[pmwiki-users] Rating cookbooks

[Eemeli Aro]

2009/1/24 Petko Yotov <5ko at 5ko.fr>:
> Hi. I have this comparison with another online wiki community that uses and
> creates MediaWiki. They have often votes or polls on various decisions, and
> all arguments are weighted instead of just the vote count.
>
> It is however Simpler than the SimpleRating implementation:
>
>  * {support} Great recipe. [[~John]] January 2, 2009 at 16:31 pm
>  * {oppose} Didn't work for me (PHP5 required). ~~~~
>  * {strong support} I love love love love L.O.V.E. it!!! :-x ~~~~
>
>  Comments made after version 20090124 :
>  * {comment} Please note that it requires SQLite. ~~~~
>  * {support} ~~~~
>  * Simple comment: the previous user didn't argument his vote. ~~~~

This makes whole lot of sense. Especially given this argument that's
been raging about just which term to use, having a variety of options
is the only logical answer. Not sure about it being simpler than
SimpleRating, given that this thing ought to have some kind of
interface for quickly adding one's opinion.

So, I'd propose something like the following, expressed in pmwiki markup:

!! Comments
* '''support''' - Great recipe. [[~John]] January 2, 2009 at 16:31 pm
* '''oppose''' - Didn't work for me (PHP5 required). ~~~~
* '''comment''' - Please note that it requires SQLite. ~~~~

(:input form "{$PageUrl}?action=addcomment":)
(:input submit value="Add my opinion":)
(:input select name=type value=1 label=support :)
(:input select name=type value=2 label=endorse :)
(:input select name=type value=3 label=oppose :)
(:input select name=type value=4 label=dislike :)
(:input select name=type value=5 label=comment :)\\
Comment: (:input text comment:)\\
Author: (:input text author:)
(:input end:)

Not sure about what the {support}  or '''support''' markup ought to
be, but I don't think it's very important. Really, what's needed is a
minimal threshold for a recipe's user to quickly voice an opinion on
it. Is it at all important to express this data in a way that's easy
to summarize or translate to a one-dimensional "rating"? From what I
gather, the current problem isn't that accessing the comparative data
is difficult, it's the near-complete absence of that data in the first
place.

eemeli