Indian Kanoon: Sushant Sinha on Innovation and Free Law in India

Indian Kanoon, Dr. Sushant Sinha‘s innovative free-access-to-law service for India, has recently received attention from several sources in the technology world.

First, Indian Kanoon (the Hindi word “kanoon” means “law”) has been included by La Chaire en information juridique de la Faculté de droit de l’Université de Montréal in its IDRC-funded of the sustainability of free-access-to-law services. Second, MIT’s Technology Review India in March 2011 cited Indian Kanoonin its recognition of Dr.Sinha as one of India’s “Top Innovators Under 35.” Most recently, Indian Kanoon was the topic of Dr. Sinha’s VoxPopuLII post entitled

In May, Dr. Sinha spoke with me from Bangalore about Indian Kanoon‘s technology and services, and about his views concerning innovation in legal technology and the significance of free access to law in India. The following is an account of our discussion.

Product Differentiation

We began by discussing product differentiation: How does Indian Kanoon distinguish itself from other free-access-to-law services in India, such as India Legal Information Institute (ILII) and Legal Information Institute of India (LIII)? Dr. Sinha responded by pointing to two factors: search functionality, and comprehensiveness of full-text content. The other free-law services for India “are not there,” said Dr. Sinha, in terms of inclusiveness of legal content — there is “not much comprehensiveness” in their content, he said — or sophisticated information retrieval. He underscored the role that innovation plays in this kind of function-based product differentiation. Given the pace of technological change, he said, he must continually innovate to improve Indian Kanoon‘s search functionality, in order for that functionality to remain qualitatively distinct from his competitors’ services.

Advertising

I asked Dr. Sinha about advertising, since one notable way in which Indian Kanoon differs from ILII is that the former does not display advertising on its user interface, whereas the latter does. In explaining his experiences with advertising on Indian Kanoon, Dr. Sinha noted that his advertising experiments thus far had not yielded desirable results, because the ad strategy was not targeted. “Advertising has to be targeted to be delivered” effectively, he said. Dr. Sinha said that he hopes soon to explore more targeted advertising in the form of an online lawyer directory. He explained that the availability of listings in such a directory might also offer incentives for lawyers to contribute content to Indian Kanoon‘s online forums, where citizens can ask legal questions and receive answers from legal professionals. He noted that India’s laws prohibiting lawyer advertising posed a challenge respecting designing a compliant online lawyer directory and offering other ways in which lawyers could identify themselves on the site.

Dr. Sinha noted that he promotes Indian Kanoon via social media (such as his Twitter and Facebook accounts and his blog), through the indexing of the site’s full text documents and forum comments on Google and other search engines, and via display of the “Indian Kanoon” brand name on documents downloaded from the site. Dr. Sinha told of his delight at learning recently that groups of lawyers in Delhi regularly download large quantities of PDF documents from Indian Kanoon at cyber cafés, since these documents, by displaying the Indian Kanoon brand name, serve as advertisements for the site.

The Forums

Currently, Indian Kanoon offers several online forums on substantive legal topics — where citizens can post legal questions to be answered by lawyers — as well as forums about the site’s functionality, where Dr. Sinha responds to users’ questions about Indian Kanoon‘s features and services. I noted that some of the forum discussion concerns not just the meaning of existing law, but also criticism of that law, and proposals for revising or reforming the law. In that way, the forums seem to provide a venue for public participation in policy making, albeit on a site hosted not by government but by a civil society organization. Dr. Sinha acknowledged this. He then expressed a desire to improve the functionality of the forum software, which he has written himself, and which he described as “not up to the mark.” When I asked what improvements he planned to make to the forum software, he identified “threaded discussion, email notification [when a new post has been added to a thread], voting for comments, characterization of comments, search in the forums,” and ways for lawyers to list their “affiliations” and “phone numbers.” Dr. Sinha said that these features, which he plans to code himself and to introduce in the next few months, would improve interactivity and data transparency in the forums.

Users and Usage

I asked Dr. Sinha what types of users he had originally hoped to attract to Indian Kanoon when it launched in 2008, and what types of users most frequently use the site today. “In the beginning, of course, I hoped that common people,” ordinary citizens, “would use the site,” he replied. But today, he acknowledged, the data he has gathered suggest that “more than half [of the site’s users] are lawyers or law students.”

