Ten Ideas


Being a hacker is all about the open sharing of ideas. So why do I keep my list of ‘projects in development’ so close to my chest? Inspired by tales of R&D departments with security measures the military would weep at? Enchanted by the notion that my ideas are worth millions, I just need to unleash them, then sit back and watch the cash roll in? Yeah, that’s pretty much it!

Yep, until very recently I was an idea hoarder. But inspired by Jacques Mattheij’s recent outpouring of his ideas, I’ve changed my attitude. I’m in good company – the folks at ycombinator have shared their list of “ideas we want to fund“, the people at halfbakery.com have an entire social ecosystem based around sharing ideas, and the Six Month MBA team have listed a whopping 999 business ideas for anyone to pick up and use.

Why share my ideas? Ideas are often said to be worthless until implemented. I’d objected to that sentiment in the past, being a big ideas person. But now I can see there’s truth in it – a bad idea implemented excellently will trump a good idea implemented poorly, and as Paul Graham says: “imaginative people will take (the ideas) in directions we didn’t anticipate”, and “No matter what your idea, there’s someone else out there working on the same thing”. Sharing something multiplies its value.

I encourage you to share your ideas with the community too, because:

  1. Someone’s probably already thought of it anyway – no need to keep it secret
  2. You haven’t done anything with it yet – so maybe you’re not the right person to bring it forward
  3. Inspiring others benefits everyone- let’s talk about these ideas, and create new ones
  4. You’re not as clever as you’d like to think – others can see problems and opportunities that you can’t
  5. Sharing ideas can kickstart the product – if everyone says “wow I like this”, then you know what to do

So without further ado, ten ideas I’m thinking about:

1) Swiss Army Knife Selector.

There’s a genuine need among swiss army knife fans for an app where they can enter in their requirements and it will tell them which swiss army knife has all those features (you’d be amazed how many different makes and models there have been over the years). There used to be sakselector.info for this job, but it’s now out of date (and the design leaves much to be desired).

Comments: It would be pretty simple to create the search system, a little harder to populate the database. The interface will be trickier. How to present something that’s not a mess of tickboxes and selections? Could be expanded to allow other communities to create their own selectors. I think you would have to understand the audience well in order to succeed.

Profit Potential: affiliate links, marketplace, ebay store, adverts.

Competitors: sakselector.info, but it’s outdated. Otherwise none that I’m aware of.

2) Historlytics.

Analytics for your history. A website supported by an add-on, which monitors every URL you visit and creates a graph of which sites you’re visiting over time, how your browsing habits change, how much time you’re spending on particular sites or classes of site, how much time you spend online per day, when your peak productivity times are and so on.

Comments: Whoa the privacy nuts are going to love this! Although keeping your history on a third-party, password protected account is probably more private than keeping it in your browser. See also voyurl.com.

Profit Potential: Ummm. Obviously you can’t sell the data, that’s just evil and no-one would sign up anyway. A freemium model might work if you built a kind of sync service into it; like dropbox for your web history.. You can’t really sell ads. Collate the data like alexa do and sell that somehow? Publish a list of top trends, like a kind of automatically populated digg/reddit social media outlet?

Competitors: Google web history, Voyurl (if they ever launch), Blingo.

3) Matlab in JavaScript.

Matlab is a numerical analysis tool used in high-end research and education. It’s a downloadable piece of software. There’s room to create an online competitor which enables open-source JS plugins, sharing results/experiments with co-workers, cloud-based computing or processing data in javascript, embedding in web sites and so on and so on. I don’t think you could fully take on matlab’s market – I can’t see toyota or MIT doing supercomputer-level analysis in a web app, but for students and small research projects it would be great.

Comments: I’m not the right person to take this forward. I don’t use matlab much, I’m not mathematically minded. It’s a winning concept though. If you decide to take this on, then for gods sake allow people to import from matlab. That’s the one essential feature. Would openoffice be popular if it didn’t open MS-Word files? No.

Profit Potential: SaaS, tiered pricing, all kinds of models could be applied to this.

Competitors: None that I know of; haven’t looked very hard though.

4) tldr

Automatic summarisation web service – you provide the URL, it gives you the one-sentence version of it for people who think it’s “too long – didn’t read”.

Comments: Two possible approaches: 1) Automatic summarisation / keyword extraction. This is sort of possible, but technically difficult and not of good enough quality yet, 2) User-submitted summaries. I would suggest going down the user submitted route while we work on auto-generation. Users can be incentivised via a kind of points/rep/karma system.

Profit Potential: Could be sold as a webservice for copy writers perhaps, but this is more a ‘cool idea’ than a ’3 years on I’m retiring’ product. Although there is a genuine need for this as a service, I’m sure money could be made.

