headers
Luke Maurits | 3 Jun 2009 08:06
Picon
Gravatar

[foaf-dev] Introduction

Greetings list,

I've recently taken an interest in getting involved in the FOAF
project, so this email is just me saying "hi" to everyone and trying to
get a feel for the level of activity in the community and this list.

I've started working on a Python library (see
http://code.google.com/p/foaflib/) to make developing FOAF-based
applications easier (it's really just a wrapper around the rdflib
library).  Assitance from any fellow Pythonistas would be very
welcome!  Once this library is in good shape I'm interested in the
possibility of developing a desktop application for (i) managing one's
one FOAF profile, (ii) keeping up to date on friends' FOAF profiles and
other things pointed to by their FOAF profiles, like blogs, Twitter
feeds, etc. - a sort of social RSS aggregator, and (iii) (and this is
the interesting part) doing peer to peer searches of FOAF-space by
communicating with other people running the application.  This last
idea is intended to solve the problem of searching for FOAFers without
someone having to put in the time/effort/money to set up a reliable
server with a huge database and constantly busy scutters.  I'm quite
interested in discussing this idea more with people, and also in
hearing if anybody has tried anything like it before.

Could someone give me a frank assessment of the vitality of the FOAF
community?  I notice that the vocab spec has remained unchanged for
something like 18 months, despite containing some obvious but
superficial problems - like the redundancy and inconistency of
givenname vs firstName and surname vs family_name.  I see that this
list only had 8 posts for all of last month, and I've lurked in #foaf
on freenode a few times without seeing anybody say a word (and the logs
(Continue reading)

Permalink | Reply | 3 comments |
headers
Dan Brickley | 3 Jun 2009 12:02

Re: [foaf-dev] Introduction

On 3/6/09 08:06, Luke Maurits wrote:
> Have I just started paying attention at a bad time (perhaps this apparent inactivity is correlated
> with the recent server downtime?)

You're absolutely right. Things are too quiet lately!

Basically FOAF was most active during the period when Libby Miller and 
myself were fortunate enough to have dayjobs that were all about 
building tools and community around practical RDF technologies. In 2001 
we wrote a proposal for an EU project, SWAD-Europe which got accepted 
and gave us a few years where we could both spend a lot of time hanging 
around with the early-years RDF developer community, building 
collaborations with folk from rdfweb-dev, foaf-dev, demos etc. Also the 
SKOS work now being finalised at W3C was a product of that same project.

Since 2005 we've both been away from the SemWeb scene a bit, and FOAF 
developments slowed down. I rev'd the spec mildly to accomodate OpenID 
in 2007, but we didn't have the 100% semwebbing lifestyle of previous 
years!

Things are changing a bit, recently.

Firstly, at W3C there is now a Social Web incubator group, 
http://www.w3.org/2005/Incubator/socialweb/ which meets weekly and which 
I'm co-chairing with Daniel Appelquist and Harry Halpin. This gives a 
broader forum through which FOAF developments can have a public 
heartbeat, and through which wider issues (privacy, I18N etc) can be 
raised and managed. There have also been many developments in the Web 2 
scene, such as the excellent Portable Contacts work. I propose that in 
any future revs to the FOAF spec we defer as much as possible to the
(Continue reading)

Permalink | Reply | 0 comments |
headers
Dan Brickley | 3 Jun 2009 12:42

Re: [foaf-dev] Introduction

On 3/6/09 08:06, Luke Maurits wrote:
> Greetings list,
>
> I've recently taken an interest in getting involved in the FOAF
> project, so this email is just me saying "hi" to everyone and trying to
> get a feel for the level of activity in the community and this list.
>
> I've started working on a Python library (see
> http://code.google.com/p/foaflib/) to make developing FOAF-based
> applications easier (it's really just a wrapper around the rdflib
> library).  Assitance from any fellow Pythonistas would be very
> welcome!

Just to follow up on this part too for those not in IRC - chat logs are 
here: http://chatlogs.planetrdf.com/foaf/2009-06-03#T06-15-34

One theme for discussion here: what would a "smarter" FOAF crawler look 
like? How can we spec things such that you can pull in a useful but not 
overwhelming piece of the nearby machine-readable Web? There's too much 
data out there to harvest everything, but pulling it all 
while-user-waits doesn't work either. How deep to crawl, what kinds of 
links to traverse?

cheers,

Dan
Permalink | Reply | 0 comments |
headers
Marco Neumann | 4 Jun 2009 16:59
Picon

[foaf-dev] foaf:Person vs crm:E21.Person

Hi foaf team,

I am modeling with an ontology for the cultural and natural heritage domain. We use a class called crm:E21.Person to achieve something similar to what foaf:Person gives us.

