. Food
4. VCs
5. Karma Systems
6. Speakers
7. Debates
8. People in a photo or video
9. Parts of speech
10. Named entities   
11. Lists
12. Tables
13. wikis
14. Videos
15. Maps
16. Circuit Diagrams
17. Events
18. PCB layouts.
19. Prices
20. Reviews
21. 3D objects
22. Blueprints
23. Lyrics
24. RankLists
25. Quoted Strings
26. Sequences of pages
27. Trees of Pages
28. Wikis
29. Trademarks
30. Reusable slides
31. Line drawings => Structure of everything
32. People
33. Artifacts
34. Antiques
35. Memorabilia
36. Collectors.
37. Pamphlets/Banners
38. Predictions
39. "N ways to"/"N tips"/"tricks"/tweaks advice with votable 
40. What did u do for X where X = Environment/Open Source/Country/Religion/Science/AI etc. Responses could be like blogged,donated,wrote code,protested,stayed in jail,got injured in battle
41. Polls
42. Interviews
43. Debates
44. Experiments n activities
45. Just born pets for sale
46. Forums
47. Q & A => FAQ,StackOverflow,Yahoo Answers,Google Answers,Forums,Ask HN/Slashdot, Any threaded discussion 
48. Manuals
49. Tutorials
50. Real Stories
51. Tagables => First list all things that have skos working for them.
52. Datasources/Structured Data and APIs
53. Plants I want to have/Plants I will sell/Plants I want to barter.
54. Stuff I made.
55. Bad Road Bumps
56. Mapping environmental damage.
57. Technical Documentation Spotter
58. blogpost classifier
59. Pivots by startups
 
The list is endless.
 
Spotting shall be done by humans as well as machines.
 
Ad revenues from mashups will fuel facebook credits.

[Cloud of Data] => [Spotter] => [Split/Merge/Redirect/Curation] => [Annotation] => [Database Model Creation] => [Migration] => [Startup Creation]

[Photo/Video Camera+Smartphone App]^

 

Startup Idea: News I may have missed.

The job of a good newspaper is to sift through tons of possible news and extract the most valuable subset that fits into 16 or so pages that suits the demographic of the news paper.

With the web this can be defined as choosing the most valuable subset of possible news and extract the most valuable subset that fits into the TIME that a PARTICULAR USER chooses to spend on THAT DAY.

Suggestion: A browser plugin watches what you browse. It observes ur various actions like your twitter feed, people you follow on twitter, your buzz stream, your blog posts and the time you spend on various webpages. The mails you compose, the forums you participate in and so on. All this gives the plugin a feel of your choices.

Now it looks at your tastes and determines what other similar people like and compares it with what you have spent your time on. It spots NEWS article CLUSTERS popular among similar people but may have been missed by you.

There will be tabls like: News I missed today, News I missed yesterday, News I missed this week,last week, News I missed this Month, News I missed this year, News I missed last year

Idea: ProxyPC: A seamless integration between Tablets and PCs

We are seeing a lot of activity in tablets lately starting with the iPad. Google has promised us the ChromeOS tablet. There are other contenders and plans like HP+WebOS. With the flurry of activity, it has become almost certain that each one of us who is not a passive consumer of information will now need to have a PC and tablet and a phone.

The tablets are highly constrained devices but do what they do extremely well. They have to be rugged, light, consume as little power as possible, typically support multitouch and so on. PCs or Macs on the other hand do not have weight restrictions, OS constraints, power constraints and typically run apps that rely on a mouse pointer and have awesome power-hungry graphics cards.

Things like AutoCAD/Photoshop/CS4/Visual Studio/Eclipse will probably never ever run on iPad/ChomeOS tablet. For the guys who view the computer as a consumption device it is a non issue. But for those who "create stuff" or work in an enterprise using legacy applications or develop code, they will find themselves switching back and forth between the tablet and the PC often and there is a lot of scope for improvement here.    

