The objective of this paper is to determine under what circumstances individual organizations would be able to rely on cross-company-based estimation models. We performed a systematic review of studies that compared predictions from cross-company models with predictions from within-company models based on analysis of project data. Ten papers compared cross-company and within-company estimation models; however, only seven presented independent results. Of those seven, three found that cross-company models were not significantly different from within-company models, and four found that cross-company models were significantly worse than within-company models. Experimental procedures used by the studies differed making it impossible to undertake formal meta-analysis of the results. The main trend distinguishing study results was that studies with small within-company data sets (i.e., $20 projects) that used leave-one-out cross validation all found that the within-company model was significantly different (better) from the cross-company model. The results of this review are inconclusive. It is clear that some organizations would be ill-served by cross-company models whereas others would benefit. Further studies are needed, but they must be independent (i.e., based on different data bases or at least different single company data sets) and should address specific hypotheses concerning the conditions that would favor cross-company or within-company models. In addition, experimenters need to standardize their experimental procedures to enable formal meta-analysis, and recommendations are made in Section 3.

Inspections can be used to identify defects in software artifacts. In this way, inspection methods help to improve software quality, especially when used early in software development. Inspections of software design may be especially crucial since design defects (problems of correctness and completeness with respect to the requirements, internal consistency, or other quality attributes) can directly affect the quality of, and effort required for, the implementation. We have created a set of “reading techniques” (so called because they help a reviewer to “read” a design artifact for the purpose of finding relevant information) that gives specific and practical guidance for identifying defects in Object-Oriented designs. Each reading technique in the family focuses the reviewer on some aspect of the design, with the goal that an inspection team applying the entire family should achieve a high degree of coverage of the design defects. In this paper, we present an overview of this new set of reading techniques. We discuss how some elements of these techniques are based on empirical results concerning an analogous set of reading techniques that supports defect detection in requirements documents. We present an initial empirical study that was run to assess the feasibility of these new techniques, and discuss the changes made to the latest version of the techniques based on the results of this study.

Recently the awareness of the importance of replicating studies has been growing in the empirical software engineering community. The results of any one study cannot simply be extrapolated to all environments because there are many uncontrollable sources of variation between different environments. In our work, we have reasoned that the availability of laboratory packages for experiments can encourage better replications and complementary studies. However, even with effectively specified laboratory packages, transfer of experimental know-how can still be difficult. A cooperation between Brazilian and American researchers addressing effective running of replications was formed in 1999. One of the specific issues being addressed is the problem of transferring tacit knowledge. We discuss what we learned about the tacit knowledge transfer problem and the evolution of laboratory packages in the description of a replication performed in Brazil using a PBR (Perspective Based Reading) laboratory package; also how further issues are addressed