On Sun, 2010-07-18 at 01:20 -0400, Jonathan DePeri wrote: > Gnumeric, most of its competitors, is based on the deeply-flawed > approach of a structureless expanse of cells. Now, there have been > "multidimensional spreadsheets" whose paradigm was to operate on > different objects -- time series, data tables, etc. -- using formulas > to relate the objects to one another without even bothering with the > expanse of cells that at any rate often results only in duplication > and errors. Thus data, formulas, and presentation were decoupled, and > great gains in productivity were possible. A number of these issues is > discussed at http://cbbrowne.com/info/spreadsheets.html -- but of > course, we're all familiar with them. > > A particularly good product was produced by Javelin Software > (http://en.wikipedia.org/wiki/Javelin_Software), but it died out. > Lotus Improv (http://en.wikipedia.org/wiki/Lotus_Improv) tried the > same approach, but also failed apparently due to lack of popularity. > The only similar product in existence today appears to be Quantrix > (http://en.wikipedia.org/wiki/Quantrix); cf. the demo and tutorial at > http://www.quantrix.com/Quantrix_Video_Demos.htm?url_text=Product_Tours. > > Anybody thought about this? We could add the whole structured, > data/model/presentation-decoupled paradigm to Gnumeric. Who said it > has to be just a traditional spreadsheet?
"but it died out" & "failed apparently due to lack of popularity". Shouldn't this tell us something? Andreas _______________________________________________ gnumeric-list mailing list [email protected] http://mail.gnome.org/mailman/listinfo/gnumeric-list
