On 11/7/06, Tom Horsley <tomhorsley@xxxxxxxxxxxx> wrote:
Treating the names that happen to have digits as part of their string content differently than the ones that have no digits makes no sense at all.
I disagree. Numbers in model names can and do represent incremental revisioning in a numerical scheme.. depending on the vendor. I won't even get started in the complication of how generally expected and intuitive alphanumerical sorting rules differ for different language groups.
(The first thing I do when I install Windows is go into explorer and turn off the "intuitive" sorting since it isn't remotely intuitive and results in hours of extra time trying to find files :-).
Perhaps the solution that best fits for people who think differently than me, people like you, is to have a text entry box where you get to enter the model string(perhaps with wildcards) so you can get a short list of known printer models which best match your string fragment, so that you can avoid having to scroll through the full list of HP printer models. -jef"There is no single 'intuitive' way to parse and sort any collection of data which is not programatically consistent. To debate such details as which lexicon rules to use to sort printer model names for presentation misses the larger point entirely... which is.. no pre-defined sorting ruleset will not be the 'best' way to parse data for everyone. We should suppliment the sorted list with other. less structured forms of information retrieval, which appeal to the different 'intuitive' ways that different people think"spaleta
