The captions accompanying displayed sample images in the browser are static and often have been updated in the museum database. These captions which in the case of the Rijksmuseum are the object titles from the museum’s database and sometimes date from the 1970’s, have often been updated, for various reasons. Most important here is the replacement of outdated and sometimes offensive terminology in titles. So the browser displays as captions old titles which sometimes have been replaced with new titles in the museum’s database.
For example the code [46A81(GIPSIES) wandering communities: gipsies] shows as an example the etching (RP-P-OB-926) by the Master of the Amsterdam Cabinet of a family of ‘gipsies’ with the old title ‘Zigeunerfamilie’ in the caption, but which now carries the new titel ‘Roma familie’ in the museum’s database. Since 2018 the Rijksmuseum does not use the term ‘zigeuner’ which is regarded as derogatory by the Roma or Romani people. Instead terms as ‘Roma’ or ‘Roma and Sinti’ are used.
So how to solve this? Can Iconclass update their imported captions/titles? Or does the Rijksmuseum have to check all examples where offensive termninology might appear?
Daniel, this is indeed a complex issue.
On the one hand it is a terminology issue. Iconclass does indeed still have outdated concept definitions that can be experienced as offensive. Your “gipsies” example is a case in point. However, researchers who will try to retrieve objects on the basis of older catalogues or secondary literature may still want to use offensive terminology. I am assuming that the Rijksmuseum still warrants that type of usage. Hence, “zigeuner*” still retrieves 59 objects from the Rijksmuseum collection.
It is indeed important to use the forum to point to this type of terminological problems, so thanks for the observation! Others can chime in.
On the other hand it is a synchronisation issue. The examples of Iconclass usage which are taken from the Rijksmuseum open access dataset, are not dynamically updated, which means that editorial corrections and changes at the museum’s end are not automatically harvested by the Iconclass software. Undoubtedly the outdated Rijksmuseum’s metadata will also be found on other websites that have harvested the RM’s dataset, but have done so a while ago.
This synchronisation problem also works the other way around. The Rijksmuseum software does not make use of the LOD options available at the Iconclass end. So, editorial corrections at the Iconclass end are not automatically picked up by the museum.
An example can be found in the metadata for the portrait of Jacobus Elisa Johannes Capitein.
If you look under “more object details” on the website’s catalogue entry you will see the term “Africans” as a subject term. This is a concept definition in the Rijksmuseum’s local Adlib database. It is not the definition of 32B321 (indigenous) peoples of sub-Saharan Africa that is now included in the Iconclass Browser.
In short: we are looking at a two-sided synchronisation problem that can only be solved by action on both sides. I’m sure other websites and databases using Iconclass will have similar issues, and I would very much welcome their comments here.
At the same time, the technical aspects of the issues can be very specific, so it may not be very efficient to discuss them here.
Let’s wait and see …
This is a question of someone sitting down and actually doing the work of re-downloading, and importing the data from the Rijksmuseum into the Iconclass database. If there were a magic button that I could press to have that done, please know that I would have hammered it long ago.
It has been on my todo-list for a long time, but truth be told, there has always been many other things crowding it out. If we had a larger team, or even more volunteers willing to help that would be great. But alas, to date, all my requests for help from many large institutions who happily use Iconclass in their projects have fallen on deaf ears. Everyone claims to be understaffed and underfunded.
How to crack this nut remains a mystery. So all I can offer for now is: yes, I will get round to it, eventually. Please send encouragement! 
Know that your blood, sweat and tears are greatly appreciated Etienne. Of course we hope that a re-download and update will be possible sometime soon. In the meantime i can make inquiries into the possibilities of updating by hand the most offensive terminologie.