[uf-discuss] GRDDL (Microformats->Semantic Web) call for comments

[Harry Halpin]

Hey everyone,

Has anyone here heard of GRDDL? It's a simple method of converting
microformats to RDF, the basic data format of the Semantic Web, and then
one can use the Semantic Web toolset to do some interesting things like
merge microformat-based data from different sources and then query the
"mashed-up" microformat data. GRDDL allows one to link a transformation
to RDF (either from the document directly or the profile URI) to some
micro-format enabled data. The technique also works with plain-vanilla
XML documents. I think with GRDDL, we get both additional value from
microformats and the Semantic Web.

I'm the chair of the W3C GRDDL WG, and I want for people who are
active in the microformat community to give us comments on our
specification. We would highly value your comments on any of the
following docs:

1) GRDDL specification, which is for implementers and power-users, and
as such is formally written.
This document is a Candidate Recommendation, so we'd like comments
within the month:

http://www.w3.org/2004/01/rdxh/spec

2) GRDDL Primer, which is for end-users and is supposed to be
easy-to-read and informal:

http://www.w3.org/2001/sw/grddl-wg/doc29/primer.html

3) The specification has test-cases and a test-harness (we have
implementations in Python, C, and Java), a Last Call Working Draft:

http://www.w3.org/2001/sw/grddl-wg/td/grddl-tests

4) Use-case scenarios:

http://www.w3.org/2001/sw/grddl-wg/doc43/scenario-gallery.htm

thanks!

--
-harry

Harry Halpin, University of Edinburgh
http://www.ibiblio.org/hhalpin 6B522426

_______________________________________________
microformats-discuss mailing list
microforma...@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

[Ottevanger, Jeremy]

Dear all,

Raising my head above what I hope is the correct parapet to ask, does
anyone know if Tim Gambell, who was seemingly leading work on the
proposed work-of-art uf a year ago, is still taking it forward? It looks
as though the last thing that happened with it was April 2006. If it's
croaked then I'll take myself over to the new ufs list and perhaps
propose something over there, but if it's alive, or if anyone here is
interested in it, do let me know.

Thanks,

Jeremy

Jeremy Ottevanger
Web Developer, Museum Systems Team
Museum of London Group
46 Eagle Wharf Road
London. N1 7ED
Tel: 020 7410 2207
Fax: 020 7600 1058
Email: jotte...@museumoflondon.org.uk
www.museumoflondon.org.uk
Museum of London is changing; our lower galleries will be closed while they undergo a major new development. Visit www.museumoflondon.org.uk to find out more.
London's Burning - explore how the Great Fire of London shaped the city we see today www.museumoflondon.org.uk/londonsburning

[Ottevanger, Jeremy]

Thank you Breton. I did see that referred to so I wondered (and found,
as you say, that hCite also seems defunct). The last post in the thread
I saw seemed to be along the lines of "maybe I'd better check out hCite"
and then radio silence. OK, I'll get myself over to the new ufs list.

Thanks again, Jeremy

Jeremy Ottevanger
Web Developer, Museum Systems Team
Museum of London Group
46 Eagle Wharf Road
London. N1 7ED
Tel: 020 7410 2207
Fax: 020 7600 1058
Email: jotte...@museumoflondon.org.uk
www.museumoflondon.org.uk
Museum of London is changing; our lower galleries will be closed while they undergo a major new development. Visit www.museumoflondon.org.uk to find out more.
London's Burning - explore how the Great Fire of London shaped the city we see today www.museumoflondon.org.uk/londonsburning

-----Original Message-----
From: microformats-d...@microformats.org
[mailto:microformats-d...@microformats.org] On Behalf Of
Breton Slivka
Sent: 15 May 2007 13:41
To: Microformats Discuss
Subject: Re: [uf-discuss] Work-of-art/Tim Gambell

I believe it was agreed to use the also stalled hCite instead.

-Breton

[Breton Slivka]

I believe it was agreed to use the also stalled hCite instead.