I am contemplating to make one class a subclass of the other. The crm community prefers to make foaf:Person a subclass of crm:E21.Person. Or should we consider equivalence between the two class?

http://www8.informatik.uni-erlangen.de/IMMD8/Services/cidoc_crm/docu/081216/classes/E21.Person.html

Best,
Marco


_______________________________________________
foaf-dev mailing list
foaf-dev@...
http://lists.foaf-project.org/mailman/listinfo/foaf-dev
Permalink | Reply | 6 comments |
headers
Dan Brickley | 4 Jun 2009 17:59

Re: [foaf-dev] foaf:Person vs crm:E21.Person

On 4/6/09 16:59, Marco Neumann wrote:
> Hi foaf team,
>
> I am modeling with an ontology for the cultural and natural heritage
> domain. We use a class called crm:E21.Person to achieve something
> similar to what foaf:Person gives us.
>
> I am contemplating to make one class a subclass of the other. The crm
> community prefers to make foaf:Person a subclass of crm:E21.Person. Or
> should we consider equivalence between the two class?
>
> http://www8.informatik.uni-erlangen.de/IMMD8/Services/cidoc_crm/docu/081216/classesE/21.Person.html

Hey Marco,

In an idea world, you shouldn't need to ask. Just read the human and 
formal documentation for the two classes, and try to find corner cases 
that fit in one class but not the other. That should give you your answer.

Here's CRM's E21:

"This class comprises real persons who live or are assumed to have 
lived. Legendary figures that may have existed, such as Ulysses and King 
Arthur, fall into this class if the documentation refers to them as 
historical figures. In cases where doubt exists as to whether several 
persons are in fact identical, multiple instances can be created and 
linked to indicate their relationship. The CRM does not propose a 
specific form to support reasoning about possible identity. Examples: 
Tut-Ankh-Amun, Nelson Mandela."

And here is FOAF's Person:

via http://xmlns.com/foaf/spec/#term_Person
"The foaf:Person class represents people. Something is a foaf:Person if 
it is a person. We don't nitpic about whether they're alive, dead, real, 
or imaginary. The foaf:Person class is a sub-class of the foaf:Agent 
class, since all people are considered 'agents' in FOAF."

Pretty much the same approach. FOAF sounds somewhat more liberal though, 
as it admits fictional people. So you could use it for example when 
describing characters in a drama - this was done for example in the SUDS 
experiments by BBC people, modelling Eastenders. See also 
http://www.r4isstatic.com/?p=18 for more discussion.

Since crm:e21.Person is smaller than foaf:Person, since it doesn't 
include fictional and imaginary people, it would be natural to express 
that in terms of e21 being the sub-class. Of course in your application 
you might chose to insulate yourself from all this and define your own 
classes directly by reference to CRM and FOAF. That's of course fine too!

cheers,

Dan
Permalink | Reply | 5 comments |
headers
Marco Neumann | 4 Jun 2009 20:12
Picon

Re: [foaf-dev] foaf:Person vs crm:E21.Person

Thanks Dan for the elaboration.

In the spirit of the FOAF vocabulary doesn't it go to far to consider an imaginary person?

If a foaf:Person can be a fictional person not just a historical figure but a truly imaginary one, I can never be a friend of that person expect in case of a mental extension to reality not to say a borderline schizophrenic situation.

That said I typically prefer foaf:Person over crm:E21 for web resources since it's more commonly used. But we might need a companion or extension for a fictional character.

I think we don't capture imaginary characters as typed entities in the crm at this point. I need to confirm this though with the SIG.

Best,
Marco

 




On Thu, Jun 4, 2009 at 11:59 AM, Dan Brickley <danbri-EeGx4OaoQ4Adnm+yROfE0A@public.gmane.org> wrote:
On 4/6/09 16:59, Marco Neumann wrote:
Hi foaf team,

I am modeling with an ontology for the cultural and natural heritage
domain. We use a class called crm:E21.Person to achieve something
similar to what foaf:Person gives us.

I am contemplating to make one class a subclass of the other. The crm
community prefers to make foaf:Person a subclass of crm:E21.Person. Or
should we consider equivalence between the two class?

http://www8.informatik.uni-erlangen.de/IMMD8/Services/cidoc_crm/docu/081216/classesE/21.Person.html

Hey Marco,

In an idea world, you shouldn't need to ask. Just read the human and formal documentation for the two classes, and try to find corner cases that fit in one class but not the other. That should give you your answer.

Here's CRM's E21:

"This class comprises real persons who live or are assumed to have lived. Legendary figures that may have existed, such as Ulysses and King Arthur, fall into this class if the documentation refers to them as historical figures. In cases where doubt exists as to whether several persons are in fact identical, multiple instances can be created and linked to indicate their relationship. The CRM does not propose a specific form to support reasoning about possible identity. Examples: Tut-Ankh-Amun, Nelson Mandela."

