Re: Draft schema for QA sites

On Fri, Feb 7, 2014 at 3:38 PM, Stéphane Corlosquet
<scorlosquet@gmail.com>wrote:

> Hi Aamod,
>
>
> On Thu, Feb 6, 2014 at 11:44 AM, Aamod Sane <aamod_sane@yahoo.com> wrote:
>
>>  Some comments on the Q&A schema mentioned earlier on the list
>> https://dvcs.w3.org/hg/webschema/file/7befd20efb66/schema.org/ext/QA.html
>>
>> * 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.
>>
>
> http://sdo-wip1.appspot.com/QAPage 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). Drupal.org 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 http://schema.org/follows.
>
>
>>
>> * 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.
>

I was on the fence about this, but after working on an example for the Q&A
proposal inspired from [1], I'm realizing that StackOverflow does not show
the number of upvotes and downvotes, but only the delta, which we don't
have in the proposal. Downvote and Upvote are both mentionned in the
StackOverflow API [2] but these are operations to vote which are irrelevant
for schema.org.

Reddit has an interesting take on upvotes and downvotes, and scores
(upvotes minus downvotes), where the upvotes and downvotes are fuzzy (can
be inaccurate) but the score is correct, see [3]. Reddit shows the score,
the number of upvotes and the number of downvotes, see an example on the
right sidebar at [4].

I agree with Aamod: I think we need to add a new property to Question (and
possibly to Answer) called "voteCount" or "score".

Steph.

[1]
https://stackoverflow.com/questions/4370960/what-is-attr-accessor-in-ruby
[2] https://api.stackexchange.com/docs
[3]
http://www.reddit.com/wiki/faq#wiki_how_is_a_submission.27s_score_determined.3F
[4]
http://www.reddit.com/r/AskReddit/comments/1xpkvl/what_is_something_that_doesnt_make_sense_to_you/