On Sep 28 at 2:54pm, Bruce Dawson wrote:
The only real problem I see is: Maintaining the conventions of a flat
namespace.
Why do we need a convention for that?
A bit of explanation: Having a "flat wiki" implies that a naming convention
be followed for link names. For instance: SummerSummit2005 could just as
rationally be Summer2005Summit, or even Summit2005Summer. Which should it
be?
Whatever you feel like. Links and searches will get people there.
How can the convention be enforced without a lot of time-consuming
refactoring of the entire web?
Any conventions we might have don't need to be enforced, for the same
reasons. Should someone want to clean up things, the software makes it easy
to do.
How will people feel when their pages are suddenly refactored and they can
no longer find evidence to the point they are trying to assert?
I don't know how they will feel. TWiki maintains a nice revision history,
so nothing should be "lost". Beyond that, what people feel is really their
problem, not ours. :-)
The external link problem is about the only thing that really worries me.
It sucks to have someone link into OurFooBar only to find it renamed to FooBar
in the name of reductionism. But honestly, web links tend to break like
spiderwebs anyway, so I'm not sure we're really making that much of a
difference. I honestly don't think name churn is going to be *that* big a
problem for GNHLUG. And we can always use the TWiki Redirect plug-in to help
that problem (like Wikipedia does with their redirections).
I agree with the above. Having different webs is a "semantic crutch" until
some better technique comes up. There are two alternatives that I can think
of:
* Naming conventions for links (that I mentioned above).
* Having a some specialized TWiki code that automatically creates
pages and their links. An example from another site I set up:
I'm a little confused as to why we need code to create pages and links for
us. Most of the time, any topic page should already be linked to by at least
one other page. If that somehow fails, the search function should find
whatever got lost. If *that* somehow fails, I imagine it should be easy
enough to have TWiki list "orphaned" pages for us. For that matter, I imagine
someone else has already done this. :)
[blob of code deleted]
Ask me to demo this at the Autumnal meeting if you don't understand the
above.
I'm afraid I don't. I would welcome the demonstration. Despite all the
experience I have with our wiki, I don't actually know that much about TWiki.
Speaking of busy pages, should I turn on the "Release Edit Lock" feature so
that pages won't be locked for an hour after someone has editted it? I've
already bumped into it several times.
Yah, me too. I'm thinking having that release-edit-lock check box default to
on would be better. One can always turn it off if one is doing multiple
edits. Maybe reduce the timeout to something like 20 minutes, too?
Cheers,
--
Ben <[EMAIL PROTECTED]>
_______________________________________________
gnhlug-org mailing list
[email protected]
http://mail.gnhlug.org/mailman/listinfo/gnhlug-org