And here is FOAF's Person:

via http://xmlns.com/foaf/spec/#term_Person
"The foaf:Person class represents people. Something is a foaf:Person if it is a person. We don't nitpic about whether they're alive, dead, real, or imaginary. The foaf:Person class is a sub-class of the foaf:Agent class, since all people are considered 'agents' in FOAF."

Pretty much the same approach. FOAF sounds somewhat more liberal though, as it admits fictional people. So you could use it for example when describing characters in a drama - this was done for example in the SUDS experiments by BBC people, modelling Eastenders. See also http://www.r4isstatic.com/?p=18 for more discussion.


Since crm:e21.Person is smaller than foaf:Person, since it doesn't include fictional and imaginary people, it would be natural to express that in terms of e21 being the sub-class. Of course in your application you might chose to insulate yourself from all this and define your own classes directly by reference to CRM and FOAF. That's of course fine too!

cheers,

Dan

_______________________________________________
foaf-dev mailing list
foaf-dev@...
http://lists.foaf-project.org/mailman/listinfo/foaf-dev
Permalink | Reply | 2 comments |
headers
Dan Brickley | 4 Jun 2009 21:03

Re: [foaf-dev] foaf:Person vs crm:E21.Person

On 4/6/09 20:12, Marco Neumann wrote:
> Thanks Dan for the elaboration.
>
> In the spirit of the FOAF vocabulary doesn't it go to far to consider an
> imaginary person?

On the Web, not every page is true. Just as HTML doesn't require every 
page to tell the truth, neither should FOAF or RDF. Another reason is 
that sometimes, the description doesn't contain enough information for 
anyone to know whether the Person was real or not.

"This is a picture of Marco stood next to a painting of someone called 
John", "This is a photo of a sculpture of a young man", "This is a story 
about George Bush Jr's great-great-grandson", "This is a blog comment 
made by someone who said their name was Anonymous Al."...  All of those 
statements need some way of mentioning the idea of a person. I don't see 
any great need for FOAF to say some of them are out-of-scope. So they're 
all in scope, that's all we have for now...

> If a foaf:Person can be a fictional person not just a historical figure
> but a truly imaginary one, I can never be a friend of that person expect
> in case of a mental extension to reality not to say a borderline
> schizophrenic situation.

Agreed. Being able to talk about non-existent, fictitious, idealised or 
dead people doesn't bring them to life!

> That said I typically prefer foaf:Person over crm:E21 for web resources
> since it's more commonly used. But we might need a companion or
> extension for a fictional character.

I agree that more conventions are needed around fiction. I hope the nice 
folk at the BBC will supply some concrete use cases here, eg. around 
fan-maintained life histories in a Semantic Wiki. When a soap runs for 
30+ years, keeping track of that alternate world is quite hard; and is a 
real issue when new writers need to catch up on things the fan-base know 
intimately (including past mental states of the fictional people). FOAF 
doesn't solve all these problems out of the box, for sure!

Re crm:E21 and cultural heritage apps., there are also some developments 
in the library world around the FRBR and RDA initiatives. I'd like to 
allign FOAF's classes with the needs of that community too. From a quick 
look, it might involve adding a Family class and explaining how it 
relates to foaf:Group, and how Organization relates to corporate entities...

> I think we don't capture imaginary characters as typed entities in the
> crm at this point. I need to confirm this though with the SIG.

It's probably a huge can of worms if you want to model it "properly". 
The FOAF hack is just to say "if you use the Person class in this way, 
nothing's going to explode"...

What're you building, may I ask? :)

Dan
Permalink | Reply | 1 comment |
headers
Simon Reinhardt | 4 Jun 2009 21:57
Picon

[foaf-dev] Fictional stuff (was: Re: foaf:Person vs crm:E21.Person)

Dan Brickley wrote:
> Pretty much the same approach. FOAF sounds somewhat more liberal though, 
> as it admits fictional people. So you could use it for example when 
> describing characters in a drama - this was done for example in the SUDS 
> experiments by BBC people, modelling Eastenders. See also 
> http://www.r4isstatic.com/?p=18 for more discussion.

Speaking of which: does anyone know if there are any vocabularies that model fiction? Fictional
universes, places, times, characters? Structure of fictional narration?
I think it's worth having those as classes of their own. Especially characters deserve a special treatment
I think. According to Wikipedia [1]: "A character is any person, persona, identity, or entity that exists
in a work of art. The process of conveying information about characters in fiction is called
characterisation. Characters may be entirely fictional or based on real, historical entities.
Characters may be human, supernatural, mythical, divine, animal, or personifications of an abstraction."
So I would say the character John Malkovich in "Being John Malkovich" is a separate entity from the
real-life person John Malkovich. This enables you to describe the special characteristics given to him
in the film and to describe his actions in it without implying that the real person did that.
I've done some initial work on that sort of thing. I was just wondering if there is any existing stuff before
pursuing it.

