• Company A buys a software license from Company B, a startup.
• Company A crosses its fingers that Company B doesn’t go bankrupt and disappear, along with the source code for Company A’s mission-critical software.
• Company B goes kaput.
• Company A is left with some machine-readable binary code that it is powerless to develop or use.

Source code escrow has changed the outcome of this sticky situation for the better, and here’s how: Countless software companies go out of business every year, and either their code disappears entirely or goes to another company that doesn’t do any development or maintenance on it. The concept of escrow is one way in which open source gives companies a chance to continue their contribution and innovation, because the code they wrote can outlive them and continue to be evolved by the community.  I covered this topic in my most recent post on the Network World open source subnet. I invite your feedback: what’s your experience with source code or open source escrow? Any best practices or cautionary tales to share? Looking forward to hearing from you.

All of the users who downloaded LucidGaze for Lucene in the last 24 hours are kindly requested to re-download it from our downloads page

What can the bug do?

It can impact large Lucene implementations, as it has an issue with the way the debug log is produced (may miss some stats). the fix improves the tracking of IndexReader/IndexWriter memory consumption.

Please accept our apology for any inconvenience it may have caused you.

Yours

The Lucid Imagination team

Because search applications can address an ambitious diversity of document indexes, searches, and queries, native Lucene Java application developers are often challenged to account for variations in performance of their Lucene application. It is vital to the developers to understand how quickly a native java application building in Lucene search can build/update an index, return a search result, process a query, and similar search operations.

By introducing LucidGaze for Lucene, Lucid Imagination offers the Lucene Java Application developer community a unique, application centric perspective on their search performance. While great advances have been made in recent years in java performance and java analysis at the JVM and OS level, experts such as Kirk Pepperdine (see http://www.infoq.com/articles/the-box ) recommend that analysis focus on the nature of the application behavior.

Understanding the relative efficiencies of Lucene operations is a vital task in analyzing and optimizing search, to deliver faster, more efficient results, both in terms of faster/better search output and in terms of making sure that search applications built with Lucene deliver superior business results. LucidGaze for Lucene offers a broad set of instrumentation into Lucene Java application operations, so developers can see exactly what their search app is doing, where it can be improved and where it is sensitive to various user needs. It can isolate long running queries that consume an unfair share of resources, expose inefficiencies in design, pinpoint fields or documents that cause processing bottlenecks — all in order to find where performance improvement work will have the greatest impact.

Lucid Imagination made extra efforts to make LucidWorks for Lucene easy to install and use everywhere Lucene is used. We assumed no prerequisites other than Lucene, and insisted on zero search application code changes in order to serve every stack configuration out there, to allow existing Lucene applications to readily benefit from it.

LucidGaze for Lucene is now freely available; check it out here. Please feel free to leave us feedback, either as a comment here, or directly to me, at eran.yaniv at lucidimagination.com