Jon Richfield schreef:
In our by no means random, but hardly unrealistic example,several questions arise, including the role of various non-alphabetic characters, and the artificial concentration of surnames under the initial letters of prefixes such as de, der, du, van, van der, von den, and no end of etcs. By sorting by the terminal alphabetic string, we remove ambiguity and even out the spread of names through the alphabet. In simple information theory this optimises search time and sort efficiency. The above example becomes:
Aswagen Aspoestertjie Sinnerella Katrina van
Bismarkharing Otto Werther von und zu
Brien Johanna Kakebeenwania van der Merwe O'
Ewen Paulette Marmorella Bridhedia Paul-
Heever Gehardus Johannes Katwimpers Janse van Vuuren van den
Mally Jolien Gertina van der Poel O'
Swizarminife Truupsvor Theooseov
Urtel Neville McSnurtle Quentin
Ürtur Xavier Ypres Zulrich
Vandaaigoed Jakobus Johannes Joumoerus
Vanderker Lelie Belladonna Nerina
Vyfer Johannes Gehardus du Toit van der
The head benefit is in the de tailing.
Not that anyone asked.
Since you've picked a bunch of mostly Dutch and German authors or at least authors whose ancestors happened to be Dutch or German, I'd like to point out that a rather common way in Dutch databases is to do it slightly different: Sinerella Katrina van Aswagen Aspostertjie would become: Aswagen Aspostertjie, van, Sinerella Katrina This prevents alphabetically sorting all surnames from becoming a massive series of entries starting with a 'V'. I'm rather sure Marcello can provide the answer on whether our Eastern brethren do it the same. Regards, Walter van Holst