[pmwiki-users] Selecting a Wiki engine...
Patrick R. Michaud
pmichaud at pobox.com
Tue Oct 3 14:27:02 CDT 2006
On Tue, Oct 03, 2006 at 09:19:18PM +0200, Mark Trumpold wrote:
> What is the ETA on this current development?
If by "current development" we mean "hierarchical groups",
I really doubt it'll happen in 2006 unless we come up with
some breakthrough ideas for the design.
Pm
> On 3/10/06 9:07 PM, "Henrik Bechmann" <henrik.bechmann at sympatico.ca> wrote:
>
> > I personally think the thing to do is wait until your current round of
> > development is over, and then revisit the whole subject of hierarchies
> > in consideration of all the tools available. Yet another opportunity for
> > a magazine article!
> >
> > Thanks,
> >
> > - Henrik
> >
> > Patrick R. Michaud wrote:
> >> On Tue, Oct 03, 2006 at 12:22:54PM -0400, Henrik Bechmann wrote:
> >>
> >>> Patrick,
> >>>
> >>> See http://wiki.splitbrain.org/wiki:namespaces
> >>>
> >>
> >> Ahhh, I see. They essentially manage this by avoiding the problem
> >> of parent, ancestor, uncle, or cousin links -- i.e., you can
> >> address pages at or below the current group, otherwise you have
> >> to start at the root.
> >>
> >> Could work. It still has a bunch of issues in a PmWiki context,
> >> however. For example, a single name could be referring to either
> >> a group or a page, and it's not clear how one would manage group
> >> attributes.
> >>
> >> (I can give a longer detailed list if people think it would
> >> be helpful.)
> >>
> >> Thanks!
> >>
> >>
> >>> Personally, I think it would be more fruitful (and less brittle) to
> >>> think of the current PmWiki implementation as a *repository* of pages,
> >>> with some way of structuring hierarchies on top of that with reference
> >>> to the underlying repository.
> >>>
> >>
> >> This is what categories, trails, and other items are intended to
> >> help resolve.
> >>
> >>
> >>> My hunch is that your page text vars may
> >>> actually help in that regard (eg (:parentnode ParentGroup.PageName:)).
> >>>
> >>
> >> I think you may be right -- I'm still learning what all is possible
> >> within the context of page text vars.
> >>
> >> Pm
> >>
> >>
> >>
> >>
> >>> Patrick R. Michaud wrote:
> >>>
> >>>> On Tue, Oct 03, 2006 at 08:11:25AM -0400, The Editor wrote:
> >>>>
> >>>>
> >>>>> On 10/3/06, Thomas Voghera <thomasvoghera at gmail.com> wrote:
> >>>>>
> >>>>>
> >>>>>>> - limitation to a two-level hierarchy (groups and pages)
> >>>>>>>
> >>>>>>>
> >>>>>> Is this about how pages can be organized? carved in stone?
> >>>>>>
> >>>>>>
> >>>>> But it is *possible* this hierarchy problem will be solved soon.
> >>>>> There was a very long list discussion on the subject a few months
> >>>>> back, and the resolution seemed to be there was not a simple way (yet)
> >>>>> to adequately qualify relative links (and avoid resulting
> >>>>> ambiguities).
> >>>>>
> >>>>>
> >>>> It's possible but not likely. I've spent a *lot* of time over the
> >>>> past five years thinking about hierarchical group implementations,
> >>>> and I don't like anything we've come up with better than the current
> >>>> Group.Name system.
> >>>>
> >>>> That said, this past week I heard through a couple of off-list
> >>>> discussions that DocuWiki now has a good hierarchical grouping
> >>>> system -- if someone wanted to check it out and report back it
> >>>> might be worth looking at.
> >>>>
> >>>>
> >>>>
> >>>>> I'm wondering if in the back of Pm's mind the new syntax for *$:vars
> >>>>> was a step toward solving this problem.
> >>>>>
> >>>>>
> >>>> Nope, I haven't consciously made a connection between vars and
> >>>> hierarchical groups. Sorry.
> >>>>
> >>>> Pm
> >>>>
> >>>> _______________________________________________
> >>>> pmwiki-users mailing list
> >>>> pmwiki-users at pmichaud.com
> >>>> http://www.pmichaud.com/mailman/listinfo/pmwiki-users
> >>>>
> >>>>
> >>>>
> >>> --
> >>>
> >>> Henrik Bechmann
> >>> www.osscommons.ca
> >>> www.bechmannsoftware.com
> >>> Webmaster, www.dufferinpark.ca
> >>>
> >>>
> >>>
> >>
> >>
>
>
>
>
More information about the pmwiki-users
mailing list