Topic Based Review

Rhea in her article discusses the old custodian based review vs the Topic Based Review. It is alright to say that a law-firm will review all of the data that is "potential" and would not like to take chances with inadvertant culling.
But before the review starts, a review plan has to be prepared and it is very important to know the scope of the review, magnitude of the data and the time and recources that would be required. With the current methodology law-firms use keyword searches, teams of lawyers sit together to come up with the most appropriate set of terms. But more often than not these terms are over-inclusive. Other major indicator to them is document count per custodian. All these turn out pretty pale when compared to a topic-based review assignment. Automatic clustering of documents, as Rhea says, "understands" the documents and is a much accurate predictor of it's relevance. It surely gives a more accurate picture of the review that will happen in the future.

What's in a million?

A million documents, that is. A colossal amount of words, for one. For most Indian corporations, it still means some serious IT effort implementing "Knowledge Management" solutions. For us lowly users, our 4 GB Microsoft Outlook PST contains a piffling 50,000 documents, and our desktop maybe a couple of thousand more. Of course, a million documents means some trivial fraction of a percentage point of Google's overall crawl. But in the corporation, it still appears to be... weighty, doesn't it?

Not if you ask a legal technology company, not any more. Apparently, that's a couple of days' output for these companies - on a single project at that. Some of the large players in the space - Kroll, Attenex, Stratify, Applied Discovery - have reportedly handled in excess of 20 million individual documents in a single project. According to them, this number constitutes a small fraction of the actual electronic data present in any mid-sized corporation. One executive drew attention to a recent ruling against Morgan Stanley, where the judge ruled that the corporation had wantonly done an incomplete job of locating and restoring critical historical email data. The corporation's attempted defense was that it had been unable to sift through the tens of thousands of backup tapes located in multiple warehouses. In document terms, this translates to a few hundred billion individual documents.

How, then, does one deal with a million documents? Some legal discovery surveys report that the average contract attorney (a document review expert hired for first-cut review at an average of $100 an hour) chews through 500-800 documents on a given day. That translates to close to $1,000,000 merely for the review work, notwithstanding the software and infrastructure expenses to facilitate this review. This is clearly a very expensive proposition.

In recent years, legal technology companies like Zantaz, Attenex and Stratify have each tried to bring their own approach to the problem. Zantaz, recently acquired by Autonomy, now claims to leverage Autonomy's content management capabilities along with its visualization offerings. Stratify attacks the problem using its clustering technology, where documents self-organize themselves into topical "clusters". Attenex permits the creation of "pattern rules", thus getting documents to group together based on their content.

Certainly, more companies will approach legal discovery in different ways in coming days. But given the vast volumes of electronic content being generated in more and more corporations, the thrust will shift from pure scale to facilitating more efficient ways to deal with the content storm. Already, companies with intelligent technology are fast becoming bigger players in the space. With the LPO industry gaining momentum in India, we may very well see various pieces of the legal discovery pie being cooked in this country.

Attenex, a leading e-discovery technology firm partners an Indian company for services

SQ Global Solutions, is partnering with Attenex to provide Legal review services from India.

Stratify already has its 100% subsidiary in India for e-discovery support. However, Stratify does not provide review services from India and seems to be looking to acquire or partner an LPO firm.

India: At both ends of Legal Domain

Indian companies are building strength at both ends of the e-discovery domains.

There are LPO firms that can do the first level review in less than 1/5th the cost. And then there are product technology firms that utilize the cutting edge technology to improvise the review process. This article in BPO watch elaborates further.

2006 e-discovery report

This is a small summary of a Socha-Gelbman E-discovery survey. E-discovery market was 1.5b$ in 2005, it is 3.1b$ in 2007. The market is growing at 35% YoY.

Delayed Justice

The backlog and speed of justice is a well known anathema in India. In this article Ashwinie Kumar discusses some of the symptoms of justice backlog in India comparing them with other developed countries. But more judges, stringent timeliness enforcement, higher penalties for the loser etc. alone will not suffice. We need to accentuate the depth, accuracy and speed of investigations too. Recently SEBI lifted ban on three brokerage houses though the investigation was still incomplete. It is not justified to curtail business while a lengthy investigation goes on nor is it appropriate to allow crooks to continue in business in the aegis of incomplete investigation. Forensic discovery can get us the data alright. But how do we make sense of that data. How can we speed up the process of making sense out of data. Organizing unstructured data can deal with that. Of course it will require competent tools and computing power, but that is a small price to pay for justice; delayed, incomprehensive justice is no justice.

Ludhiana City Scam and Digital Investigation

Amarinder Singh and his family might be having sleepless nights (refer: TimesOfIndia 14th June '07) but it could be an awakening for a new era of digital investigation. Now Chetan Gupta had just one pen-drive where he stored Amarinder Singh and his associated accounts' transactions. And we could hire data forensic teams to cull out the data from the (un)fortunate drive.
Now imagine if Chetan had a network of associates/partners and Amarinder Singh too had a network of accompolices; implying that instead of a few files that can be accomodated in a pen drive we were trying to hunt through millions of documents in 'n' different kinds of formats, implying that we only had an idea of what might have happened and need to figure out more but don't know how. This is not too far from reality. It's already happening in US with E-discovery becoming a major part of any litigation. It has opened doors for a host of vendors (Attenex, CA/iLumin, DOAR, EDDix, Fios, Forensics Consulting, Guidance Software, inData, Interwoven, Iron Mountain, Kroll Ontrack, MetaLINCS, Renew Data, Stellent, StoredIQ, Stratify ZANTAZ) in this space. Gartner estimates this space to grow 30% YoY through 2010. It is time the Indian Legal Sector take note.

Stratify, an e-Discovery technology vendor, in SmartTechie's Top 25 emerging technology companies

Stratify has been growing at over 100% per annum for the last two years and has focussed on providing Legal Discovery, email analystics solution. Stratify seemms to be the only Legal Discovery software Product firm with presence in India (Bangalore), though it is headquartered in the Silicon Valley.

It has recently launched its flagship product -Stratify Legal Discovery Service (SLDS) 7.0.

SmartTechie has chosen Stratify as one of the top 25 emerging technology companies in India.

At last an industry initiative on Data Security Standards

Nasscom wakes up after numerous data security thefts in outsourcing firms located in India including at HSBC resulting in loss of around $400,000

The independent self-regulatory organisation mooted by Nasscom, to ensure data security standards across member-companies, is expected to be named Data Security Council of India (DSCI).