-Breton

On 15/05/2007, at 7:03 PM, Ottevanger, Jeremy wrote:

[Andy Mabbett]

In message
<CBE3ED7D5509D54F9C4E...@mail-mwh-2k3.museumoflondon.org
.uk>, "Ottevanger, Jeremy" <JOtte...@museumoflondon.org.uk> writes

>proposed work-of-art uf

What would be the use-case?

--
Andy Mabbett
* Say "NO!" to compulsory ID Cards: <http://www.no2id.net/>
* Free Our Data: <http://www.freeourdata.org.uk>
* Are you using Microformats, yet: <http://microformats.org/> ?

[Ottevanger, Jeremy]

Hi Andy, thanks for your interest.

At the risk of going outside the bounds of this list I'll give you an
outline of why I feel such a format might be useful. I should say that
the idea arose from a thinktank exploring the possibilities of the
semantic web in museums, following a session when Jeremy Keith, our
guest of the day, inspired us all with what was for most of us the
pretty unfamiliar world of microformats. The concerns that many of us
had about the practicality of hard-core SW approaches seemed to be at
least partially addressed by the idea of embedding some lightweight
semantics in our web pages. Clearly lots of what we might do with
microformats can be done with existing ones, as indeed we have now done
at my own institution (for events, contact and location information).
However our core activities, revolving around the physical objects in
our collections, cannot be usefully reflected by existing formats.
Museums and galleries have (typically) collections of unique objects. If
they are not unique in the sense that they are not duplicates, then it
is still important to be able to distinguish between one pot, say, and
another identical one in the same or another collection. In this, and in
the fact that their history and provenance (and current ownership) is
deemed valuable, they are unlike, say, products in an online shop. A
number of sophisticated metadata standards exist for museum objects,
ranging from reference models like CIDOC-CRM, through data structure
standards like SPECTRUM and CDWA[Lite] to data content standards like
CCO, but even the lightest of these is not suited to manual authoring
or, really, to embedding in web pages.

The role I envisage for a museum object microformat would be:
- to identify a unique object on the web, tied (ideally) to an
institution
- to enable the capture of a very basic set of metadata about that
object
- optionally, to point at a source of fuller, structured metadata,
thereby making a bridge between the light, author-friendly and flexible
microformatted semantic web and the stricter, machine-facing Semantic
Web

And the use-case? Catalogue pages like this:
http://www.museumoflondon.org.uk/English/Collections/OnlineResources/X20
L/objects/record.htm?type=object&id=742656. Actually the basic idea is
to scratch an itch I've long had, to be able to gather together things
that I find on multiple websites and do my own thing with them. I've
been messing around building a little app that lets me pick up
microformatted museum objects from any page (via a bookmarklet) and
gather them in one place, kind of like Tails crossed with del.icio.us.
That's just what I have long wanted, and I figure that other
museum-lovers may feel the same way.

The way I've been going with this does take me rather outside the normal
practice for microformats but to an extent I'm not concerned as I thing
the most important thing is for it to do what is useful to the sector;
all the same it's important to, firstly, establish if there's much
support in that sector (which I'm trying to do!) and secondly to make
sure that wherever possible I inherit from established formats. I'm very
interested to hear your thoughts as this is the first time I've tried to
explain to people outside the museum community why I think this is worth
pursuing.

All the best, Jeremy

Jeremy Ottevanger
Web Developer, Museum Systems Team
Museum of London Group
46 Eagle Wharf Road
London. N1 7ED
Tel: 020 7410 2207
Fax: 020 7600 1058
Email: jotte...@museumoflondon.org.uk
www.museumoflondon.org.uk
Museum of London is changing; our lower galleries will be closed while they undergo a major new development. Visit www.museumoflondon.org.uk to find out more.
London's Burning - explore how the Great Fire of London shaped the city we see today www.museumoflondon.org.uk/londonsburning

-----Original Message-----
From: microformats-d...@microformats.org
[mailto:microformats-d...@microformats.org] On Behalf Of Andy
Mabbett
Sent: 15 May 2007 18:52
To: Microformats Discuss
Subject: SPAM:Re: [uf-discuss] Work-of-art/Tim Gambell

In message
<CBE3ED7D5509D54F9C4E...@mail-mwh-2k3.museumoflondon.or
g

[Andy Mabbett]

