[rdfweb-dev] Relationship Schema Updated

Michael Bauser michael at bauser.com
Sun Mar 14 09:09:27 UTC 2004


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ian Davis wrote:

> Well it doesn't claim to be comprehensive or definitive.

The problem is that its vague when it should be definitive (kinship),
and definitive where it should be vague (friendship).

> It's just useful and fun.

- From where I stand, it's not that useful, because it encourages the
creation of less-than-desciptive datasets.

> We're taking tiny steps here so we get _something_ that
> works even if it's wrong. Then we learn from that and make it better.
> Please propose some additions or alternatives that will improve it.

OK.

1) Kill those hideously ethnocentric attempts at kinship terms and
replace them with terms that are cross-cultural and actually
descriptive: brother, daughter, father, husband, mother, son, sister,
and wife.) They've worked for scientists describing real-world social
networks for decades, I don't see why they wouldn't work on the
Internet.) See my response to Morten's message for the gory details.

2) Kill FriendOf, CloseFriendOf, EnemyOf, AntagonistOf, and
AmibivilantOf. They're just going to create unhelpful ambiguities and
social awkwardness. (It's like an out-of-balance Likert scale, or
something.) If you really want "tiny steps", try
"postive/negative/neutral opinion" (with no assertions about
reciprocity). That's where a social scientist would start, because
informants alway understand binary questions.

3) Speaking of "tiny steps", how is "mentorOf/apprenticeOf" better than
"teacherOf/studentOf"? More people understand teacher/student than
mentor/apprentice. (It'll probably translate better, as well, when it
comes time to translate the documentation into other languages.) The
relationship vocabulary may be lessening its own usuability here by
being just a bit too clever. Remember, when and if somebody creates
applications that help people create FOAF files, these relationship
terms stuff will proably be an a drop-down menus: In a drop-down,
simpler is usually better.

4) Lose the entire "Using With HTML and XHTML" nonsense. You're taking
link relationships in exactly the opposite direction the newer HTML
recommendations do; link relationships define the relationship between
documents, not people. You're imitating one of the worst aspects of XFN.

5) knowsInPassing and knowsByReputation: The first in nearly
nonfunctional, and the second is completely non-functional. Where does
knowsByReputation begin and end? Should I just go ahead and list all the
United States Presidents that can I remember? They have reputations.

6) wouldLikeToKnow is also nearly-useless, and just a little creepy.
What's next? rel:stalkerTo? rel:patheticCrushOn?
rel:waitingForHerToDumpHerBoyfriendSoICanBagHerOnTheRebound? I'm just
getting too many bad vibes from wouldLikeToKnow; its primary users will
be tactless high school students and creepy guys who don't bathe enough.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (MingW32)

iD8DBQFAVCFGcpuEJT2bpHsRAi8XAJ0XMnVBbFwpAmKLcuEXejw/oeqEZQCg3c+4
BM5Amz5hOdd00h6WOPPq95I=
=p1vS
-----END PGP SIGNATURE-----



More information about the foaf-dev mailing list