Dr. Sinha then discussed usage data for Indian Kanoon. One month after the site launched in 2008, Indian Kanoon received “500 page views per day, and 100-200 [daily] visitors.” A year later, in 2009, the figures had risen to “8,000 page views and 2,000 users” per day. Today, the figures are “80,000-90,000 page views per day, and 15,000-20,000 users” per day; or more than 400,000 users and approximately 2.5 million page views per month.

He said that page views per person have remained steady at 4 to 5, a figure that he interprets as indicating that most users are not just downloading documents, but actually reading documents on the site.

Dr. Sinha related that “most users are from Delhi,” India’s “litigation hub,” and noted again the anecdote described earlier of lawyers’ downloading Indian Kanoon documents at Delhi cyber cafés. “Ninety percent [of Indian Kanoon‘s Web traffic] comes from India,” he explained, while “UK and US users make up 10%.”

He noted that Indian Kanoon‘s “search” page views — that is, the number of pages of lists of search results viewed by users — seem to be increasing, compared to “document” page views — that is, the number of pages of retrieved documents viewed by users. Today, he said, approximately “50% [of total page views] are search page views, and 50% are document page views,” and “search [page views] used to be lower.” He attributed this shift to improvements in Indian Kanoon‘s search functionality that he has introduced over time, improvements that he announces in the forums. (See, e.g., this post explaining changes to the search engine’s stemming function, and the expansion of abbreviations in the site’s index.)

Dr. Sinha observed that lawyers and law students had contributed much feedback respecting proposed improvements to Indian Kanoon‘s search functionality, and that most of the improvements to that functionality had been made in response to that feedback.

Positive user reception of these search enhancements also seems to be reflected in the fact that today there is “more searching on the site” than before. Earlier, most retrieval from the site occurred through queries executed in Google, which indexes Indian Kanoon‘s content. But today, a higher percentage of search queries are being executed directly on the site. In addition, today more queries take the form of “navigation queries,” indicating that more users are engaging directly with Indian Kanoon — as opposed to retrieving its content through Google or another search engine — and using Indian Kanoon‘s search and navigation functionality to identify and use desired information.

How did Dr. Sinha receive feedback from users about Indian Kanoon? I asked. He replied that, although social media services had not yielded much user feedback, the Indian Kanoon forums had produced “very good feedback,” respecting both ideas for enhancing software functionality, and the identification of content problems, such as “outdated laws.” He added that “email and phone” are also valuable channels of user feedback. He related that many of the ideas about software enhancements that lawyers had contributed — notably suggestions for adding advanced search features, such as restricting search queries to the “author” field (to enable searches of opinions by particular judges) and the “citation” field (to enable quick retrieval of known sources of law) — had come via telephone and email inbox.

Technology

Dr. Sinha then described the technology underlying Indian Kanoon. Scraping programs that he has written harvest — on a daily basis — full-text court decisions, administrative decisions, statutes, and regulations from dozens of Indian federal and state government Websites, and report in a log problems with or changes to these government sites’ interfaces. I asked Dr. Sinha about the challenge of frequently updating scraper programs to fit constantly changing government Websites. He acknowledged that at first this had been quite difficult, but that it had recently grown much easier, due to automatic notification of government Website changes via the log, and due to the relative stability of many government Websites, especially those of the judiciary. “Many [government Websites] are pretty stable, pretty consistent,” he said.

I asked Dr. Sinha whether he believed that the Indian national and state governments would continue in the future to publish full-text law free of charge on their Websites. He responded optimistically, noting what he perceived to be a strong commitment on the part of many Indian government technologists — and particularly the “in house technology” personnel at India’s Ministry of Law and Justice and the Indian national judiciary, to whom, he said, “much credit must be given” — to make full-text law available free of charge to the public. I asked Dr. Sinha whether he had met with government technologists to discuss harvesting data from their sites. He said that no such discussions had occurred, and that Indian government technologists tended to keep their distance from civil society technologists like himself. He asserted that, nonetheless, he had ascertained indirectly that many of these government technologists were committed to making law available to the public free of charge via the Internet.

After Indian Kanoon‘s scrapers have harvested full-text legal documents from government Websites, the documents — which originally are in a variety of formats, including PDF, HTML, and word processed formats — are processed and stored in a PostgreSQL relational database, which Dr. Sinha has customized with some of his own code. (Click here for details on these modifications.) Some of the document metadata are stored separately in XML, Dr. Sinha said, but the main full text documents are not marked up in XML.