In message
<CBE3ED7D5509D54F9C4E...@mail-mwh-2k3.museumoflondon.org
.uk>, "Ottevanger, Jeremy" <JOtte...@museumoflondon.org.uk> writes

>The role I envisage for a museum object microformat would be: - to
>identify a unique object on the web, tied (ideally) to an institution -
>to enable the capture of a very basic set of metadata about that object
>- optionally, to point at a source of fuller, structured metadata,
>thereby making a bridge between the light, author-friendly and flexible
>microformatted semantic web and the stricter, machine-facing Semantic
>Web
>
>And the use-case? Catalogue pages like this:
>http://www.museumoflondon.org.uk/English/Collections/OnlineResources/X2
>0L/objects/record.htm?type=object&id=742656.

Thank you. I don't wish to dissuade you from continuing, but I'm really
having difficulty seeing how this would be used.

A simple example use-case would be, for hCard:

"a user can add contact details to their address book, or look
up places with coordinates or postal codes on an on-line map".

For the "Currency" proposal:

"a user agent can convert an amount of money, encoded in one
currency, into an alternative currency, by looking up the
exchange rate open a nominated website".

In the case of an art object, surely there will only be one primary
source of information, and that can simply be linked to?

Can you provide examples of webmasters or software, currently delivering
the kind of services which you envisage that a microformat would
facilitate?

BTW, this conversation probably ought to be taking place on the "new
microformats" mailing list; I've cross-posted and set follow-ups, so
please reply there. Thank you.

[Ted Drake]

-----Original Message-----
From: microformats-d...@microformats.org
[mailto:microformats-d...@microformats.org] On Behalf Of

Ottevanger, Jeremy
Sent: Wednesday, May 16, 2007 1:59 AM
To: Microformats Discuss
Subject: Re: [uf-discuss] Work-of-art/Tim Gambell

Hi Andy, thanks for your interest.

At the risk of going outside the bounds of this list I'll give you an

outline of why I feel such a format might be useful....

-------------------------------------------------------------------------

I would agree that museums and archives show great potential for
microformats. As historians, the museum curators and registrars are
exceedingly careful to use best standards in documenting their objects.

I worked in the museum world for five years and the Dublin core was always
held up as a goal for metadata. Unfortunately, the existing collection
management software packages utilized it effectively. Further, the museum
web sites didn't adopt them.

I do believe the museums would adopt a standardized collections microformat
quickly. Further, there are some fantastic conferences held worldwide that
would allow the sages to meet and agree upon a pattern quickly.

However, museum collections don't differ greatly from other archives. The
proposed microformat should work with library and digital archive
collections. Is there already something like this? Could the Dublin core be
converted into a microformat.

Here's a list of previous Dublin core discussions at the Archimuse web site
(hosts of the museums on the web conferences and more)
http://www2.archimuse.com/cgi-bin/htsearch

Ted Drake
www.last-child.com

[Toby A Inkster]

Ted Drake wrote:

> Could the Dublin core be converted into a microformat.

<h1>New Additions to the Museum Collection</h1>
<ul>
<li class="objet">
<h2 class="dc:title">Comb</h2>
<ul>
<li class="dc:date:created">circa 3200 BCE; Predynastic period</li>
<li class="geo">Egypt (<span class="latitude" title="30">N
30&deg;</span>, <span class="latitude" title="31.2">E
31&deg;12'</span>)
</li>
<li>
<span class="dc:type scheme:dcmitype" title="PhysicalObject"></span>
<span class="dc:format">Ivory; 5.7 cm</span>
</li>
<li class="dc:publisher vcard">
<span class="fn">The Metropolitan Museum of Art</span>
<span class="adr">
<span class="locality">New York</span>,
<span class="country-name">USA</span>
</span>
</li>
</ul>
<p class="dc:description">
Finely carved ivory combs and knife handles produced toward the end of
Egypt's prehistory demonstrate the high standards Egyptian artists had
achieved, even before the Old Kingdom. This comb may have been part of
the funeral equipment of an elite person who lived about 5,200 years
ago. Parts of the comb's teeth, now missing, can be seen along the
bottom edge. The detailed decoration suggests that it was a ceremonial
object, not just an instrument for arranging the hair. On both sides are
figures of animals in horizontal rows, a spatial organization familiar
from later Egyptian art. The animals include elephants and snakes;
wading birds and a giraffe; hyenas; cattle; and perhaps boars. Similar
arrangements of these creatures on other carved ivory implements suggest
that the arrangement and choice of animals were not haphazard. Elephants
treading on snakes suggest that this part of the scene was symbolic. The
mythologies of many African peoples associate elephants and serpents
with the creation of the universe. The uppermost row of this comb may
symbolize a creative deity to whom the rest of the animals owe their
existence.
</p>
</li>
<li class="objet"><!-- another objet --></li>
</ul>

