Re: Draft schema for QA sites

Hi Aamod,

On Thu, Feb 6, 2014 at 11:44 AM, Aamod Sane <> wrote:

> Some comments on the Q&A schema mentioned earlier on the list
> * Q&A, the complete collection of question and all answers, is not
> represented in this proposal.
>    The set could be called simply Q&A or QuestionsAndAnswers.
> was introduced in the latest update of
the proposal.

> * A question should have a field called "clarifications" where the
> questioner further
>    explains aspects of the question.
>    Any CreativeWork may have some amendments, clarifications from the
> author after
>    they see some response of the audience.

I'm not sure if this is easy to model accurately. In many cases, the author
will edit the question itself, it's sometimes easier and less confusing to
update and correct/expand the original question than having to re-explain
everything in a follow up "clarifications" field. Also, I'm not sure the QA
applications have such clarifications field. There are sometimes comments
available on the question itself (like on StackOverflow) but I don't know
of a way to distinguish between a comment on the question and a
clarification on the question. Sure, you could infer that any comment by
the original author is a clarification, but is that always the case, and is
it useful to distinguish.

> * Social features that are used in Q&A sites should find a place in the
> schema,
>    something like InterestedPersons who follow this question/answers
> session.
>    Since we already represent Authors, and given that Q&A is a useful
> unit,
>    there is already an implicit set of people associated with the work.
>    So it does not seem like too much of a stretch to include "followers"
> or similar.

I've never seen the "followers" of a question being available publicly on a
site like StackOverflow. In fact I think that most sites don't publish this
information (possibly for privacy concerns). for example allows
people to follow tickets and support questions, but only the numbers of
followers is displayed. Twitter display the list of people following each
other though, but for that there is

> * There are separate Upvote and Downvote, what about VoteCount
> (Upvote-Downvote)?
>    Are such derived fields ok in a schema?

Well, that's something that can easily be computed by the consumer based on
the upvote and downvote values. That said I'm not feeling strongly so
either way is fine by me.


Received on Friday, 7 February 2014 20:38:32 UTC