Although our general knowledge about open source communities is extensive, we are only beginning to understand the increasingly common practices by which corporations design software through engagement with these communities. In response, we combine design theorizing with field-study research (1) to analyze rich qualitative data from over 40 corporations participating in the Linux open source community and (2) to synthesize the observed corporate-open source community engagements into a new type of information systems design theory that we call responsive design. Empirically, we document how corporate participants in these contexts respond to market decisions, interdependent ideologies, and distributed relationships by continuously establishing and maintaining connections with community members; connections that stem from the social and material rules inherent in the open source community. Based on these observations, we create the theory of responsive design as a particular form of corporate software design which, beyond the inclusion of external participants, distinguishes itself from traditional monocentric design in which one corporation controls a dedicated team of software designers focused on solving an isolated and singular organizational problem. Guided by the principles of interconnection, opportunism, and domestication, we define responsive design as the kind of design approach that enables corporate participants to create and maintain productive design practices in response to the complex and dynamic landscapes of activities that are the foundation of corporate-communal engagements. We conclude with a discussion of the theoretical and practical implications of this new form of corporate software design.

The performance of the Coulter STKS (Coulter, Hialeah, FL) was evaluated in a busy computerized teaching hospital laboratory. The STKS was compared with a Coulter S Plus IV and manually performed 400 white blood-cell differentials. The measured blood-count parameters (i.e., white blood cells [WBCs], red blood cells [RBCs], hemoglobulin [Hb], mean corpuscular volume [MCV], and platelets [PLTs]), compared very well between the two aperture impedance-based systems; precision, linearity, and lack of carryover were excellent. The STKS WBC differential (DIFF), derived from a combination of aperture impedance, aperture conductance, and laser light scatter, also was precise; linear and carryover were insignificant. The DIFFs (n = 424) compared well to the manual WBC differentials, with r values of 0.97, 0.97, 0.73, and 0.86 for neutrophils, lymphocytes, monocytes, and eosinophils, respectively. The DIFF and Suspect Flagging system produced 6.2% false negatives and 2.6% false positives when compared with the manual technique. These were further investigated and discussed. STKS DIFFs were stable for 18 to 24 hours in normal samples anticoagulated with K2EDTA and stored at 20 degrees C prior to analysis. Storage in the same anticoagulant at 4 degrees C and immediate aspiration preserved the DIFF analysis for considerably longer than 24 hours. These performance characteristics make the STKS a significant advancement in automated hematology.

Platelet mass (mean platelet volume x platelet count) can be derived from data obtained from the routine full blood count and separates patients with myelodysplastic syndromes (MDS) at diagnosis into three distinct prognostic groups: low platelet mass group - median survival 5 months and 5-year survival 0%; intermediate platelet mass group - median survival 30 months and 5-year survival 34%; high platelet mass group median survival - not reached at 82 months follow-up with a 5-year survival of 82%. These data provide a simple rapid prognostic index at the time of diagnosis in MDS

inte rac ti o ns N ove m b e r + D e c e m b e r 2 013 Foundation. inte rac ti o ns N ove m b e r + D e c e m b e r 2 013

No abstract