[uf-discuss] Q&

[Benjamin West]

I think you guys are on the right track.  I'd like to encourage you to
do some "market research".  Start collecting examples and see what you
can distill.  Here are some questions I've got:

* Are lots of people publishing questions and answers?
  - My bias is yes!
* How are they doing it?
  - My bias favors the <dl> idiom, but it'd be interesting to find out
how widely it's used.  You might ask Ian Hixie what research he's
uncovered wrt to class="question" and its ilk.  There are also tools
my employer offers that would help with this research, but I don't
want to mention it inapropriately, and I'm working on ways to benefit
open source communities with this tool.
  - Browse around and see if we can collect a handful of idioms used
for this.  I suspect that there are a few classes of sites publishing
QnA (which we should verify through research):
    * Commercial sites offering Q&A to inform the public of their products
    * Project/personal sites offering Q&A to help with encountered problems
    * Informative sites whose focus is Q&A
  I bet we can find common idioms and patterns for publishing this
kind of material.

Finally, there are a few things keeping me from starting a wiki page:

1.) What is the scope of this format? Is it strictly questions and
answers?  Is there a slightly more general concept that would yield
much more benefit without a corresponding increase in complexity?
2.) What are the use cases for this format?
3.) Are there any other formats that cover the would-be use cases/problem space?


Finally, on a more personal note:
I'd like to encourage the community to help with this research.  There
have been some negative things going on, and this is a good
opportunity to reset our expectations:
* Be positive
* Do research
* Build consensus
* Constructive collaboration.
There will be more on this in a separate thread.

-Ben

On 12/15/06, Paul Kinlan <paul.kinlan at gmail.com> wrote:
> >>Paul, what do think?
>
> I personally think that the qa is a good idea,  I belive that you
> would be easily able to seperate questions and answers out and you
> will be able to start infering meaning from the text inside the qa
> section, however like with all microformats it is useless unless
> people use it (and if it is only you and me then there is little point
> in having a microformat because only ourselves will be publishing and
> consuming our own data). I don't belive at the moment that people will
> be bothered with microformats unless the tools are there that create
> them without people knowing about them, but obviously when you get to
> that level of integration I don't think microformats will be needed at
> all.
>
> However on a lighter note, as far as I am aware the dl, dt suffice
> (although it looks like dt is not ment for questions) I don't think
> classes are needed to distinguish questions and answers, and if this
> can start to get used by people I have lots of ideas for it.
>
> Paul
>
>
>
> On 14/12/06, Ciaran McNulty <mail at ciaranmcnulty.com> wrote:
> > On 12/14/06, Taylor Cowan <taylor_cowan at yahoo.com> wrote:
> > > This might break when there are multiple answers, not sure if one to many dt 2 dd is ok, but a surrounding <di> would help.
> >
> > One-to-many DT/DD is allowed, as are many-to-many.
> >
> > <dl>
> >  <dt>A term</dt>
> >  <dt>Another term</dt>
> >  <dd>A definition</dd>
> >  <dd>Another definition</dd>
> > </dl>
> >
> > It's a DT that follows a DD that 'starts' a new block, if that makes sense?
> >
> >
> > -Ciaran McNulty
> > _______________________________________________
> > microformats-discuss mailing list
> > microformats-discuss at microformats.org
> > http://microformats.org/mailman/listinfo/microformats-discuss
> >
> _______________________________________________
> microformats-discuss mailing list
> microformats-discuss at microformats.org
> http://microformats.org/mailman/listinfo/microformats-discuss
>