PostgreSQL also furnishes the search engine for Indian Kanoon. As noted above, Dr. Sinha has modified the search engine with his own code, in order to customize it to accommodate queries from non-lawyers, and in response to requests by users who are lawyers or law students. Notable functions of the Indian Kanoon search engine include stemming, expansion of abbreviations, and relevancy ranking. The advanced search page lets users search by keyword, words in the title, citation, author, venue, and date; limit the search to statutes, court decisions, administrative decisions, or other material types; and specify whether results should be sorted by relevance or date. If the latter function is not used, search results by default are ranked by relevance, but the results display allows users to re-sort results by date — from most recent to oldest, or from oldest to newest — with a single mouseclick.

Respecting document output, Dr. Sinha uses HTMLDOC open source conversion software to deliver individual documents as HTML and PDF. Individual documents are output initially as HTML, and the interface displays a “Get the document in PDF” button that enables rapid delivery of a PDF version of the document.

Other components of Indian Kanoon‘s architecture include Django open source Web applications software, the Linux kernel, and the Apache Web server.

Respecting new technology to be added to the site in the near future, Dr. Sinha identified — in addition to the improved forums and possibly a directory of lawyers — an online document assembly service. This service, which Dr. Sinha is developing from scratch, would enable legal professionals, for a fee, to assemble and download litigation, transactional, or advisory documents that include content from Indian Kanoon. Dr. Sinha said that he hopes to launch this new service in the next year or two.

Open Source

I asked Dr. Sinha whether, as the operator of a free-access-to-law service, he felt an ideological commitment to using free, open-source software (FOSS) in the architecture of that service. His answer was complex, and reflected an interesting perspective on the role of innovation, entrepreneurship, and information sharing in the free-law sector. Dr. Sinha acknowledged that practical considerations played a “very important” role in his decision to build Indian Kanoon with FOSS. Yet he said that ideology also played some part in this decision. Characterizing the entire Indian Kanoon code base as “open source,” he observed that he adheres to a principle of publicly disclosing all of his original code (much of the code has been posted on Google Project Hosting) — including the code for the forums, which he has developed from scratch — and, where that code consists of modifications to FOSS, of always contributing the code back to the relevant FOSS community. (Dr. Sinha provides more detail in this post.) He admitted to taking pride in the fact that several of his original PostgreSQL patches had been incorporated into the main PostgreSQL code. He said that his motivation for fully disclosing his code and contributing it back was to participate in the common enterprise of innovation that characterized these FOSS communities. Open sharing of code “helps a lot of other products to come in, new products that could not be envisioned” by programmers working entirely on their own, he observed.

Dr. Sinha further underlined the benefits of the division of labor within the FOSS communities. Each developer acquires expertise in particular areas or functions, or gains in-depth knowledge of the needs of particular user communities, and that distinctive knowledge gives rise to fine-grained innovations. If all of the resulting innovations are shared with the entire FOSS community, then community members can pick and choose from a great variety of finely customized features, to meet their particular local needs. It would be inefficient for a programmer who is unfamiliar with users of legal information to create new software features adapted to those users’ needs, he observed. “I’m the best person to do that,” he concluded.

“The Thirst for Law”

In , Dr. Sinha interpreted the Indian Kanoon usage data as suggesting a remarkable hypothesis: that the provision of free access to law via the Internet was actually increasing demand for access to law among ordinary citizens; Dr. Sinha called this rise in demand “the thirst for law.”

I asked Dr. Sinha about the evidence that had given rise to that hypothesis. He replied that he had observed in the Indian Kanoon transaction logs a consistently high percentage of “descriptive queries” — search statements consisting of non-legal language used to represent legal issues. He noted that he was struck by the contrast between these queries and the many queries in the log that used legal language with great precision. He concluded that the descriptive queries must be entered by non-lawyer citizens with no formal knowledge of law, whereas the precise legal queries must be coming from legal professionals, or citizens with substantial legal knowledge. In addition to this, he said that he had seen much anecdotal evidence, in the form of comments in the forums — such as this — and direct communications from users.

But the most persuasive evidence, he said, came from two kinds of usage statistics. The first is the very large numbers of users — which Dr. Sinha interprets as indicative of a user community numerically much greater than the community of lawyers in India. The second is the consistently high rate of page views per user, which suggests that a very large percentage of users are reading full-text documents online. Dr. Sinha expressed his belief that the latter behavior is more characteristic of non-lawyers than lawyers, who might be expected most often to search for and download known-item documents to be read off-line. Dr. Sinha concluded that, taken together, this evidence suggested that a very large and growing number of users of Indian Kanoon were non-lawyer citizens who had little knowledge of the law, and whose interest in acquiring legal information was being fueled by free access to such information through services like Indian Kanoon.