Competitors: Hah, just goes to show that there’s always someone working on your idea – tldr.it, based on the the open source libots, launched earlier today and is now on techcrunch! I guess I’ll scratch this one off the list then ;)

5) imCaptcha

reCaptcha helps computers read books, imCaptcha helps computers read images. The user is presented with an image, they have to give a one-word description of the image. We aggregate responses and use that to provide an image recognition API – with enough users it could be near realtime. I have already written a machine learning system which can automatically tag the colours in an image. I’d love to be able to add genuine machine vision to it.

Comments: Lots of problems here – how do you know if the word is accurate, use reCaptcha’s approach and give them two images, one we know the words for and one we don’t? Wait till 2 users tag an image with the same word? Images have a lot more correct words than words do – take the favicon for this site – correct words could be: orange, green, mango, fruit, food, icon… detecting correct vs incorrect is a problem.

Profit Potential: solve the problems and you’ve got yourself a something that image libraries will pay richly for. (Solution, as always, limit the domain)

Competitors: Microsoft Assira, Imagination

6) pH – Password Handler

I use the same password on almost every site. Some of you don’t, but I bet most of you do, or at best you cycle through a few variations. pH is a bookmarklet; when you want to register or log in, you enter your one master password. It hashes that with the website’s URL to create a unique password for that website. The master password never leaves your computer, you end up using a different password for every site.

Comments: What if you forget your master password? Unlucky! What if the website’s URL changes? Then change your password. If you lose the bookmarklet, just download it again – it’s the same algorithm for everyone. What we’re talking about here really is a salted hash on the URL where the salt is your master password. What about keyloggers? Yeah, well what about Boris waving a .44 in your face?

Profit Potential: none. (unless you want to go down some kind of freemium model or something, but that kinda defeats the purpose). You could package it up as an add-on and make money by doing something evil with adverts I suppose.

Competitors: various password managers.

7) tinySketch.com

Web based whiteboard. You go there, you get a short url like tinySketch.com/g8k4 and you can share the URL with colleagues etc. You get a simple canvas, and updates are pushed to everyone viewing that URL in real-time. So many times I’m speaking to my boss on the phone and wishing I could just show him what I mean.

Comments: done lots of times before, but it’s a good domain name at least (want to buy it? email puremango.co.uk@gmail.com)

Profit Potential: adverts, or preferably a freemium model with better features for paid users.

Competitors: loads!

8) Sketch Forker

You create a drawing using an online paint clone, then other people can come along and create variations based on your drawing, ‘forking’ it, to borrow a computing term, you can join in the new version or carry on with yours. drawings with more contributors are ranked higher, etc etc.

Comments: You’d have to have a few people on board who can actually draw, unlike me! Also: would this even be fun?

Profit Potential: adverts, freemium, products (posters, prints of collaborated sketches)

Competitors: There are a few collaborative drawing games around, eg isketch.net, but this would be more open, like twitter.

9) Expandable Desktop

Google maps meets your desktop. Imagine your icons and bookmarks laid out on a zoomable, pannable, draggable landscape. You can arrange your icons by theme, and zoom into a topic when you want to work with it. In fact you don’t need to imagine – this was my entry to the An Event Apart 10k webapp challenge, check it out: http://10k.aneventapart.com/entry/details/243 (with 17 votes and an average of 2.6 out of 5, it’s not bad, but maybe this isn’t the user interface revolution I’d expected! I was in a huge rush to finish it though, so with some polishing I still think it has potential)

Comments: Do we need a better interface? Can users be convinced to invest their time learning a new way to interact with their computers? I think the answer to both is yes, but there are those who’ll disagree. Deep integration is key here. My 10k app only worked for bookmarks and so had limited applicability. If you wrote a desktop replacement which handled programs, documents, bookmarks and open applications in a unified interface, that would be pretty cool.

Profit Potential: well, bumptop was acquired by Google so there’s got to be some money in desktop alternatives! Failing that, adverts and freemium models.

Competitors: Raskin, probably others.

10) Augmented Web

This one’s not an idea for a web site, it’s more a trend. Imagine a web where you’re not tied to a keyboard and mouse, where you don’t have to be sitting down to use it, and where there’s no division between audience and actor. A web which incorporates augmented reality in a natural useful way. This is the web of the future. Where I point with my hands, rotate objects in a projected 3d space, move my face closer to the screen to zoom, and become a part of the place I’m visiting, instead of a passive observer. I’m talking gesture recognition, face tracking, eye tracking and all the juicy “around the corner” technologies we see at TED demos.

