Open Source Search and the Analysts’ Radar

Two analysts reports were published recently by specialists who focus on the search space. Lucene/Solr search (and Lucid Imagination) figured in both of them, in a manner of speaking.

First, Ovum published a report on the state of enterprise search, which you can download here; (I’ll let you draw your own conclusions about the relationship between the organization that hosts the report and its evaluation in the report). While the scoring methodology favors bundled solutions with features you may or may not need, they offer the following interesting observations about Solr/Lucene:

Apache Lucene Solr 1.4 is rated ‘explore’ in this assessment. Lucene Solr must be appreciated for what it truly represents: a strong open-source library and toolset that is the most customizable of all solutions in the market. Lucene Solr 1.4 is truly enterprise grade and its open source code can be immensely helpful for organizations that are trying to explore facets of search not provided by commercial vendors. Ovum strongly recommends it for large organizations that have an expertise in development and coding and sufficient IT manpower to customize the solution to their unique needs. The solution is also recommended for IT departments that are true ISVs in themselves and can manage large scale projects internally. Even for routine tasks such as the simple indexing of a massive library of documents running into billions of documents, this solution can prove extremely effective given that the end-user will not need to pay by the number of documents indexed (as opposed to pricing models followed by the likes of Google).

The second analysts report comes from an organization you may have heard of, called The Gartner Group (they used to publish a ‘magic quadrant’, now replaced by a ‘market scope’). Their analysis of the pace can be found here (again, draw your own conclusions about how the organization sponsoring publication fared in the report). While the Ovum report focuses on technologies and vendors, Gartner constrains themselves to vendors alone. Here are their ratings criteria:

I, for one, find it interesting that with all these criteria about organizations and vendors, little attention is paid to technologies. Perhaps this is driven by the fact that Gartner’s subscribers want to know who to buy from, not what to buy. Odd, to say the least, that one might believe that _what_ one buys matters less than _who_ one buys it from. Could this be a weakness of the approach? Let the following picture allow me to skip the next thousand words:

Source: http://www.slideshare.net/sogrady/meet-redmonk-analytics

I don’t know about you, but when I shop, I like to know what I’m going to buy and use, not just who’s selling it. And there’s not a lot of mystery about the use of open source technologies, not least Lucene/Solr. Though to read these reports, it seems some might wonder if the jury is still out. (It’s not; there’s an analyst report for that, too).

What does it take to help decision makers to see the value and opportunity in open source search technology? I’ll give the last word to Ovum:

For any open source product to be considered enterprise grade, possessing a viable and strong third party service backbone is necessary. Apart from a great product, enterprises need guarantees of assisted deployment, support, bug fixes, updates, and ongoing consulting.

That sounds familiar: a great product, guarantees of assisted deployment, support, bug fixes, updates, and ongoing consulting.