DEV2013

'Practitioner Session' Janastu Poster at DEV 2013, Bangalore

Ongoing Janastu activities that are of potential interest to DEV'13

This page is an elaboration of the poster at http://janastu.org/dev2013.pdf

Alipi - Re-narration Web.

Over 10% of India (120 million) have accessed Internet by December 2011 where 90% of these 10% are from urban areas [iamai]. Mobile penetration, however, has reached 900 million Indians in 10 years. Due to the proliferation of mobile devices in remote areas [iamai] and the smart phones becoming affordable [aakash], Internet access by rural agricultural and pastoral nomads is becoming a reality. While Internet accessibility groups have developed authoring guidelines and standards for "disabled" Internet users, they do assume that such a user is a (considerably) literate person. What would it be to provision Internet accessibility to non-literates?

We explore re-narration as a basis for ”designing Internet for inclusion.” In the renarration model, any web page or even an element of it can be /re-narrated/, to make it accessible to a target audience of users in a completely decentralized way. The notion of re-narration is completely general. It could, for example, mean translating a page automatically to another language. Or it could mean creating a more accessible version of a technical document, even if it is in the same language by an expert for a layperson.

   See the dev2013.org program re-narrated to some Indian language contexts. 

Or see http://mitan.in/bcp/raika. More examples at Alipi

Technically, Re-narration Web is effectively a social semantic web. The "alipi.us" (the non-literate us) can be seen as a third party service that is served by collecting the "narration" type semantic tweets on the Web. Such "tweets" are a result of someone re-narrating some content for a specific community context.

• W4A paper http://servelots.com/d/W4Aalipi.pdf
• W4A slides http://servelots.com/d/alipi/w4a-slides/
• M4D paper http://servelots.com/d/m4d-feb12delhi-ASocialWebforAnotherBillionFinalSubmission.pdf
• Alipi report http://j.mp/alipi2011

---

Mural Annotation

http://vijayanagara.in describes a project initiated by Deparment of Science and Technology, Govt of India to seek inputs from cultural heritage groups in putting to gether a demonstration case that brings various image processing technologies to show case Indian Digital Heritage. Using the heritage site Hampi to illustrate the possibilities.

Lepakshi, a temple site near Bangalore, has many large murals from the Vijayanagara period. We are experimenting on how large murals can be annotated by experts and others so that these annotations also contribute to the knowledge bank of Indian Digital Heritage.

A demo is available at http://j.mp/temple-mural We start with annotating a mural with text. See Annotation of A2A mural at http://iiacd.org/lepakshi/lepakshimap/ Then MOWL team will process the text and help situate it into a language that can be used by the tool. Once such a language is codified, the Sweet Web infrastructure is assumed as available to inform all the IDH participants and others of these annotations. These SWEETs can be used by presentation apps to render a selection of these specific to a context.

This work is a collaboration with the International Institute of Art, Culture and Democracy (http://iiacd.org) and the MOWL (multi-media ontology) group at IIT Delhi.

Related links: Lepakshi Temple Map and iiacd.org/lepakshi

---

SWeeT Web

Web as we know it gets annotated/updated with our conversations

Here is a recent sampling of a few SWeeTs.

@Anon annotated http://folkhampi.openrun.net/mural-annotation/#[top=-3756,bottom=-3892,left=5070,right=5238] 
as anklet name:anklet, character:, material:Silver #swtr

@Amrapali narrated http://dev2013.org/#//*[@id='content']/DIV/DIV/DIV/DIV[2]/P[4] 
at http://alipi123.blogspot.com/2013/01/acm-dev-2013_10.html#//*[@id='post-body-7823136983528562417']/p[7] 
for Software in Gujarati

@Amrapali narrated http://dev2013.org/#//*[@id='content']/DIV/DIV/DIV/DIV[2]/P[4] 
at http://alipi123.blogspot.com/2013/01/acm-dev-2013_10.html #//*[@id='post-body-7823136983528562417']/p[7] 
for Software in Gujarati

@Arvind annotated http://folkhampi.openrun.net/mural-annotation/#[top=-960,bottom=1856,left=13696,right=14432] 
as Vishnu name:Vishnu, character:Vishnu, material: #swtr

Some of them say they are annotations. Some say they are narrations. In each one, someone is saying that it is an annotation or a narration. And what it is an annotation or a narration of. The first SWeeT here can be read as: the person "Anon" says that an object at the URL http://folkhampi.openrun.net/mural-annotation/ has been annotated as "anklet". Here "annotate" is a type whose definition can be found on the site somewhere (typically at site/type/annotate). The type specific attributes (in this case that it is an anklet with certain properties) of this SWeeT can be interpreted by looking up the type information.

Similarly the second SWeeT says it is a narration of an object (could be a paragraph) at http://dev2013.org To see how one can develop a Web application using these "narrate" SWeeTs, visit http://alipi.us and look at the narrations at http://dev2013.org

These SWeeTs have an object and a type they refer to. This is not trivial for any individual to key in. These SWeeTs are typically also assisted by Browser apps that attend to what a user says and generate an appropriate SWeeT. For example, one can see that dev2013.org allows one to re-narrate and object and on publishing these alternative narratives, the SWeeTs are generated (see http://y.a11y.in/web/feeds)

More information at Sweet Web and http://swtr.us

---

IDH Knowledge Bank

http://vijayanagara.in

Socionity iiit-h

---

Other links

Follow the Sheep, http://mitan.in/followsheep and http://mitan.in/bcp/raika

POP Principles of Programming for Web 2.0 students

Pantoto Community managed community knowledge

CR and Digital Media Rights

Inventory Management - A visual web interface for small scale local industries