I'd like to start off some discussion of how DSpace can provide better statistics on its content and functioning. Here are some thoughts to spark the conversation:
o "Statistics" is a broad and diffuse topic. We will probably identify more than one distinct area of focus. What contributors need to see is not what the site's providers and champions need to see, is not what senior administrators need to see, is not what sysadmin.s need to see, is not what Promotion and Tenure committees need to see.... o Different sites may well have different bundles of needs -- different enough that one package may not cover both sets of requirements. At least in the short to medium term, we may profit from encouraging several complementary approaches. o It may be possible to slot in more than one package within a single repository. Maybe "statistics" should be significantly modular within itself. o There's a lot of good work out there already. We may get good results sooner from contributions to some of them, than from an attempt to start from scratch. o There are excellent external tools, whether you want to produce giant all-encompassing reports or run simple ad-hoc experiments. How can we support people who want to siphon DSpace's observations off into SAS, SPSS, R, Stata, Excel, whatever? -- Mark H. Wood, Lead System Programmer [email protected] Friends don't let friends publish revisable-form documents.
pgpyDRBSQpcmP.pgp
Description: PGP signature
_______________________________________________ Dspace-general mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/dspace-general
