The link to the website with the presentation at the bottom doesn't work >From: Jim Connelly <[log in to unmask]> >Reply-To: Records Management Program <[log in to unmask]> >To: [log in to unmask] >Subject: [RM] Sample Taxonomies >Date: Thu, 27 Sep 2007 08:51:29 -0600 > >Weighing in late ... with a few comments on taxonomies > >In the eighties, Canadian governments developed taxonomies based primarily >on known subject areas, the federal GRDS was an example and the BC >ARCS/ORCS and Alberta ARDA were provincial examples. For the most part >these were simplistic but workable systems and they were focused on the >hard >copy world. > >In the nineties a trend towards function based taxonomies was seen. Even >the ISO15489 "suggests" that functional designs may be more durable and >effective. These systems satisfy archives and RM staff in that they can >easily be mapped to retentions, but in most of the "real world" ... these >function based systems have yet to be proven as effective and user-friendly >systems. (In Scottish law there can be 3 verdicts - Guilty - Not Proven - >Innocent.) Most of the anecdotal evidence that I have seen to date >suggests >that users are not impressed with most of the functional designs with which >they are presented. And as audits continue in some jurisdictions, there is >mounting evidence that something is missing. Some work was done in >Britain >regarding hybrid designs in the nineties ...., Malcolm Todd's article (UK >Archives) was particularly impressive. I did enjoy Carol Choksy's IMJ >article also, as it was very thorough and will help many organizations >develop better designs, but I remain unconvinced that the methodology >outlined therein is appropriate for the current business environment. > >I believe there are three (3) reasons why our functional systems run into >problems. One is that in most commonly used methodologies, users are not >sufficiently consulted or engaged in the design process. The second reason >is that the records management world is now primarily electronic and that >ECMs and file server structures and needs (mapping, security, naming etc) >are not taken into consideration. A third reason is that the relationship >between developing business architectures and document architectures >(taxonomies) is not being defined and developed. > >In the past 5 years, I have yet to encounter an organization that had less >than 70% of its unstructured information in electronic form in various >servers/domains. In designing systems, new methods of document inventory >and analysis of server structures is necessary. Even though the old >inventory methods are needed for hard copy and remain valid, most taxonomy >designers do not delve sufficiently into the depths of our file servers. >Also, when mapping function based or hybrid taxonomies to these structures, >serious adjustments have to be made to naming, abbreviating and identifying >documents or files. Considerable work remains in these areas before RM >can >hope to be relevant in the electronic work environment. > >On my web site www.cccrecords.com <http://www.cccrecords.com/> is a link >to >a pdf of a seminar on taxonomies, I gave some 6 months ago in Calgary ... >some of the information may be useful to people working on taxonomy >designs. >Also the link to Malcolm Todd's article is there as well. Finally I >apologize to John O'Brien in advance for not taking more space to delve >into >the difference between ontologies, taxonomies and faceted classification >systems :-) > >Regards > >Jim > >Jim Connelly, CRM >[log in to unmask] >8 Oakdale Place >St. Albert, Alberta >T8N 6K6 >1-780-460-7089 > >List archives at http://lists.ufl.edu/archives/recmgmt-l.html >Contact [log in to unmask] for assistance >To unsubscribe from this list, click the below link. If not already >present, place UNSUBSCRIBE RECMGMT-L or UNSUB RECMGMT-L in the body of the >message. >mailto:[log in to unmask] List archives at http://lists.ufl.edu/archives/recmgmt-l.html Contact [log in to unmask] for assistance To unsubscribe from this list, click the below link. If not already present, place UNSUBSCRIBE RECMGMT-L or UNSUB RECMGMT-L in the body of the message. mailto:[log in to unmask]