[Web4lib] OPAC user interface - design trends?

Mike Rylander mrylander at gmail.com
Fri Jan 27 00:17:30 EST 2006

On 1/27/06, Alexander Johannesen <alexander.johannesen at gmail.com> wrote:
> Hi there,
> > > And clicking the link for the upper-cased version of my search yielded
> > > nothing as well. :)
> >
> > I'm not surprised by that.  We haven't created our authority based
> > dictionary yet (it's a standard English aspell dictionary), and only
> > about 1/4 of our data is loaded into the development server.
> Shouldn't these things by case-insensitive? Or are there compelling
> reasons not to?

The searches are case insensitive, but the default dictionary is not. 
Proper names in the spellcheck dictionary retain their case, however,
and are seen as different words.  That will be fixed when we generate
our own dictionary, though we could probably just check the result of
the spellcheck case-insensitively and discard the suggestion if it
matches the original query.

> > [...looks in current PINES catalog...]
> >
> > And it seems we don't have any copies of that in all of PINES,
> > according to the current catalog.  ;-)
> True. But wouldn't it be great if it could find some stuff about
> Monteverdi even if it didn't have that exact one? :)

That is another search hint that's planned.  It will check for the
existance records that match each word in the original search and
suggest searches that would succeed based on that.  We don't want to
simply /do/ the search, though, as that's not what the user asked for,
and has ended up confusing users in our (informal) tests.

> > I'm most interested in what confusion there was about the back button.
> >  Had you expected the chosen library to stay, er, chosen after
> > "back"ing past it?
> Hmm, I think I clicked the link and saw the big list of libraries, but
> instead hit the back button, and it redid the whole search (meaning,
> you've put in some doo-da to ignore the browser cache). It feels a bit
> like you're trying to do DHTML things. Maybe it's just me being old
> (in internet terms, at least).

The entire OPAC is nearly all DHTML, actually.  That library tree is
just a hidden <div> that swaps out the main content of the page. 
There is a "Cancel" link above the tree if you decide not to use it,
though it may not prominent enough.

> > Thanks for taking a look, and thanks for the input!
> No worries. I'll probably put my own little experiment up here for
> poking soon too, so you can have your revenge. :) (Lucence over MARC
> as a Java app, exposed as WebServices, user interface on a different
> server in PHP, mixing in Amazon stuff, really fast on huge
> collections, usability-designed, fast development time, open-source,
> etc)

It'll be fun to compare speed /and/ features.  :)

> Regards,
> Alex
> --
> "Ultimately, all things are known because you want to believe you know."
>                                                          - Frank Herbert
> __ http://shelter.nu/ __________________________________________________

Mike Rylander
mrylander at gmail.com
GPLS -- PINES Development
Database Developer

More information about the Web4lib mailing list