Centre for Internet and Society

Train the Trainer: Running effective outreach activities in India

tanvir — 2015-07-30T15:20:46Z

It is heartening to report that many Wikimedia projects in Indian languages have sustained, and even experienced an upward trend in, editor engagement. However, in terms of content creation, the majority of these projects are still facing grave challenges that put their very existence at risk.


February’s Train the Trainer program—which aims to increase the number of new editors and ‘ambassadors’ for the movement at large—proved a rewarding experience for attendees. Photo by U.B. Pavanaja, freely licensed under CC-BY-SA 4.0.

Pageview statistics for Indian-language Wikipedias are pleasantly surprising. Almost all exceed one million unique views every month—but despite these positive readership figures, very few of these readers become actively involved in the project’s communities. There is almost no increase in the number of active and very active editors on a month-to-month basis.

These statistics are alarming. They suggest a very real possibility of volunteer burnout, a dearth of second-generation editors who might continue established work, and, perhaps most importantly, the projects losing their reputation as frequently-updated and reliable encyclopedias.

The most realistic way of dealing with this problem is to bring in new volunteers who will be guided by more experienced users. They would, eventually, fill the shoes of senior Wikimedians and continue to fight for free and open knowledge.

The Centre for Internet and Society – Access to Knowledge (CIS-A2K)—a campaign to promote the fundamental principles of justice, freedom, and economic development—realised as part of its needs-assessment program that although outreach activities are being conducted to attract more volunteers to Wikipedia, they had not been as successful as expected.

To address this problem, CIS-A2K came up with the ‘Train the Trainer’ program (TTT). The program is designed to teach volunteers essential skills and abilities to, in turn, train the general public on all things Wikipedia.

These volunteers, or “trainers,” develop key competencies that will allow them to conduct a successful outreach workshop, such as public speaking, presentation skills, peer-to-peer learning, effective communication, reporting, and followup strategies.

To take part in the TTT program, it is imperative that participants be active Wikipedians. CIS-A2K is angling TTT as both a skill-building initiative amongst Indian-language Wikimedians, as well as a platform where Indian-language Wikipedians can meet and greet each other in-person. This allows participants to interact with Wikimedians from many different communities, to understand their nature of engagement, and share the challenges they have faced and overcome.

The contextual learning and exchange of ideas at these events, similar to editathons, are very special. They help participants feel like they are a part of both their linguistic community and a greater Indian-language community, opening up new opportunities of collaboration, project development, and friendship.

TTT intends to train Indian-language Wikimedians into effective ambassadors of the movement—keen and able to spread the goals and mission of the open knowledge movement. The program also strives to combine best practices from all over the world, taking cues from various chapters, user groups, and thematic organisations. It builds bridges between communities in terms of communication, encouraging partnerships and collaborations that can result in long term rewards.

Link to the original entry on Wikimedia Blog

For more details visit http://editors.cis-india.org/openness/blog-old/train-the-trainer-running-effective-outreach-activities-in-india

The vandals of Wiki

tanvir — 2015-07-30T16:22:08Z

For the better part of June 30, Kolkata was India's largest city and Bangla was the government's official language along with Hindi and Indian English.

The article was published in the Times of India on July 4, 2015. Tanveer Hasan was quoted.

The sly attempt at misinformation also featured the national motto 'Satyameva Jayate' written in Bangla script on Wikipedia's 'India' page. Many hours later, an administrator who goes by the username SpacemanSpiff landed on the page, saw all the telltale signs of a vandal attack, and scrubbed it clean of the errors. This was right after a round of firefighting on the pages of Jawaharlal Nehru and his family, where an anonymous vandal had tried to pass off the late PM's grandfather as a Muslim.

Vandalism is a fact of life at Wikipedia, an online encyclopedia that anyone can edit. Administrators like Spacemanspiff are senior editors who, over the years, have been entrusted by the Wikipedia community to protect, delete or restore pages, and even block notorious editors.

As SpacemanSpiff reminded fellow volunteers on the India notice board about the need to keep a watch on the Nehru wikis, another editor remembered that similar attempts were made in 2014. "This is not a new meme on or off-Wiki. Parallel attempts to establish that Feroze Gandhi was Muslim are even more common," writes Abecedare.