Notes:

1. An object's class is intentionally "objet", from "objet d'art" to avoid
collisions with existing use of class="object".
2. I'm not 100% sure about the best use of "geo" here. Should it be used
to markup where the object was created (as I have done here); where the
object currently is; or both?
3. Note the empty span to represent the dc:type metadata. Virtually all
items in museums are physical objects, so I didn't think it necessary to
show this information to human users.
4. I'm not sure about scheme:dcmitype. Are schemes necessary, or is that
going too far?
5. Could somehow integrate with XOXO?

--
Toby A Inkster BSc (Hons) ARCS
[Geek of HTML/SQL/Perl/PHP/Python/Apache/Linux]
[OS: Linux 2.6.12-12mdksmp, up 104 days, 16:20.]

URLs in demiblog
http://tobyinkster.co.uk/blog/2007/05/31/demiblog-urls/

[Andy Mabbett]

In message <sljnj4-...@ophelia.g5n.co.uk>, Toby A Inkster
<ma...@tobyinkster.co.uk> writes

>Ted Drake wrote:
>
>> Could the Dublin core be converted into a microformat.
>
><h1>New Additions to the Museum Collection</h1>
><ul>
> <li class="objet">
> <h2 class="dc:title">Comb</h2>
> <ul>
> <li class="dc:date:created">circa 3200 BCE; Predynastic period</li>
> <li class="geo">Egypt (<span class="latitude" title="30">N
> 30&deg;</span>, <span class="latitude" title="31.2">E
> 31&deg;12'</span>)
> </li>

[snip rest of markup]

Haven't you just reinvented RDFa?

>Notes:
>
>1. An object's class is intentionally "objet", from "objet d'art" to
>avoid collisions with existing use of class="object".

I think that would generate confusion.

>2. I'm not 100% sure about the best use of "geo" here. Should it be
>used to markup where the object was created (as I have done here);
>where the object currently is; or both?

Either, or both, but with an additional class of "made" and/or
"located", or some such; possibly also "found" (imagine a coin made in
Rome, found in York, and on display in London). They and the place names
should be included in hCards.

--
Andy Mabbett

[Toby A Inkster]

Andy Mabbett wrote:

> Haven't you just reinvented RDFa?

There are a few important differences:

* RDFa adds new attributes to XHTML -- my proposal doesn't.

* RDFa uses namespaces to potentially "import" many different
metadata schemes. Mine doesn't. The "dc:" prefix on certain
class names should not be seen as a true namespace -- just a
name prefix to avoid collisions.

* Mine leverages existing microformats, such as hcard and geo.
I'd also intended to include an example which leveraged
hcalendar to describe the history of a particular object,
but didn't have time to write it.

--
Toby A Inkster BSc (Hons) ARCS
[Geek of HTML/SQL/Perl/PHP/Python/Apache/Linux]

[OS: Linux 2.6.12-12mdksmp, up 104 days, 23:55.]

URLs in demiblog
http://tobyinkster.co.uk/blog/2007/05/31/demiblog-urls/

_______________________________________________

[Tantek Çelik]

On 6/8/07 2:21 AM, "Toby A Inkster" <ma...@tobyinkster.co.uk> wrote:

> Ted Drake wrote:
>
>> Could the Dublin core be converted into a microformat.
>

In short no, however, it could be converted to POSH.

Dublin Core is one of many citation-like previous formats, and thus best
serves as source research for the citation microformat[1].

Tantek

[1] http://microformats.org/wiki/citation