This is the P2PU Archive. If you want the current site, go to www.p2pu.org!
Mashing Up The Open Web - Mar 2010
My recent threads
You haven't posted any discussions yet.
Recently updated threads
- Course Recap Blog Post Thu, 2010-05-27 13:57
- Reflections on the course Mon, 2010-05-10 02:09
- Notes Mon, 2010-04-26 22:00
- Project Ideas Fri, 2010-04-02 18:55
- Pairing Mon, 2010-03-29 00:23
Project Ideas
Go back to: General discussion
Please post a detailed description of your project here and comment on the projects proposed by the other participants.
I have an idea of designing an portal which stores the basic first-aid details like blood group, Blood Pressure other critical infos which can aid in faster information retrival through Short messaging system from the client mobile at the time of accident . This can help public to send a notification ID to a common gateway (say 53000-twitter like) and recieve the stored critical data of the person who encountered with an accident. This time-critical data can help in faster first-aid treatments by the doctor which can save many lives at the time of accident.
This is a fantastic idea. I wonder if you'd run into any privacy concerns surrounding the storage of personal medical info (e.g blood type)? Overall though, I think is a really interesting idea with a lot of utility. Best of luck.
REPLY to Santhosh V
Health is a very interesting subject. TEDMed gives a lot of interesting talks available on the TED website where they talk about leaving institutionalized health care behind and focus on "health care at home". There is a range of hardware available that helps doing this and additional software, also for the iPhone.
I haven´t seen APIs on the web so far which could help creating some interesting services together with mobile applications or browser based apps.
Have you found any?
My project is very much still in the planning stages, but I'm interested in creating a 'Crisis Alert' application that can calculate the average number of social media mentions (from publicly availalbe feed data) for a brand per day and then allow users to set an "alert me" notice that sends a user an alert if there is a new surge in volume....
REPLY to John Stauffer
Crisis Alert sounds interesting. When using Twitter and their new GeoLocation information, you could visualize the Tweets on Google Maps to locate "Hotspots". I did not understand what you mean with "social media mentions for a brand per day" in relation to "Crisis Alert".
I'm planning to create a web application (tentatively calling it 'FreshScript') that allows a user to connect their FreshBooks account (using OAuth) and create custom workflows using the FreshBooks API and Webhooks implementation. An example of a workflow would be: when an estimate is updated, if the status has changed from 'draft' to 'accepted', create an invoice and send it to the client.
My goal is to create a useful application, but also to demonstrate how to write a sample application that uses OAuth and Webhooks.
REPLY to Paul Osman
Very interesting. Never heard of the term "Webhooks", so I looked it up. Sounds awesome and looking forward to see how that stuff works and what scenarios it is used for.
http://timothyfitz.wordpress.com/2009/02/09/what-webhooks-are-and-why-yo...
I thought of a simpler one than the one I signed up with, but still fun:
"Wimf: What's In My Fridge?"
Recipe and grocery suggestions, with nutritional data
APIs:
recipes - http://api.bigoven.com/
products & data - http://platform.fatsecret.com/api/
Implementation:
Linux/Apache virtual server, using jQuery & rails (new to both!)
Specifications: forthcoming ;-)
REPLY to Oren
Nice idea! I guess you will build some kind of community around it?
I am thinking of creating a mashup of the existing OER (Open Educational Resources) - feeds for a lot of them (or probably most of them) are found here:
http://www.ocwconsortium.org/use/use-dynamic.html
Educational resources can be organized and connected by country/ university, subject, lecture media type (audio, video, text). Linking similar lectures through tags or something else would be useful too.
Academic Earth (http://academicearth.org/) is an example of such an aggregation - they have brought together lectures of the leading universities of US.
My idea consists of on online shop that uses social media APIs for its online marketing efforts, e.g Facebook Connect, Twitter and Google Base Data Syndication.
The shop itself is a local platform for small manufacturers and artists to help offer their products to a wider audience. It should contain community features like comments, voting and sharing and as far as possible, these features should be provided by APIs, e.g. Disqus or GetSatisfaction.
For sellers and buyers to register, it accepts OpenID and OAuth services.
For payment the sellers have to provide their own account form PayPal (Standard or Express) or Google Checkout.
Wider research has to be done on available Data APIs like Google Base Data API to provide sellers with tools for comparison with similar products and hints on what keywords ("tags") to use for their products (SEO). This actually would be my contribution to the assignment of collecting and using "Open Data". I am not sure whether this is enough and be happy about any comments and suggestions on this.
As this platform does not generate its own content, but relies entirely on "user generated content", see seller´s products and buyers comments, it needs to be flexible and reliable in the syndication of page content to avoid its duplication and to find customers also over "the long-tail".
Web programming languages to be used: PHP, JavaScript, HTML/CSS
Considered Frameworks as foundation: Symfony 1.4, Drupal 6 or TYPO3 4.3
I also considered on working with data from the german government, but it is not existing! No Open Govnerment initiative like http://www.whitehouse.gov/Open
But they are working on it. Anyone here considering doing something with US data?
