> What do you think about future versions of swish-e? ...
I'll confess to being slightly alarmed (as well as excited by the
possibilities!) One of the great strengths of swish-e to date has been that it
is moderately easy to install and use. This is a great help to the many web
admins like me that don't really know C -- I know just enough to tinker with
obvious things and to compile and install. Adding many additional features is
going to make things a little harder for us "dummies", so a good, well
organised Makefile and instructions are going to needed too.
Now, a couple of things I'd like to see included:
1. An option to limit indexing of html files to just the header -- in fact
just the metanames. If you've built pages with good metadata in their header,
then you don't want to also index the content -- because that's what the meta
data is for! [I apologise if this has been done already -- haven't been paying
attention for a while.]
2. Include support for rdf files! Then I can put my metadata into rdf instead
of the html header, and I can also index non-textual files (e.g. images). If
you do this, you also need to have the search return the URL of the actual
resource, instead of (or as well as?) the rdf filename.
3. Split the indexing and searching into two separate programs. They do
different things, so why not have two progs?
All the other suggestions are good too, I'm sure -- but the above, especiall 1
& 2, are what I need.
Received on Wed Jul 26 03:19:47 2000