encoding biographies

[eroux]

I'm not sure the best way to indicate that a work is a biography of a person. Currently we have workIsAbout which has no domain nor range but only topics as examples. Should we use that? Or maybe make a subproperty workBibliographyOf? Maybe it's not useful if we have

bdr:W123 :workHasGenre bdr:T123 ;
                 :workIsAbout bdr:P123 .

with bdr:T123 being "namthar". Although one could argue that in this case, the fact that bdr:W123 is a biography of bdr:P123 could be fully express by the ontology without relying on user-defined genres, as it's an important data... but we can leave that for later as it's unlikely to change the migration.

[xristy]

I think we're almost there, I think.

Checking bdr:W20443 I see:

bdo:workIsAbout bdr:P4252 ;
bdo:workIsAbout bdr:T3 ;

which came from:

<w:subject type="isAboutPerson"  class="P4252">dbang phyug brtson 'grus</w:subject>
<w:subject type="isInstanceOfGenre"  class="T3">rnam thar</w:subject>

so if the @type="isInstanceOfGenre" is mapped to bdo:workGenre then:

bdo:workIsAbout bdr:P4252 ;
bdo:workGenre bdr:T3 ;

and all is good?

[eroux]

oh thanks, I just fixed that.

In the piece of code where the bug was, I realize there was a grey area: how should I encode @type="isCommentaryOn"?

I suppose using :workIsAbout is necessary, but maybe I should add a :workGenre linking to the appropriate topic? If so, which one would that be?

In a perfect world, an xml file with @type="isCommentaryOn" has a @type="isInstanceOfGenre" with the appropriate commentary type, but my suspicion is that it's not always the case...

[xristy]

In looking at the data I see there are two cases: 1) W1KG12097 where the genre is a commentarial type, in this case T132 ('grel pa), there are also T61 (rnam bshad) and so on; and 2) W20400 where the genre is more like a tradition or school, in this case T2055 (sems tsam/cittamatra).

The first case is simple:

bdo:workGenre bdr:T132 ;
bdo:workIsAbout bdr:W20804 ;

In the second case adding:

bdr: W20400 bdo:workGenre bdr:T132

to stand-in for the (now implicit) "isCommentaryOn" would perhaps be appropriate.

There's another sort of type 1 situation: W20401 which has:

<w:subject class="T233"  type="isAboutUncontrolled">dbu ma</w:subject>
<w:subject class="T132"  type="isInstanceOfGenre">'grel pa </w:subject>
<w:subject class="W19458"  type="isCommentaryOn">madhyamakavatara </w:subject>

which I think just maps to:

bdo:workGenre bdr:T132 ;
bdo:workIsAbout bdr:W19458 ;
bdo:workIsAbout bdr:T233 ;

There may be other cases to consider out of the 970 occurrences of isCommentaryOn that I haven't found yet.

The commentary topics that I am aware of are:

T304         mchan 'grel

T3JT5054     don 'grel

T61          rnam bshad

T4JW5424     rang 'grel

T10MS12837   dgongs 'grel

T132         'grel pa

T1488        'bru 'grel

T1491        dka' 'grel

T2397        tshig 'grel