Thanks,
  Simon

[1] http://en.wikipedia.org/wiki/Character_(arts)
Permalink | Reply | 1 comment |
headers
Danny Ayers | 4 Jun 2009 22:39
Picon
Gravatar

Re: [foaf-dev] Fictional stuff (was: Re: foaf:Person vs crm:E21.Person)

2009/6/4 Simon Reinhardt <simon.reinhardt@...>:

> Speaking of which: does anyone know if there are any vocabularies that model fiction? Fictional
universes, places, times, characters? Structure of fictional narration?
> I think it's worth having those as classes of their own. Especially characters deserve a special
treatment I think.

Interesting point.

> So I would say the character John Malkovich in "Being John Malkovich" is a separate entity from the
real-life person John Malkovich. This enables you to describe the special characteristics given to him
in the film and to describe his actions in it without implying that the real person did that.

Might that not be possible by the use of a single class?

_:him
       a foaf:Person;
       a x:FictionalEntity;
       foaf:name "John Malkovich";
...

(loved that film btw, even though John Malkovich usually irritates the
heck out of me :)

Cheers,
Danny.

--

-- 
http://danny.ayers.name
Permalink | Reply | 0 comments |
headers
Michael Smethurst | 4 Jun 2009 23:28
Picon
Picon
Favicon

Re: [foaf-dev] Fictional stuff (was: Re: foaf:Person vs crm:E21.Person)

Whoops, meant to send that to list too. Sorry - sleepy :-)


-----Original Message-----
From: Michael Smethurst


> Speaking of which: does anyone know if there are any vocabularies that model fiction? Fictional universes, places, times, characters? Structure of fictional narration?
> I think it's worth having those as classes of their own. Especially characters deserve a special treatment I think. According to Wikipedia [1]: "A character is any person, persona, identity, or entity that exists in a work of art. The process of conveying information about characters in fiction is called characterisation. Characters may be entirely fictional or based on real, historical entities. Characters may be human, supernatural, mythical, divine, animal, or personifications of an abstraction."
> So I would say the character John Malkovich in "Being John Malkovich" is a separate entity from the real-life person John Malkovich. This enables you to describe the special characteristics given to him in the film and to describe his actions in it without implying that the real person did that. I've done some initial work on that sort of thing. I was just wondering if there is any existing stuff before pursuing it.

Hi Simon

We (the BBC) are just kicking of a project to make 'character pages' and are coming across all the issues you describe here. Obviously being the bbc we tend to get stuck on doctor who (no bbc meeting ever happens without a mention of doctor who [2] ;-) ). Is each doctor a different character, different portrayal, different persona etc...?

Similar problems with different dramatisations or sketches based on novels. Is character x from the tv adaption the same as character x from the film or character x from the radio drama or character x from the comedy sketch or character x from the opera performance? are they different portrayals of the same character? can 2 different actors make the same 'portrayal'? as ever you rapidly run of of labels to capture the concepts...

our favourite edge case for the moment is Miley Cyrus who portrays Miley Stewart who portrays Hannah Montana (also played by Miley Cyrus) [3]. luckily it's not broadcast by the bbc so we're ducking such pomo horrors for now ;-)

Another problem we've had in the past has been the typing of 'people' as real or fictional. it starts out fairly easy and rapidly heads into tricky territory when u hit religion. however u handle it, dealing with "human, supernatural, mythical, divine [...] or personifications of an abstraction" ends up upsetting someone...

Outside drama you get the same kind of issues in music with multiple aliases often taking the form of characters (aladin sane etc). But you also get aliases that are only borderline characters (ringo star is an alias but is he a character?!? freddy mercury? in some ways they're both characters but then so are some artists who use their given and family names....)

Haven't come across much outside SUDs and Paul Rissen's work that tries to capture fictional universes, timelines etc. If you do go ahead with this work it'd be really good if you kept us informed of progress. We can supply endless use cases and getting external input would help us get our modelling sorted. Sorting definitions for character, portrayal, persona etc feels like the first step

Hope to hear from you
Michael



> Thanks,
> Simon


> [1] http://en.wikipedia.org/wiki/Character_(arts)
[2] http://www.currybet.net/cbet_blog/2009/03/currybets_law.php
[3] http://en.wikipedia.org/wiki/Hannah_Montana



http://www.bbc.co.uk
This e-mail (and any attachments) is confidential and may contain personal views which are not the views of the BBC unless specifically stated.
If you have received it in error, please delete it from your system.
Do not use, copy or disclose the information in any way nor act in reliance on it and notify the sender immediately.
Please note that the BBC monitors e-mails sent or received.
Further communication will signify your consent to this.
_______________________________________________
foaf-dev mailing list
foaf-dev@...
http://lists.foaf-project.org/mailman/listinfo/foaf-dev
Permalink | Reply | 0 comments |

Gmane