Pretty much the same as Google labs. You will be able to see what interesting things people are doing with the Instagram API. The sites could be sorted into categories; sound, location, events etc
Users could leave feedback and vote for there top Instagram sites.
Create a free, open-source global project to map all known human ancestry
It seems like a finite problem (there is only so much information available on who’s related to who); why not aggregate it once and for all, and allow people to add to it over time?
I’m not entirely sure what that could be used for, but my gut says there are some very big, very practical applications. Especially if it expanded to collect different types of information (e.g. medical) over time.
Why should this just be the domain of private companies?
A system into which you could supply JSON-formatted twitter data (pasted/uploaded/search API URL) which would analyse it and produce interesting stats. Most retweets (old quote-style RT and new RT); distribution through time; number of conversations; long-distance conversations vs geographical neighbourhood discussions; longest conversation; most popular hashtags/words/phrases.
Basically, anything cool or esoteric that can be learnt from stats beyond the initial top-level analysis.
When the TED API opens up this summer, it'd be nice to see a website dedicated to finding videos using terms that describe how a user is feeling at a particular moment, i.e. a site with a single search field and label:
"Right now I'm ________ [Search]"
Expected terms are things like 'bored', 'tired' but could equally be 'inspired', 'happy'... 'fat'.
Depending on the matched term the service would attempt to work out what the user needs to see. Could be an interesting experiment, if nothing else.
Made by Ideas to incorporate networking tools to foster collaboration.
Some great ideas are too good to be given away, but also too much for one person to tackle alone. By introducing a networking facility within Made by Ideas, collaborators can network, gain introductions and make these ideas happen.
A nifty app that captures information from your receipts / internet orders / bank transfers and gives you weekly updates on your outgoings versus your bank balance.
It might report that you've spent 45% more on food this week than usual; on more on evening entertainment. It's not about prescriptive budgeting; it's about staying on track.
A basic building block approach to letting people build their own location based service... Smaller groups will create more interesting, fun experiences, and remain passionate about them. Taking the notional fun part of foursquare that's actually not fun, and letting people create better games for whatever purpose.
Basic service free. Sell additional features (e.g. cluster bomb check-ins that change the game for all, timed checkins that give you fifteen minutes to check in elsewhere) on a hipstamatic style 'package' basis.
Only make it for mobile. Forget the website. You build and play and upgrade through the app.
A shared browser window which two (or more) people can connect to simultaneously. Only mouse events, viewport scroll and highlights would be transmitted to save on bandwidth and allow for more responsive interaction. Actual rendering would be handled at each viewer's end. It'd be superficially similar to a standard shared desktop but without the overhead.
It could be used to demonstrate how a site works or to provide feedback on bugs. Or click through a flickr gallery so you can talk about the photos. Or watch youtube. Loads of things.
Instagram project where users take pictures of the weather and tag it; #sunny #raining #cloudy #insta-nt-weather.......you can then search by location or weather preference.
The name Insta-nt-weather was an afterthought, it could be anything.
Possible future additions:
1. Favourite places - Mark down your favourite places and see images of the weather come in for those spots.
2. Push through notifications if an image is added to one of your favourite places
3. Maybe nice to keep an archive of the images so you can see what the weather was like back in the day.
It would be nice to have a Twitter desktop client that will give you tweets only after a 25-minute timer runs out -- interval in which you're supposed to work, as per the pomodoro method.
An app/site where you can explore different events based on similarities in their sound (e.g. falling rain and tapping fingers on cardboard). Potential "shazam" functionality only the return is similar sounding events.
A web service allowing you to add links or notes to the
Instapaper-like service of your choice, using various common
communication methods (email, Twitter, etc.) rather than having to
rely on your chosen service being integrated into all of the content
providers you happen to want to save content from.
Would also help content providers, as they would only have to provide
email or Tweet functionality, and not worry about supporting the
myriad Instapaper clones.
Primarily a mobile app that mimics the success of Instagram (i.e.
making bad pictures beautiful) but instead of adding filters, it
can help suggest how to take a picture with a perfect composition.
Thus putting the final nail in the coffin of photography as art.