If we assume this theory to be correct, I then asked Dr. Sinha, were there particular attributes of free-law services that might foster or inhibit this “thirst for law”? Dr. Sinha pointed to the two factors he identified in his VoxPopuLII post as encouraging the increase in desire for access to law: the absence of paywalls, and improved search technology. Dr. Sinha speculated that the mere removal of pay barriers initially attracts large numbers of non-lawyer users to a free-law service. Then, when non-lawyers use the high quality — and very “forgiving” — search functionality of the site, their experience of quickly receiving relevant information, coupled with the absence of a charge for that information, gives rise to a desire for additional legal information.

Dr. Sinha’s notion of the “thirst for law” reminded me of the role that knowledge of legal rules plays in Professor Amartya Sen’s theory of “development as freedom.” In his book of that title, Professor Sen asserts that citizens’ access to law, and knowledge of law, are necessary conditions of certain basic human freedoms or capabilities for living. These include the freedom to participate effectively in markets, to make informed political decisions, to hold government officials accountable, and to vindicate one’s rights to basic social services such as education, health care, and emergency food supplies. This framework suggests how a “thirst for law” might arise in a developing nation: As citizens and small business owners use free-law sites to increase their knowledge of laws that relate to their lives and their freedoms, those users may come to believe that such knowledge of law contributes significantly towards increasing their freedom and their capacity to live well. That belief could well create a desire for access to other laws that relate to the users’ lives and freedoms. Thus, within the “development as freedom” framework, free access to law via the Internet could generate increased demand for access to law.

I asked Dr. Sinha whether he thought the “development as freedom” framework was consistent with the evidence he had seen for the “thirst for law.” He acknowledged that he had “no evidence” directly supporting the “development as freedom” explanation for the “thirst for law” phenomenon. However, he said, he found the explanation quite plausible, because of particular attributes of India’s governments and society. Free access to legal information, he asserted, “is far more important in India” than in the U.S., because of the vast social inequality, great disparities in power, and the pervasiveness of government abuse and corruption in India.

Dr. Sinha illustrated this point by citing two examples. The first concerned Mohammad Adil Hossain, a student in mass communication at Aligarh Muslim University, who, after criticizing university officials online, reportedly was suspended and banned from campus in 2010 on the ground of “‘tarnish[ing] the image of the University by resorting to and misusing the internet.” During the administrative proceedings, the university purportedly refused to disclose its evidence to Mr. Adil Hossain. Mr. Adil Hossain researched the law, using services including Indian Kanoon, and found precedents in which courts had upheld students’ freedom of expression and required due process in university administrative proceedings respecting discipline of students. Mr. Adil Hossain appealed the university’s decision to the Allahabad High Court, which reportedly reversed and remanded the university’s decision. On remand, the university reportedly “revoked the suspension” and reinstated Mr. Adil Hossain.

Dr. Sinha then related a second circumstance, in which gated communities in India purportedly assert control illegally over public roads, and prevent members of the auto-rickshaw union from using those roads in the course of business. In his recent post on this topic, Dr. Sinha explained that he had researched the law, and found that the applicable law prohibited assertions of private control over public roads, and that such assertions of control may also violate the right to freedom of movement protected by the Indian Constitution. This legal information could potentially be used by members of the auto-rickshaw union to compel gated communities to open their public roads to union members and other members of the public.

Both of these instances, Dr. Sinha contended, were consistent with the “development as freedom” view of free access to law. Not only did free access to legal information enable citizens — actually in the first instance and at least potentially in the second — to vindicate their legal rights. One could also plausibly argue that the citizens in both examples could experience such access as affording them knowledge that increases their freedom, their capability to live fully, in one instance free from unfair restraints on speech and from unfair treatment by university officials, and in the other, free from impediments to their rights to move freely and to earn their livelihood. “This [free access to law] is very important” to ordinary citizens, he concluded.

Partnerships

I noted that, after several years of operating independently, Indian Kanoon had recently reached out to another organization: Dr. Sinha agreed to allow the new Legal Information Institute of India to harvest content from Indian Kanoon.

I asked whether Dr. Sinha foresaw Indian Kanoon‘s engaging in additional partnerships in the near future. Dr. Sinha replied that Indian Kanoon was currently involved in such a partnership, with PRS Legislative Research, which provides research and legislative history services to members of India’s national and state parliaments. In its partnership with PRS, Indian Kanoon is adding to its content the full text of debates of the Indian National Parliament from 1998 to the present. Dr. Sinha said that this content was currently being loaded and indexed in Indian Kanoon, and should be available in the coming months.