Open Government, Government 2.0, Open Democracy
As I am from Germany I looked up whether there are APIs from government institutions and offices, but nothing with real datasets so far. Was interested in creating something in this area.
I have seen that this is more developed in the US. Has anybody already made experience with their APIs, like http://www.whitehouse.gov/Open
REPLY TO Dennis Riedel idea on Democracy 2.0
Whilst Germany itself hasn't released any public data sets, the UK has:
http://data.gov.uk/
A substantial amount of this data is already available to mashup through restful api's. There's also a significant amount of data that has been released by not currently converted over to RDF or another more accessible format.
test
I am thinking of creating a mashup of the existing OER (Open Educational Resources) - feeds for a lot of them (or probably most of them) are found here:
http://www.ocwconsortium.org/use/use-dynamic.html
Educational resources can be organized and connected by country/ university, subject, lecture media type (audio, video, text). Linking similar lectures through tags or something else would be useful too.
Academic Earth (http://academicearth.org/) is an example of such an aggregation - they have brought together lectures of the leading universities of US.
This is in reply to Tanushree Jindal's project Idea.
A very useful data set for enabling you to do this is the creative commons oer search engine: http://discovered.creativecommons.org/search/
Whilst the underlying data is stored as RDF, this hasnt yet been exposed through an api ( it will be soon ) however in the interim theres no reason why you couldnt parse information out pages such as this:
( I did a search for Hannibal and clicked on the (explain) link under the first search result. The data is structured, and contains many attributes you could use for gauging similarity.
http://discovered.creativecommons.org/search/explain.jsp?idx=0&id=17651&...
http://tr.im/TsT7 are resources that would be good if mashed.
Lots of cool ideas! The webhook thing does sound cool! So I am supposed to have it up and running between Github and Redmine, but it isn't, don't know why and don't know how to troubleshoot it.
I have 3 projects that I'm working on and need to decide which one to do first:
I'm tempted to do this one - because it's easy and fun for me to illustrate:
weatherMom or weatherDog
an iPhone or web widget that dials up the weather in your city and shows you how to dress most comfortably - layers, raincoats, boots, umbrella, wintercoat + on johns, excellent for parents or newbies to winter climates
Done with cartoon characters or manga for the older crowd
No more - "put your snowpants on, it's cold outside!!" Dial up weatherBunny to show them YUP they need their snowpants, or sunscreen or at -1 degrees it's too cold for a sweatshirt.
Technically I know I need to start with the Database of images with a weather feed, but I'd love suggestions on how to make them work together "mash them"?
The one that I proposed - which is to set up a a neighbourhood communications forum - needed to motivate and engage neighbours to solve problems in the area.
By invitation based on location.
I want to use it to build a case to change traffic light locations and add sleeping policemen to this community.
Optimum would be to get feeds from site that are based on studying traffic solutions, city resources and guides, solving community problems like traffic, crime and entertaining your kids in the neighbourhood
city calendar listings, community events guides and rec centre info would be a bonus but would need to be very neighbourhood specific.
Another project is a storytelling site that would use the SIMILE Timeline code so I'm looking forward to next weeks class!
Hi all,
How many of you do programing, listen to music, browse and also update the same as your status in your social network?
Would it not be easy if all this could happen in a single page!?
My project idea is to do a tab within Firefox, which allows you to manage all our social network, listen to music and also code!
It would be even better if we could sync this page with something like weave, so that we can emulate our machines to behave as a sever over dyndns kina effect and we could use all our files in our desktop anytime anywhere.
When it comes to social netwroking there are many websites available and the API's normally allow us to update our status easily, based on the selection the options must vary, for ex : if the user selects FB the be will have extra options specific to FB, may be like upload video and so on.
The user might also select all and update his status.
When is comes to music, two options must be provided one to listen it from the local HDD and other of-course over the internet. the user must have the flexibility to configure the player with bookmarking feature
When it comes to coding a terminal will be provided where user can do his hardcore coding stuff.And one selecting a piece of text/code on any site, the selection must be saved to a specific directory with appropriate extension.
For EX : When i select a ruby code in a site with my mouse, the selected code might be saved in a folder like ~/code/ruby/Hello.rb.
The file name will be the class name and deciding the extension must be done using few parameters like may be the interpretive line or so.
And each window in the tab can be maximized so that we can concentrate one a particular aspect, so when we have a dead line ;)
Here is a preview of the design http://is.gd/b1vyD
I have come up with another project idea which might suit better the given timeframe and subject.
The application allows the user to enter a search term for images. He can enter "tree" for example and the application will first look up the translations (Google Translate, ...) for "tree" in different (specify!?) languages. Armed with a bunch of multilanguage tags, it will then look up images in Flickr or Twitpic (for example and if possible) and show the user a tagcloud of languages as a result. Behind these language labels is the gallery of found images. The more images for one language have been found, the stronger and bigger the language labels are displayed in the tagcloud.
With geo information it should also be possible to show the different images in a Google Map.
First tests run with the Flickr API and Google AJAX Language API.
Will use Zend as underlying framework.