The idea being suggested here is a technique using VNC that every geek would likely reinvent, but to make it available out of the box and make it an obvious way for the non tech savvy to use these things. There is also a small piece of hardware that seems inevitable. But someone out there may be able to suggest a Software only solution.

So let me begin without much ado. Shortly after the iPad was out, we began to see solutions like : Mac OSX Running on the iPad! With Flash!

What is being shown there is actually MacOS running on a Mac being displayed on an iPad through VNC. There is also a Windows 7 on the iPad: Hack n Mod

These stories talk about a special application called Desktop Connect that costs $11.99 and enables VNC on an iPad, which should have been freely available by default(cough). 

But what these guys have done just scraped the surface of what is possible. Remotely connecting to the desktop currently has many shortcomings.

  • One needs to physically go to the PC/Mac and turn it on or leave the PC permanently on even if it is rarely used.
  • Fingers are too fat to be a substitute for a mouse pointer.
  • The experience is not seamless and one is continuously aware that he is now temporarily working remotely on another computer.

I just listed the pain points most obvious to me. I am curious to see what other hurdles affect the quality of such usage. Anyone who can attack the pain points using clever solutions has the opportunity to show off a cool hack if he is of the open source type and a source of $$$ if he is the entreprenurial type. 

ProxyPC:

So let me set the conversation going by putting out some suggestions.

Let us say there exists a PC or Mac with the following abilities. Let call it ProxyPC for now.

  • It has Bluetooth and/or WiFi capability. 
  • Another device like an iPad can launch a document on the ProxyPC by asking the Device to "wake up" from suspension or hibernation if it is not already active.
  • The iPad/ChromeOS tablet can ask the ProxyPC to launch a document in an associated application. For Eg. It can ask the PC to open a psd file in Photoshop.
  • Once the document is opened the VNC connection is seamlessly opened behind the scenes and the user gets the illusion that the file is opened locally on the tablet. 
  • The saved file can be easily synced back to the tablet or cloud. 
  • The user would actually sit in front of the PC only when the screen refreshes too rapidly AND needs a lot of computing power like in high end games. Things like movies would be watched locally on the tablet. 

This brings together the best of both worlds: We have the portability and battery life and multi-touch of the tablet... the raw CPU and graphics card power and large storage of the PC. The only time anyone would physically sit in front of the PC is to get a lot of typing done or play games at very high frame rates.

Combined input from fingers+stylus.

Fingers are too imprecise for many input activities. So we will need better pointing if the ProxyPC session has to be productive. Some novel experiments are underway combining pen and finger multitouch input. Thus we can simultaneously achieve accuracy as well as easy finger input. Will future iPads have stylus input as well? Will Chrome OS tablets have fingers+stylus? Should we demand it? I invite you to watch video for some inspiration.

Careful integration:

The real entrepreneurial challenge here is to identify each tiny pain point and make solutions where users have an out of the box fantastic experience. Just buy the ProxyPC capable computer and install an App on your tablet and get maximum integration with minimum user setup/configuration/tweaking. I would have bought such a device if it were available. Ask yourself whether you would buy such a device?

Will apple approve such a use of "its" device??

Maybe not. But there is a huge opportunity out there... and many competitors will emerge. Many enterprises could choose to use such a device where legacy tools like Lotus notes is accessed from a ProxyPC. A huge majority of people whose reliance on PCs is limited would immediately love to switch over to minimize pain. Also if users use their Tablets as the primary internet connected device and use the PC as lonely rarely booted devices, nuisances such as virus would be significantly reduced. In a corporate set up one could even use Citrix to run large number of legacy apps on a single system reducing maintenance costs in many ways.

All I can say is there appears to be an interesting opportunity here and some small fortunes to be made. While no technological miracles have been suggested here, the ideas bring value to users. The devil lies in the details. For Eg how well does the clipboard work?

I am curious to know about such products already in the market. Do you think this is a fertile ground for exploration. If not, why not? If you think it is worth it, post a comment to tell the rest the world what you care about the most. How it could be made even better?