HOW ZEBRA EARNED ITS STRIPES

Start with some Danish developers curious about connecting information systems in line with open standards and skip to the last chapter, where they've become a hot little company supporting libraries across continents.

   
 
by Nancy Cohen

September 26, 2003
   
     
  So our story begins with some Danish academics back in the mid-1990s telling people they have this business where they give away software free to libraries. Don’t think there weren’t laughs galore. Today, Index Data is no laughing matter. At the time of this writing, one of its co-founders is on a plane bound for Washington, DC and the Library of Congress, to participate in a forum talking about the next evolutionary steps of standardization to become a bit more friendly to networked environments, and also to provide tutorials for programmers about his company's own tools. In fact, libraries around the world have been turning information-management challenges into solutions, thanks to these guys’ expertise.  
         
 

Their company is Index Data, a consultancy specializing in information retrieval in networked environments. Its products and services are music to librarians’ ears: software developed around vendor-independent information retrieval standards. For e-connecting libraries, the challenges involve having a uniform simple-pass search query capability for finding articles from multiple databases based on open standards and management of meta-search Internet portals.

Index Data has the software that can help do that, in tools to save on library staff time, training costs, and to build user satisfaction. Their software portfolio includes an engine that permits one user interface—without blinking banners promising free nights in Vegas—delivering information from different databases. Its Open-Source Zebra is a one-stop indexing and retrieval engine that makes use of the industry-standard Z39.50, for searching through large numbers of information systems. With the growth in the size of libraries’ information systems indeed has come the hunt for search tools to reach all the underlying data. 

 
Z39.50

What is the Z39.50 Protocol?
ANSI/NISO standard; protocol allowing client applications to query databases on remote servers, retrieve results, and perform other retrieval-related functions. 

Documentation
http://lcweb.loc.gov/z3950/agency/document.html

Benefits

"The essential power of Z39.50 is that it allows diverse information resources to look and act the same to the individual user. At the same time, it allows each information system to assume a different interface for every user, perfectly suited to his or particular needs."—"Z39.50 and the WorldWide Web," Sebastian Hammer and John Favaro.
http://www.dlib.org/dlib/march96/briefings/03indexdata.html

 
     
 

"Information retrieval using open standards and open software is where we come from,” says Sebastian Hammer, one of the co-founders of the firm. Back in 1994, when Hammer and co-founder Adam Dickmeiss, both with backgrounds in library technology, began work on information architectures, they developed this novel idea of creating a new generation of library software.  “We didn’t have any money, though, and we figured we would support ourselves by doing some Unix consulting on the side.“ 

As they continued, they became very interested in integrating components over networks, “to build a system based on various components plucked together and using interesting standards coming out of NISO and ANSI,” as Hammer puts it. In building components to handle these standards for connecting information systems, they found it natural to make this all Open Source—“which we believed would have more people to pick up on them, and to implement the standards. Furthering the cause for standards would in turn benefit us and what we were all about.”

 
         
 

Zebra was the search-engine part of the system they wanted to develop. Index Data began to draw interest from more and more projects commanding information management and discovery via Internet and intranets, and it is now a company on the move with software services, consulting, and project management. Such as ? “We've a couple of pretty major projects in the US,” he says. Both, it turns out, are in the big and ambitious state of Texas.

Hammer says they’re working with the Texas State Library and Archives Commission to build a statewide library- search application for searching databases in parallel.  He and his team are building a turnkey application for them based on Index Data’s tools. Also under development in Texas is to help build a portal in a project that will digitize presentations of everything related to Texan history.

The goal in the latter project is for Texas students, teachers, historians, and other researchers to be able to access—through one single portal—digital copies of artifacts, manuscripts, photographs, maps, letters, and other primary source materials about the history of Texas. “We were introduced to someone at the University of North Texas looking for a content-management solution for a Texas history project. There was a bid that went out, and we got in that as well." 

While their software expertise was certainly a driver, the fact that they were offering an Open Source solution did not hurt. Hammer puts that even more succinctly. Index Data could only have reached the success it has today through doing business as Open Source.

 
Zebra

What is Zebra?
Text indexing and retrieval engine, popular with librarians needing to implement the Z39.50 ANSI/NISO standard for information retrieval.  

How It Works
Reads records in variety of input formats and allows access through Boolean search expressions and relevance-ranked queries.

Licensing
GPL
Alternatives for commercial use

Benefits
User gets query-once, search-many interface to indexes. Frees library staff time; permits easy user access without having to master varied search approaches and interfaces.

 
     
 

“Early on, when we first started and told people we were giving away software free to libraries, they laughed and laughed. We just learned to present the same situation in a different way, showing people what we could do. Making our software free and downloadable made them aware we had tools they needed."

The business model is based on principles of Open Source and developed into carefully thought-our pricing models to stay viable. Hammer and team began by licensing software for free because they were in the very business of making a name for themselves as the people who work with standardized network protocols and, unlike commercial search engines, could promise the academic community solutions for building portals without the presence of banner ads and other commercial pockets of real estate. Their low-level protocol tool kits are distributed under a  BSD-like license; Zebra is licensed under the GPL but offered under alternative licensing arrangements for commercial use, not unlike MySQL AB.

Says Hammer: “I think about the revenue that we are not getting that closed source people are getting, in the form of a bunch of licensing money that we just don’t see, and I also think about a whole bunch of expenses we don’t have.  Closed-source companies have sales people, and one thinks about how much of that licensing money is going back into the sales department. There are nine of us and most of us work in development."

The company’s profit history has been marked by what Hammer calls slow growth, an achievement not to be underestimated, considering Index Data was born way before the dot-com era and managed to ride through it, steering clear of angel dust, maintaining a careful balance between resources and reach. But Hammer now sees happenings in the wings that could turn a considerable corner for them. Simply put, large organizations whose raison d’etre is information systems need more than ever to spend their trimmed budget money very carefully. “Within the past year or so," he says, "we see how government agencies have been hit with budget cuts, and they are looking at ways to save money. Open Source is looking very appealing. I see us making bids for large-scale projects. I see us bidding against close-source solution providers. We can win."

Winning will show up handsomely on their bottom line. While some of their revenue comes from service contracts linked to their Open Source software, the lion’s share of their revenue is coming from project consulting. Opportunities that could lie ahead and their scope are not lost on Hammer. Up against closed-source vendors with similar stakes in selling library-oriented information technologies, he sees Index Data as stepping up to the plate in bids for large-scale projects and winning some of those, with double-barreled ammo of technology expertise in open standards and appealing licensing terms.

Unsurprisingly, as part of its roadmap to seize the project opportunities ahead, Hammer told Open that Index Data will open a U.S. office. Also on the roadmap: placing Zebra in a marketing context that, beyond libraries and library-oriented public agencies, will speak to all kinds of organizations. Zebra is but one of numerous software products within the company’s portfolio, in business terms, but “Zebra is a sleeper,” says Hammer. “It is a fairly powerful indexing engine, and in time we will free it from the more esoteric aspects of the Z3950 for them, and put it out there as a powerful engine.” 

Also on that wish list is to revisit the company's home page to make sure that Index Data's strengths are not lost on a wider range of prospects. “This is the backside of having a whole bunch of developers and not a big marketing staff,” says Hammer, musing over the challenges ahead. 

When Open reassures him that, for a company with a whole bunch of developers, its web site is quite readable, he laughs. “You should have seen it nine years ago.”