Freebase: Discussion about Automobile Model Year

None: Absence of information vs. information of absence

2008-11-14T16:31:17.0012Z

The bug has been created and resides here:

https://bugs.freebase.com/browse/FREEBASE-330

None: Absence of information vs. information of absence

2008-11-14T15:26:19.0012Z

I read his response but couldn't cut through the jargon, to be quite honest. It seemed as if he agreed that typing "none" and keeping it as a topic was an acceptable solution, but I just can't see how that can be. It's a novelty on Freebase, to be sure, but it certainly goes against the legitimacy of a data set when one of the items a) doesn't exist and b) doesn't exist with the types of relationships that it claims to. I'll open a ticket in the bug tracking system to this effect, but if what you say is true, I don't know that it'll do any good.

None: Absence of information vs. information of absence

2008-11-14T06:23:51.0012Z

Scott Meyer seemed to indicate on the data-modeling list that the graph-level folks were not interested in implementing this. I'll vote on this ticket if someone from Metaweb opens it but I suspect that the issue is pretty much decided now.

None: Absence of information vs. information of absence

2008-11-13T19:34:44.0012Z

I'd suggest raising a ticket in our bug tracking system, describing in detail the problem being addressed and your proposed solution. This is probably a graph-level change, so we can then pass it over to that team for evaluation.

None: Absence of information vs. information of absence

2008-11-13T14:21:41.0012Z

Yes! This is a great idea. How do we get this to happen?