[pmwiki-users] Recommended recipes?

[Twylite]

Thanks Ray,
> In my experience pmwiki lends itself well to starting simple, 
> sandboxing, adding functionality as you recognise the need.
>
> Unless your needs are very simple, or your mind is very disciplined, 
> you ain't going to get it right first time...and that is not a crime.
This is pretty much what I am planning to do, but I realised quite early 
in the process that there are a number of decisions that can be quite 
costly (time, effort) if I don't get them right from the start:

* Changing the Wiki  markup to 
http://www.pmwiki.org/wiki/Cookbook/Creole or 
http://www.pmwiki.org/wiki/Cookbook/Markdown is a decision that can only 
be taken before any content is created.

* Changing the linking scheme of the Wiki to an 
http://www.pmwiki.org/wiki/Cookbook/AlternateNamingScheme or even to 
disable CamelCaseLinks needs to be done from the start.  CamelCaseLinks 
could be enabled later, but not disabled without a significant 
probability of destroying existing links.

* PageStore changes such as 
http://www.pmwiki.org/wiki/Cookbook/PerGroupSubDirectories can be 
awkward to introduce once many pages exist.  Changing where uploads are 
stored as in http://www.pmwiki.org/wiki/Cookbook/UploadPlus would be 
extremely difficult to automate.

* Getting rid of Main (http://www.pmwiki.org/wiki/Cookbook/GetRidOfMain) 
or changing the group home page name 
(http://www.pmwiki.org/wiki/Cookbook/GroupHomePageChange) would involve 
renaming pages and altering links, and may not be straightforward 
depending on the size of the Wiki.  For a public Wiki this would also 
destroy URLs that may have been bookmarked.

* If I introduce say 
http://www.pmwiki.org/wiki/Cookbook/PageTableOfContents to create ToCs, 
but later find that it is buggy or doesn't interact well with some 
common/useful recipes, then I may need to change the markup on many 
pages in order to migrate to another ToC recipe.

* The previous point is generally true for any recipe/extension that 
adds markup and which graduates from the sandbox -- removing such 
extensions is non-trivial, so you want to avoid installing one where a 
more suitable alternative exists, suggesting a need to investigate 
alternatives.

I try to keep my approach disciplined, even if my mind isn't ;)

Regards,
Trevor