The popular site sees vandals of all kinds, says A Ravishankar, program director at the Wikimedia Chapter in India. Newbies sometimes click the edit button on a page to make some trivial changes. Some vengeful ones go on a rampage when editors reject their edits due to insufficient citation or poor sourcing. Then there are the passionate hordes that descend on their pet topics, mostly entries about religion and politics. The only way to deal with vandalism is by remaining vigilant, says Ravishankar.

What's more insidious and hard to rein in are changes by editors who are paid to twist the rules at Wikipedia so that their clients' pages sparkle. "We've had PR agencies editing pages for their clients, including actors. IIPM was a prime example of such paid editing," claims Ravishankar. The now-defunct business school allegedly got help from a senior editor called Wifione and an army of sock puppets (bogus user accounts) to keep out negative information about it between the years 2010 and 2012. They were reportedly reverted after severe backlash from within Wikipedia.

Internationally, blog writers for Wikipediocracy, a review site examining Wiki's flaws, have criticized Wikimedia Foundation, the non-profit that runs the site, for not doing much to allay allegations that Croatian and Kazakh Wikipedias have been manipulated, sometimes by the government machinery itself. Bots such as @parliamentedits (UK) and @congressedits (US) came up to track anonymous edits done from official IPs after several instances of tinkering by parliamentarians and Congressmen surfaced. A similar bot from India, @AnonGoIWPEdits, linked the Nehru edits to an IP address belonging to the National Informatics Centre.

The reason why so many people flock to get the content right on Wikipedia is the number of page views it gets. Indian users account for 4.1% of the total page views on the site. In December 2014 alone, there were 50,000 edits from the country with 'India' itself being the top edited entry for that year.

Others on the most-edited list include wikis of actors Vijay, Shah Rukh Khan, Rajinikanth and Aishwarya Rai Bachchan. Curiously, Hyderabad, Mumbai, Kerala, Mahatma Gandhi and Hinduism also made it to this list as per data generated by the Centre For Internet and Society (CIS) from the Wikimedia database. Trending news items also quickly make their way in. PM Narendra Modi's estranged wife Jashodaben got a wiki last year; so did the first Indian in NBA Satnam Singh Bhamara, says Tanveer Hasan, programme officer, CIS-Access to Knowledge.

For all this buzz, India has followed a global trend of slowdown in the number of edits on the English language Wikipedia. "When Wiki started in 2001, the first generation of editors was in their 20s. Many of them have moved out due to family and professional commitments and not many new editors have replaced them," says Ravishankar. Senior editor Tinu Cherian Abraham remembers starting off in 2005 and creating entries from scratch about anything and everything, from a historic fort near Bangalore to villages in Kerala. He also used to organize meet-ups in various cities to create awareness about the site and recruit more editors. Though he still edits, he says he doesn't have much time for anything else.

For now, the action has shifted to Indian language Wikipedias, where smaller and vibrant communities are busy creating pages and correcting errors. There are nearly 2,600 editors in English and close to 10,000 in Indian languages now. Either way, vandals have their hands full.

For more details visit http://editors.cis-india.org/openness/news/the-times-of-india-july-4-2015-sandhya-soman-the-vandals-of-wiki

Magic words in Wikipedia

tanvir — 2015-09-14T08:37:56Z

The struggle of finding solutions for replacing and retrieving content /words/facts and figures, in this day and age of machines that seem to know everything should ideally be a non-issue. Yet, for many of us who write reports based on the data available at that moment, it is nothing less than a nightmare to come to know that there has been a significant change in the data with which our reports have been written.

How does one mediate between the ever changing nature of data in general and more specifically understand the way a publicly curated knowledge ecology operates (Wikipedia can be taken as an example). It is made clear to us by the various earlier reports¹ that the amount of data exchange and content generation can be astonishingly high when one takes into acount Open Knowledge repositiries such as Wikipedia in the major languages of the world, commons.wikimedia project and other Wikimedia projects.

Our understanding of the available data can often lead us to erroneous conclusions, if one does not account for the constant updating nature of such data. The bigger risk is for a reader/researcher/user to assume that the data quoted in news reports/research or reports as static data (as against the dynamic data discussed here) and form conclusions based on the same. Another catastrophic possibility is to use the data procured in such fashion for planning and evaluation purposes. If one does not acknowledge the possibilty of change of data and plans only with the available data and does not account for the changes under contingency measures the entire planning might be off the mark and might not be successful when executed. Even for the purpose of evaluation, the constant change in data has to be tracked and monitored to appreciate the work/critical evaluation of the nature of the work.

