[uf-discuss] AP News microformat?

[Mark Ng]

Hi All,

I work for the Media Standards Trust, and I'm partly responsible for the
value added news/'hnews' work.  I speak for the MST, rather than the AP,
that said, I believe most of our views on these subjects are similar or the
same.  I'll also encourage the people from AP who are involved in this work
to get involved in this discussion too (and they've indicated to me recently
that they'd like to engage with the microformats community).

Some history :

In mid-2008, the Media Standards Trust won grants from the Knight News
Challenge and the MacArthur Foundation to "help improve the transparency of
news online" [1].

I got involved later in 2008.  At that time, the MST had already decided
that the best way forward was to either encourage the use of existing
microformats or to create their own.  They had an internal draft poshformat,
which I changed to be based on hatom (as significant portions of what it did
was covered by hAtom).  I spent a bit of time talking to people I know who
are involved in the microformats community to seek advice about the best way
of doing things (and got some very valuable feedback from Tantek at a
barcamp), and fired a couple of emails at the microformats lists and didn't
get much response.  I released what I had done at that point as a set of
recommendations on the use of, and a few extensions to hAtom [2], and a new
poshformat "rel-principles" [3].  This work can be seen at
http://newscredit.org/ (particularly under the development section).  This
site will soon be deprecated in favour of our latter work.

We then focussed on spending time talking to news organisations about
working with us on the format, and have had sets of discussions with
Reuters, The Guardian, BBC, AP and others with varying degrees of success.
When we approached the AP, we found that they were also looking at doing a
large release of content with some form of consistent markup - they were
looking at either using existing microformats or creating their own, in much
the same manner that the MST had been doing.  We then decided to merge our
efforts in order to not duplicate work.

After looking at our work, they were also convinced that extending hAtom was
the way forward.  As I understand it, due to internal commercial concerns,
they were not ready to be able to discuss this particularly publicly until
very recently (they're using this as part of their new content API release,
which at the time they were not discussing publicly, but which is currently
in private beta).  As we believe that expediency was important to keep this
project moving, we have released a early draft version of our work to the
outside world.

Because we intended on seeing whether our work would be accepted by the
microformats community, it's currently termed, within its specification, as
a microformat - if this document doesn't end up being taken into the
microformats process for whatever reason, the MST (and I believe the AP, but
I don't speak for them) would stop referring to it as a microformat.  If
changes proposed in this document are moved into hatom as suggested
elsewhere on the list, or a news-specific microformat, we'd be very happy.
However, we recognised that hatom is much more multi-purpose than the news
industries requirements, and can well see that alot of our proposed work
needs to sit outside the hatom specification.

We're sorry if we've made it appear as though we're trying to do this
unilaterally.  We very much want to work with the microformats community on
getting more machine-readable meaning into the news media, and I'll be
attending the microformatsdevcamp this weekend in that spirit.  On a related
note, is all I need to do to attend that camp to add myself to the wiki and
upcoming as attending ?

As a side-note, we're looking for people (preferably based in the UK,
potentially on a consultancy or permanent basis) to help us drive adoption
of (both 'lower' and 'upper' case) semantic web technologies in the news
media.  I would love to hear from you off-list if you'd like to be involved.

[1] : http://www.mediastandardstrust.com/projects/transparency.aspx
[2] : http://newscredit.org/development/newscredit-specification/
[3] :
http://newscredit.org/development/newscredit-specification/rel-principles-specification/