Topic on User talk:Fralambert/Structured Discussions Archive 1

Hi! Yesterday you changed Nigula Nature Reserve (Q13653426) to be an instance of its designation (nature reserve (Q45754521)). As I described earlier in Talk:Q61467056, I used generic protected area class protected area (Q473972) as P31 value so that P31 and P1435 values weren't duplicated and one statement didn't seem redundant to one another too easily. I believe this should be fine as its common that generic class is used as P31 value while values to some other properties might be considered as subclasses of this generic class (e.g. P31=Q5 vs. occupation, gender etc.) So similarly, as its useful to provide particular designation as separate property then I've used generic protected area class as P31 value consistently for protected areas in Estonia.

You also added property stating that IUCN category is unknown. As I described earlier in Property talk:P809, IUCN category is not unknown, but instead no protected area in Estonia is assigned into IUCN categories. (Overview of use of IUCN categories in Estonia can be found here.) There are many protected objects of certain designations that are without IUCN category. So as no value/unknown value statements suggest that there should be a value or that there usually is value for these types of objects then confusion arises from it. A few months ago I tried to remove associated property constraint as discussed. You did put it back and added the status of "suggestion constraint". I think here "novalue" is still far too common for suggestion status as well and it would be still better to remove this constraint as apparently it's for the most part a source of confusion without serving much of useful purpose.

Personnellement, je comprend pas pourquoi l'Estonie devrait avoir un régime différent des autres nations. Pour le Canada, le statut légal est dans P31, pour la France, les États-Unis, etc. aussi. De plus, je vois que vous n'êtes pas ouvert aux compromis.

I'm not well familiar with types of protected natural objects in Canada or France. If there are no individual protected objects (non-areas) there, then my considerations would have less relevance for protected objects in these countries, of course. Otherwise, do you disagree that at least for individual objects (like Ehalkivi (Q1299227)) it's better to provide their protections status using different property than P31? (So that my other considerations on consistency and confusing duplication also wouldn't apply.) If so, then why? Currently I don't quite understand between what considerations is P31 and P1435 value duplication a compromise.

Regarding protected objects that are protected areas, I notice that there doesn't seem to be consistent approach for objects outside Estonia either to provide all area designations as P31 values. For istance, Ramsar site or UNESCO world heritage site designation seem to be generally given as P1435 values (and not P31).

As for P814: it obviously differs between countries which types of objects are assigned into IUCN categories, e.g. (regional) nature reserves in France seems to have IUCN category, nature reserves in Estonia don't. To my understanding constraints are supposed to be something that apply (almost) universally. Not something that applys in some country (or for some types) and produces many confusing false-positives for the rest. Then again, I don't really know your consideration for adding this constraint (along many with exceptions as no value/unknown value) either. Last you told in property talk that we should probably delete the constraint.

Do you still mind if I removed IUCN category constraint from WDPA ID (P809)? As this came up again, and probably will again, then could we please seek for resolution on this. In property talk you said that we probably should delete the constraint. If you have changed your mind then why?

Considering that there whole classes of WDPA objects that are without IUCN category, then I find this constraint very confusing, as suggestion constraint (Q62026391) too. Also, description for this constraint status says that "violations are not as severe", not that the suggestion couldn't be followed on all items.

Regarding Special:Diff/1058355599. First of all, I made the change as I had thoroughly explained it in advance and since there was no reply from you then I assumed that you are alright with it now. We should be able to discuss things, in property talk page or in some other venue, rather than just ignore comments and issues. I believe this is the normal way to handle issues on wiki. Dissucsion via edit comments does not work well and is discourged to my understanding.

I probably don't fully understand your edit comment in French due to using Google Translate. You seem to tell that constraint is useful and it's a compromise (or suggestion status is compromise).

I still don't quite understand what you mean by compromise. Adding suggestion status changes barely anything. As explained, to my understaning it only says that "violations are not as severe". It still suggests that it's OK to add the statement and it should be added. So the suggestion still is clearly misleading. Or, are there some more thorough guidelines on using suggestion status which suggets that status should be understood differently?

For protected areas in Estonia I've mostly encountered bad/erroneous IUCN statements (i.e. mostly where there should be no IUCN statement; also sometimes people apparently try to quess the value, e.g. category II for national parks that is false for national parks in Estonia). Most likely these statements have been added due to the constraint, and so it's quite hard to consider the constraint as useful.

If the constraint is useful for protected areas in some other countries then this shouldn't be an excuse for messing the date in other countries. If all protected areas in, say, France or Canada have IUCN category, then why not add the IUCN statement constraint to some other national identifier (property) instead? Or, if there is no other identifer to be associted with this constraint, then you might consider using complex constraints instead (apply the constraint by appropriate region, or something like that).

The "suggested" constraint is fine, it's say that you are not obligated to fill out P814.

A posibility, we can also add a has cause (P828) qualifier to P814. It's generally wath I add when a archeological site have a address restricted (Q47460806) (ex.: Dead River Cattle Mound (Q17513341).)

Where does it say "you are not obligated to fill out P814"? I mean, I get that you are not obligated to edit Wikidata, but if you do then you probably try to get it right by following the directions and the direction due to this constraint is to fill out P814 without exceptions. So far the only clarification I've found on what does suggestion mean from item description here: it only says that "violations are not as severe".

I think we have established that it is very common and also normal that there protected areas (WDPA objects) that are without IUCN category and that are not supposed to have IUCN category. So "no value" (not "unknown value", which is just wrong) with/without qualifer would not serve meaningful purpose. Unlike e.g. coordinates in your example, where type of object is expected to have coordinates and it's useful to provide the information on special circumstances about why they are not known. These IUCN statements on the other hand to my understanding would serve only the purpose of being a workaround for hiding the constraint violation, which is counter-productive to the purpose of providing data that is useful, meaninful and relevant.