[Discuss] Creating a Wiki server

[david at thekramers.net (David Kramer)]

On 08/28/2011 12:31 PM, Tom Metro wrote:
> Kyle Leslie wrote:
>> Any suggestions on a set up.  I have seen some things about TikiWiki and
>> MediaWiki.
> 
> My vote is for Twiki (http://twiki.org/).

I used to have a TWiki instance on my server for a long time, but I
ditched it because there were an incredible number of security problems.
 They were pretty good about fixing them, but one exploit hit my server
and I lost a lot of work.  This was a long time ago, so I don't know if
it's still true, but I personally would never use it again.

I have long ago switched to PMWiki (http://www.pmwiki.org), which I find
easier to use, easier to work with as an administrator (it's written in
PHP instead of Perl), and has an extensive collection of plugins.  I use
it on many projects.

> MediaWiki obviously has the advantage that almost everyone is familiar
> with it due to Wikipedia, but I find that the markup syntax is
> inconsistent enough that I still refer to the documentation, despite
> perhaps a decade of semi-regular document authoring in MediaWiki.
> 
> Can you even get a WYSIWYG editor for MediaWiki? 

That's what I don't get about MediaWiki.  It doesn't have WYSIWYG built
in, the markup is well documented, but a bit awkward, especially for
tables, which I use a lot, but it's got this reputation for being the
best.  You can add a WYSIWYG plugin though:
http://smwforum.ontoprise.com/smwforum/index.php/WYSIWYG_extension

> It doesn't seem to
> provide much of a framework for setting up a document hierarchy with
> standard navigation controls to move up/down the hierarchy. (Maybe with
> a plugin?) Instead you get a breadcrumb trail that reflects whatever
> random path you took to the current document, with no real indication
> where the document fits in or how to navigate to siblings or parents.

I don't fault MediaWiki for that, because what you're looking for works
counter to Wiki philosophy.  To state that there is a fixed breadcrumb
trail to a given page is to state that there is only one (or one
"official" way to get to that page.  Wikis are about crosslinking from
multiple places, and that's what gives them power over a simple document
repository.

Your particular usage may lend itself to breadcrumb trails, and you can
probably create them yourself, but I wouldn't expect that to be
automatic (though it is in Confluence, which we use at work, and the
breadcrumbs often get "confused" because there isn't just one path to
most documents).

For instance, in PMWiki, pages can be divided into groups, and you can
define many things at the group level, including the header and footer.
 So for my Agile New England (http://www.agilenewengland.org) project,
the PMWiki instance we use to coordinate volunteers has a separate
WikiGroup for each team, and each team's WikiGroup has a defined header
with links back to the main project page and the team's page.  We also
define authorization at the group level.

...

> Like Dan said, wikis need maintenance, and a wiki that lets you automate
> some of that helps.

Another feature I like about PMWiki is that page content is stored in
text files on the hard drive rather than in the database.  That means I
can do many administrative tasks (eg global search and replace of a
person's email address, finding all pages that have text matching a
regular expression, etc) very easily, and backups get compressed down
very small.