Comments: The web is starting to look old. It looks like paper – flat, 2 dimensional. We still use layout principles set down in the days of newspapers – columnular text, embedded images. Where’s the interface of the future?  I mean, when you compare the web against the interfaces provided by Guitar Hero, the Wii, accelerometers, and so on, the web’s lagging behind the curve. Isn’t it?

Problems: The web’s not ready for augmentation yet from a standards perspective. HTML5 has a proposed device element, but that’s some way off yet. Also, it’s debatable whether what I describe above actually is a better interface for rich media than a two dimensional screen. But certainly for some domains – for example games and scientific visualisations – there’s going to be a need for multidimensional augmented reality delivered across the web. Within the next 3 years, someone big will start incorporating hand based gestures into their web applications. Maybe you’ll move through photos on flickr by waving your right hand, and back again using your left. Maybe you’ll be able to zoom into youtube videos by moving closer to the screen, or view different angles by tilting your head. Perhaps someone will just create a video tennis app where you and your opponent play ping pong with your hands and some JavaScript interprets your movements and draws the ball. But it’ll happen, and it’ll be big.

Profit Potential: Once the market realises what the web’s capable of in terms of immersive interfaces and AR, there’ll be a huge demand for it. At the moment it’s still an unusual interface paradigm, so people don’t expect it or request it.

Competitors: Non-web platforms are already getting incredibly advanced augmented reality, eg realtime object removal, and MrDoob is making huge steps forward in delivering various AR demos across the web using flash.


So, there are my ten ideas. I hope you’ve found some of them interesting, I’ll be very happy if I inspired you to share your own projects – please comment below with a link if you decide to open up.

If you do decide to take one of these ideas on, I’d appreciate it if you drop me a quick message just so I can check up on progress. I’d rather open up and have someone work on these ideas than keep them to myself and never have a product released.


Related Posts:

, , , , ,

  1. #1 by jonmcrawford on October 19, 2010 - 6:38 am

    Great post, exactly what I’ve been thinking lately, that even if I have what I think is a great idea, I’m not necessarily going to get anything done with it.

    Take your 9 & 10 and combine them, make your desktop appear to be in a 3D space, let people manage their icons/work by manipulating the 3D space with their hands. Find a way to hack the Kinect to make it work, and I’ll buy it from you in a heartbeat.

    Some of my ideas:
    1 – something that can be added to every vehicle on the road that transmits vector coordinates and projects position for the next x seconds (determined by the max safe stopping distance for the vehicle). Each vehicle broadcasts within that limited area and reads other vehicles within the same radius. If vehicles will collide based on projected data, they slow/speed up/change trajectory to miss. Integrates with your GPS so that if you have a planned route, that’s taken into account. (next step for this is to have the cars drive themselves, get rid of traffic lights and just integrate the traffic flows seamlessly) Priority rules would need established, examples: emergency indicator for cars in distress or emergency vehicles have highest priority, everyone moves away from them. For straight-line traffic, higher velocity has higher priority. Turning vehicles get higher priority (to allow for better integration of the traffic streams), N-bound (closest to 0 degrees) traffic gets priority if there’s a tie, etc.

    2 – Swype for desktop PCs. I am working hard on acquiring carpal tunnel, it would be great if I could use a Swype virtual keyboard on my work PC, and just use the mouse to activate it (mousedown = press finger, mouseup = lift finger). Probably creates a different kind of carpal tunnel syndrome, but at least I don’t have that one yet.

    3 – take the above, and make the letters shuffle (on startup, or when prompted) to have the most frequently used ones (for you specifically) grouped the closest, saving you motion. Could also potentially have a single key for an entire word, or maybe even phrase that you use repeatedly. I use snippets like this for coding, would be great for real life too. (be harder for us old-timers, but I’ll bet kids who don’t already know how to type would eat this up). Might also get rid of the need for text-spelling, and get back to real grammatical communication. (BRB, need 2 P, c u l8tr)

  2. #2 by Trung Huynh on February 4, 2011 - 3:07 am

    3 – I think you can export Matlab graph to svg format and embed it in HTML5, no need to reinvent the wheel.

  3. #3 by Kayla on February 4, 2011 - 11:11 pm

    I type this in on the search bar javascript:buddyList.itemOnClick(511674482); and nothing happened

  4. #4 by adam on February 7, 2011 - 8:48 pm

    hey… thanks for the mention. i’d like to happily announce that voyurl has in fact launched, recently too, in private beta.

    feel free to contact @adamcl or @voyurl on twitter for an invite. we’d love to have you on board as a charter member of the service.

    thanks,
    -a.

Comments are closed.