Here are examples of market opportunities based on our Level One technology -- compilation, indexing, search, data mining / research in collections of up to 3.2 million words each.
Ebook readers on mobile devices can handle many of the display functions of browsers. However, access to information in an ebook is anything but precise. Instead of search, there is a scan function that looks for the next instance of a sequence of letters -- possibly a word or an exact phrase. No combinations of terms are accepted. There is no relevance ranking. This inability to handle normal search (let alone serious research) is especially a handicap for educational and reference works.
Suggested solution: Identify a potential partner with established skills in programming for mobile devices, and a knowledge of C and C++ programming languages. Work with that partner to port our Level One MarpX search to portable devices. Where the programming environment accepts compiled C and C++ modules, the task is reduced essentially to work on the interface.
It is trivially simple for us to produce Level One searchable WCT (Words Close Together) index files instead of ebooks. Customers would purchase the WCT file and download from the partner the (presumably free) app that lets them browse, read, and search -- i.e., enjoy precision access to information on a hand-held device or phone.
This represents a major discontinuous innovation in the ebook market.
Input for text destined for MarpX search and text data mining always passes through a "tagged text" stage... simple printable ASCII characters with a handful of HTML tags and subject to a few guidelines. This applies to text content from whatever source ... PDF, email, HTML, Microsoft Word, Rich Text Format, and a host of others. We visualize firms and freelancers willing to provide this service, under license from a MarpexInc partner.
Back-of-the-book indexes are prepared by freelancers, by in-house units in publishing firms, or sometimes by authors. The American Society for Indexing will assure you that there is much more to the task than meets the eye. A good index makes a great difference to the usefulness of a book over time.
Indexes are prepared normally under intense time pressure, since page numbers are not finalized until near the printing date. With text data mining, we have already found ways to help indexers make their product more accurate and comprehensive. We plan to work with a handful of indexers to see whether our notions of "incremental indexing" can reduce the time crunch and enhance quality further. If this is successful, we hope to turn the entire indexing-aids project to a MarpexInc partner.
Several methods are recommended to create value for publishers using MarpX technology. When these gain sufficient traction in the marketplace, we visualize licensing the complete tool set to multiple publishers. The Bureau of Labor Statistics reports 34,726 private publishing establishments as of second quarter, 2012. That's a considerable opportunity for a firm with marketing strength and connections to the industry. We would like such a partner.