This partnership with PRS illustrates an interesting kind of relationship that often arises between civil society-based free-access-to-law services, and government users of legal information. As Tom Bruce recently noted at the Princeton Law.gov Workshop, many free-access-to-law services have reported — to researchers conducting the IDRC-University of Montreal study described earlier — that government employees constitute one of their core user groups: Many government personnel find that free-law services offer access to law superior to that offered by the government itself. Through its partnership with PRS, Indian Kanoon will offer additional evidence of this phenomenon by delivering key legal information to Indian government officials — Indian legislators and their staffs. By enhancing a government’s access to its own information, free law services like Indian Kanoon can be said to serve as components of that government’s own technology services — its “e-government” services. By supplementing inferior government information technology, free law sites can contribute to the efficiency of governmental operations and the improvement of governmental services offered to citizens. This kind of complementarity between civil society-based free-law services and government seems consistent with the model of public-sector information provision set forth by Robinson et al. in their influential paper, Government Data and the Invisible Hand.

Business Models and Sustainability

I asked Dr. Sinha to describe his current business model, and how he expected that business model to change in the coming months and years. Dr. Sinha answered that he currently funds the operations of Indian Kanoon entirely from his own resources. He said that he plans in the coming months to develop revenue streams for Indian Kanoon, notably the fee-based online document assembly service and, possibly, the lawyer directory service mentioned earlier.

Dr. Sinha set out a two-tiered business model for Indian Kanoon. “Search [of Indian Kanoon‘s full text primary legal content] will always be free,” he affirmed, explaining that charging for access to primary legal information “doesn’t make sense” and “is not going to help people.” However, he acknowledged, value-added services on Indian Kanoon — such as the planned document assembly service, and possibly also access to legal commentary, “judgment summaries,” or other kinds of secondary legal information — may be provided in the future on a fee basis. Revenue from such sources are intended to make Indian Kanoon sustainable in the long-term.

I observed that this two-tiered approach resembles models being adopted by other organizations in the online legal publishing sector, in which services for retrieving primary law are considered to have become commoditized. As a result, large, for-profit legal information providers are abandoning those services and concentrating on more lucrative value-added services, while smaller for-profit providers are experimenting with free or low-cost primary legal search, funded or supplemented by advertising or by moderately priced value-added services. Dr. Sinha affirmed that “Indian Kanoon is a for-profit” enterprise, and his business model going forward is consistent with that view.

Conclusion

Several themes emerge from Dr. Sinha’s account of Indian Kanoon. First, innovation is a value that informs several aspects of the organization. In Dr. Sinha’s account of Indian Kanoon‘s distinctiveness as a product, superior search performance is the key, but constant innovation is required to maintain that superiority. Dr. Sinha identifies the opportunity to participate in the culture of innovation that characterizes the open-source software development community as one of his primary motivations for adopting an open-source approach to Indian Kanoon‘s technology. The challenge of innovation also marks Dr. Sinha’s decision to create and improve Indian Kanoon‘s forums using his own, original code, rather than relying on software created by others.

Second, and relatedly, Dr. Sinha adopts an entrepreneurial approach to legal information provision. His model has several dimensions, including a desire to serve the needs of ordinary citizens; a willingness to take substantial risks, including funding the organization in its initial stages entirely with his own capital, creating key services with original code, and offering all code back to the community as open source; an interest in filling a market niche; a willingness to continually adapt the service to accommodate the needs and desires of key users and to stay one step ahead of competitors; a strong interest in operational independence coupled with a willingness to enter into selected, beneficial strategic partnerships; and a basic, long-term profit motivation.

Third, Dr. Sinha insists that a commitment to serving the public interest — by providing free, high quality access to primary law — is consistent with a for-profit business orientation. He argues persuasively that, particularly in the Indian context, citizens must have free access to law in order to vindicate their legal rights and enjoy their fundamental freedoms, and that both government and civil society organizations have an obligation to facilitate that access. But he also asserts confidently that such a commitment to free law accords fully with the view that charging fees for value-added legal information services or access to secondary legal information created in the private sector is appropriate. In Dr. Sinha’s view, fee-based services — by providing revenues that enable sustainability — facilitate and complement free-law services.

Within this framework of innovation, entrepreneurship, and commitment to public service rooted in sustainable business practices, Dr. Sinha envisions a bright future for Indian Kanoon.