How do we bell the Cat (read as the changing data) if not on the traditional print and digital platforms atleast in the Wikimedia universe? To do this manually is a task that is supremely tasking and prone to high error possibilities. If there is one thing researchers accross subject domains agree it is that no data is always better than wrong data. One must also think of the precious resources that would be spent on this data mining activity, the human hours, the time resources, the physical and infrastructural resources that are consumed in this process of keeping the data feed accurate and updated. I do notdeny the efficacy of systems where data mining is done manually. It is the digital researcher in me who would like to introduce to the readers a tool called 'Magic Words' https://www.mediawiki.org/wiki/Help:Magic_words used in the Wikimedia universe to plug this problem and offer researchers fewer nightmares regarding the validity of the data.

Magic words offer a one stop solution towards resolving the issues of sourcing, securing and updating our data fields. With this one can be sure that the data fields do not become obsolete and might yield to erronoeus and in worse case contradictory interpretations. If there is a research report which seeks to compare page numbers of Kannada Wikipedia with another Wikimedia project. The traditional way to do this would be to aggregate the number of articles/redirects and publish the same. The reader ends up with a number that is static in nature. Does this mean that this number is permanent, the answer is an easy no because Kannada Wikipedia is by nature and definition a live project that will be changed and added to constantly.

An ordinary reader who does not have the bandwidth to follow the researcher's footsteps in finding out the total number of articles on Kannada Wikipedia will have to be content with the same static number provided even when he knows that the number is no longer accurate. By using the magic word {{NUMBEROFARTICLES}} in a report that is written on Meta, the researcher allows the data to update automatically and changes the nature of data from static to dynamic. A classic example for the static data and the dynamic nature of the data using Magic words can be seen at https://en.wikipedia.org/wiki/List_of_Wikipedias (dynamic data represented) https://meta.wikimedia.org/wiki/List_of_Wikipedias (static data that needs to be updated manually). The change in the nature of presentation of data also implies that the facts and figures available are not just numbers but indicate factors that have driven the nature of data and influenced the formation of number of pages. Dynamic data allows us to ask interesting questions such as 'what factors contributed to the spike/decline in the number of articles' and learn from these numbers.

It becomes an essential responsibility for the researchers working with digital resources and in digital domains to broaden the scope of their research and also extend its validity to a longer course that would be difficult for quantitative research done with traditional resources and hosted on traditional platforms. Given below is a table of key magic words and its function.

Si. No	Magic Word	Used for	Used in Wikimedia Projects and Local Wiki Projects
1	{{NUMBEROFFILES}}	Number of uploaded files.	Metrics and Government Databases
2	{{NUMBEROFEDITS}}	Number of wiki edits.	Metrics and Student Evaluation
3	{{REVISIONDAY}}	Day edit was made	Metrics and Student Evaluation
4	{{REVISIONSIZE}}	The size (bytes of wikitext) of the current revision of this page	Metrics and Data inflow
5	{{NUMBEROFVIEWS}}	Number of page views.	Metrics and Website Traffic
6	{{REVISIONUSER}}	The username of the user who made the most recent edit to the page, or the current user when previewing an edit.	Metrics and Collaboratively written documents
7	{{NUMBEROFADMINS}}	Number of users in the sysop group.	Metrics and Focus group
8	{{PAGENAME}}	Full page title (including all subpage levels) without the namespace.	Metrics and Info websites

Magic words facility can be employed in many and diverse ways (as of now these are fully operational in the Wikimedia universe) if the Mediawiki software is used to build applications. For eg: Websites, evaluation programmes, databases and other applications.

Reference:

1. https://meta.wikimedia.org/wiki/CIS-A2K/Indic_Languages#State_of_Indic_Language_Projects
2. https://www.mediawiki.org/wiki/Manual:Magic_words

For more details visit http://editors.cis-india.org/openness/blog-old/magic-words

CIS-A2K financial report Jul 2014-Jun 2015

tanvir — 2016-10-27T08:48:02Z

CIS-A2K financial report Jul 2014-Jun 2015

For more details visit http://editors.cis-india.org/a2k/cis-a2k-financial-report-jul-2014-jun-2015