Re: Wiki Classification Schemes [Was: Page rename crashed second time around]

Murray Altheim <[email protected]>
Newsgroups gmane.comp.java.jspwiki.user
Message-ID <[email protected]>
Claudia Frers wrote:
[...]
> In short I just want to have some way of keeping track of what I am doing as
> the number of pages grow and I am not sure that i trust that a search will
> make the right page pop up.

Hi Claudia,

Yes, a common issue as the wiki grows.

> thus I have been poking around Multiwikis and categories and namespaces and
> i assume that it's all about trade offs....

To some extent, yes, complexity vs findability. Trying to find a balance
that doesn't require users to learn classification schemes, use complex
and problematic syntax (ala the "semantic wiki" solutions), or create a
nightmarish world of rigid rules, well, there's gotta be an 80/20 solution
here somewhere.

> Will copy my answer to him here as well:
> Dirk  wrote:
> "
> 
> Due to the absence of a hierarchical folder concept in JSPWiki, I am using
> extensively a dot syntax to group related pages together. (similar to the
> java package concept) Therefore, I needed a way to generate Index pages of a
> subset of wiki pages, based on a certain Namespace or page prefix.
>
> "
> Perhaps I am using the wrong plugin. The Namespace Index Plugin seems to
> give me what I want.
> But I still need to test it.
>
> The multiple categories is a good second choice if I cant get the
> NamespaceIndex Plugin to work.

I've not tried using the NamespacePlugin so I can't really comment on how
it works in a production environment, i.e., on whether people can figure
out how to use it effectively. My criticism of any such system is that it
does then rely to a great degree on designing and following a controlled
vocabulary of terms. For example, in the example on the NamespacePlugin
page it uses

    Zoo.Birds.Flamengo

which points out a few problems. First, why is the top level category
"Zoo"? In most zoological taxonomies the level above birds (the zoological
"Class" of "Aves") would be "Order". If one wanted to skip the various
levels the next might be "Animalia", but skipping levels would be a
problem if at some point there needed to be something between "Zoo."
and ".Birds". In the case of using "Birds" there are potentially name
conflicts with other things that use the same five characters for their
common descriptor (since "Birds" is a non-scientific, colloquial term for
"Aves" that provides no context and is only expressed in English, and in
plural -- should it be expressed as a singular or plural term?).

Then there's the issue that "Flamengo" is misspelled. I'm assuming
someone really meant "Flamingo", but I might be wrong: perhaps "Flamengo"
is a common name of some other bird or the name of the flamingo in a
different language than English. There's also six species of flamingo,
which one does this refer to?

I'm using the above example to point out the various kinds of problems
that *will* arise in any use of a simplistic classification system (i.e.,
my questions are rhetorical). There can be no simple solution to a very
complex problem. We need at a minimum either a means of expressing scope
(aka context) or having great flexibility in our navigation and search
(e.g., potentially permitting "flamengo" for "flamingo").

> To have to remember to add a link on each page is a bit of a strain on the
> old memory. If I forget to add a link to a page, that page can be easily
> forgotten. For me, its easier to attach the category as a name
> Zoo.Birds.Stork

Or how about having the page inherit characteristics, so if you're on the
"Birds" page you'd only have to attach "Stork"? So that all of the
characteristics of Birds are also true of Storks...

You may also look into using the QueryPlugin which includes regular
expression matching on wiki page names. It probably could be used
as an alternative to the NamespacePlugin, but either of them would
have many of the same issues, and would require users to learn an
arcane query syntax in order to obtain query results. Another possibility
would be to set keywords in page metadata (like tags) and query on
matching keywords/tags. There's stuff in the works for that but
nothing available right now that I'm aware of.

Murray

...........................................................................
Murray Altheim <murray07 at altheim.com>                           ===  = =
http://www.altheim.com/murray/                                     = =  ===
SGML Grease Monkey, Banjo Player, Wantanabe Zen Monk               = =  = =

       Boundless wind and moon - the eye within eyes,
       Inexhaustible heaven and earth - the light beyond light,
       The willow dark, the flower bright - ten thousand houses,
       Knock at any door - there's one who will respond.
                                       -- The Blue Cliff Record
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.