In seeding the market with a free introductory downloadable metadata catalog tool, Tamr may be unwittingly tackling one of the most stubborn hurdles to traditional master data management (MDM): the fear among data owners of losing control over their data sets.
Cataloging data doesn’t mean you have to say you’re sorry
Data wrangling – the process of making unfamiliar data sets usable for exploratory analytics – attacks the problem that most data is what Tamr characterizes as “dark.” The content, structure, and sometimes the origin of such data sets are not as explicit as those from familiar enterprise data sources, and as such they have traditionally been outside the reach of business intelligence and analytics.
Tamr tackles one portion of this process: cataloging and matching sets of data. It does so with a technology that leverages a combination of crawling, machine learning, human feedback, and collaboration to catalog and provide a preliminary matching of data. These processes complement rather than replace traditional MDM processes.
In contrast to MDM, Tamr’s approach does not strive to standardize or designate any data source as “golden” as MDM systems do. Rather, it encourages owners of data to share their data sets – and their knowledge about their data sets – minus the fear of ceding control to a data czar who will dictate what the data should look like or which source(s) are the “golden copies.”
In its second release, Tamr is adopting a common strategy of offering a more modestly packaged, free version of its offering. Specifically, it is releasing for free download a tool that performs just the cataloging portion of the process without the crawling capability. By making available a free community version, Tamr could be unleashing a tool that could help business users share data with the assurance that good deeds, in this case, will go unpunished.
On the Radar: Tamr, IT0014-002934 (August 2014)
Tony Baer, Principal Analyst, Ovum IT Information Management