On Tue, Aug 23, 2011 at 8:14 AM, Kim Bruning kim@bruning.xs4all.nl wrote:
On Tue, Aug 23, 2011 at 05:21:23PM +0200, Milos Rancic wrote:
Board was aware of that, as the first Robert Harris' report included very similar text from Canadian librarian association.
I would then like to point out that there is no practical way to make a value-neutral categorisation scheme to use for filtering.
This seems like an over-hasty statement. There are many possible categorization schemes that are neutral; the ALA in fact makes that distinction itself, since libraries (obviously) use all kinds of labeling and categorization schemes all the time. The ALA and other library organizations have taken a stand against censorious and non-neutral labeling, not all labeling. If you keep reading the ALA page you linked, it says that the kind of labels that are not appropriate are when "the prejudicial label is used to warn, discourage or prohibit users or certain groups of users from accessing the material" -- e.g. a label that reads "not appropriate for children". That does not mean that picture books for kids, or mystery novels, or large-print books, aren't labeled as such in every public library in the country -- and that is the difference between informative and prejudicial labeling.
The ALA also makes a point of stating that materials should be on open shelves and accessible to everyone regardless of labeling -- this comes out of, among other things, the once-common practice of not allowing children in the adult section of the library. The natural equivalent for us I think is to make sure that all materials we host are accessible to everyone regardless of any label, which is certainly a principle we have and continue to uphold.
The Board didn't specify any particular mechanism or system in our resolution. What we did was to ask for a particular kind of feature and spell out some principles for its development. We talked about neutral language in the interface, and our intent was exactly that distinction I noted between informative and prejudicial -- we do not wish to set up a system that privileges certain value judgments about content. We wish *readers to have a choice* when they use our projects -- one they do not have now unless they are remarkably technically inclined and forward-looking.
We didn't address the categorization system in particular because frankly, it's not our business. It's the community's, and tech's. And the Trustees didn't all agree on whether we thought categorization as proposed in the first draft of the system was the best idea, anyway; some of us thought it was appropriately in line with the principle of least astonishment, and some of us thought it could lead to problems. But we did come to consensus on the high-level idea as expressed in the resolution, and we agreed and understood that the ideas around how to implement it would have to iterate, with reevaluation along the way. But after all, developing informative, neutral and useful systems for organizing information is something that the Wikimedia projects have become world-famous for -- so if anyone can do it I have faith that we can :)
As I told DGG, there's a lot of caveats in that resolution. And those caveats are there for a reason. It should not be extrapolated that the Board as a whole *actually* supports a particular, or different, or more censorious, filtering scheme. What we want is for people to easily be able to hide images for themselves if they don't want to see them when using our projects. (And we also want other things, like better tools for Commons, that are expressed in other parts of that resolution.)
I know we are all looking forward to seeing the referendum results, and the data from it will need to be carefully considered. In the meantime I am glad to see more discussion of this, but I am remembering that it is a stressful topic!
best, -- phoebe