P: Ability to NOT import image keywords

I have to agree with Ian, there should be an append or replace metadata option

Yes, the inconsistency in behaviour is extremely irritating and has been drawn to the attention of Lr engineering folk more often than I care to count.

Lr should offer users the option to "append" or "replace" metadata, much as Bridge has done since version 1, but that's the subject of a separate and very long standing request.

Well put.

I don't think this option should be considered mutually exclusive with other keyword handling improvements. Often when I import foreign photos into my catalog, I just want to deep-6 all the keywords presently assigned, and then assign my own, or not, as I see fit. If there was a substantial opportunity cost involved then obviously that would be a consideration, but this feature would take very little time to implement - really.

I concur with Ian. Now that the image quality issues have been addressed with 3.x it is time to get serious with adding effective DAM tools. Providing a robust and flexible keyword tools is high on my priority list. I have several suggestions that I will be posting soon.

If I import photos that have embedded legacy metadata, such as Title, Caption, Copyright and Keywords, and apply a metadata template during import that includes new Title, Caption, Copyright, and Keywords, then I would expect all of the data in the metadata template to *replace* the existing metadata when I view the imported photo in Lightroom.

This is almost what happens.

I would not want to see my new title, caption or copyright appended to the embedded data in those fields, and thankfully Lightroom replaces embedded data with the data in the template as it should. Yet with keywords, the existing words are always appended to the new. I would just like a similar level of control over keywords entering the catalog that I currently have over other metadata fields during import.

Let me put it another way, if you create a metadata template that is completely empty of new data, but check the little box next to every field in the template, and then apply that metadata template to a new import, the only legacy data that is still imported into the catalog are legacy keywords.

A check box in the keyword section of the metadata template that explicitly prevented Lightroom from importing legacy keywords would be one possible solution I would support, but if it just behaved the same as all the other fields then that would work for me too.

On a keyword-related note, there is an ongoing discussion regarding better keyword management, which I fully support, if anyone has missed it:

http://feedback.photoshop.com/photosh...

It has taken me a while, on several occasions, to get rid of keywords imported from other peoples photos that did not go along with my keywording scheme - I think this is a great idea, together with an option to be prompted before new keywords are created. As has been pointed out, if one is diligent, one can delete them after import (hint: remember to purge unused keywords too). And in fact, after having made the mistake a few times already, this is exactly what I do (I usually remember now). Still, the more things that help, the better...

I agree about the problem of managing keyword lists, and I suspect you know that I've stayed with a flat list. I agree with your description of pollution as one of the tipping points that makes managing a hierarchical list impossible. List management, what you call an editor, is the real need here - not skewing the import process so it becomes a method of sorting out legacy metadata.

I'll add my voice to David's. The root problem is LR's keyword management capabilities. We are in desperate need of a more elegant way to manage and maintain keywords. Simple things like scrubbing for dupes, hierarchy management, more powerful keywording groups, and so on.

Respecting the existing metadata is, I think, fundamental to the "non-destructive" nature of LR.

I agree with both John B and Ians points. If we are to use this platform as a means to suggest where attention needs to be paid then a more robust KW editor is in order rather than a modification of an existing feature.