For information, I was given a large spreadsheet (not as big as the one mentioned, as it had only about a dozen columns but around a million rows). It took several minutes to load into Gnumeric or OOcalc. From CSV form R loaded it in 11 seconds. Then a quick "summary" almost immediately revealed a very suspicious cell with a "payment" of $ -2 billion. (This was a government accounts item.) As the colleague who asked for help said "Tomorrow is going to be a very interesting day at the office.", I'm afraid I was not privy to the consequences. Probably a fumble fingers by a data entry person, but in the meantime a lot of corrupted reports.
The message is that spreadsheets are quite good for data entry and edit, but for processing, something like R (which has good read and write csv capabilities) will be faster. JN _______________________________________________ gnumeric-list mailing list gnumeric-list@gnome.org http://mail.gnome.org/mailman/listinfo/gnumeric-list
