always a slightly controversial subject but they sometimes fix a necessary evil of added protection. The question as always been about their respective clarity and any adverse affect on images.
The higher end lenses usually come with their own protective meniscus front lens. This is meant to be sacrificed in an unfortunate situation to protect the Special glass lenses that usually sit immediately behind them (as well as elsewhere in the lens). So, they are never a real issue or concern. It is the other medium grade lenses that do not come with a meniscus lens and that you want to protect; especially wildlife lenses used in environment that are likely to attract an accident or excess dust, moisture etc, that the issue might arise as to filter or not to filter. The answer to this usually depends on how good and non-destructive the filter is to the image quality.
Mongo had this issue when he got his 200-500 Nikkor. It is used in anything but a studio-like environment so protection was a high priority. Research at the time revealed that some of the older Hoya UVs had some of the best figures ever for light transmission, UV cover over the relevant spectrum range and least distortion. However, they are no longer made and almost impossible to find in 95mm size. Mongo found that the best available at the time based on lab test and figures was the new Hoya Fusion Antistatic filter and so invested in one. Even so, its figures may still not have been quite as good as a particular older Hoya UV"0" which is no longer available. When looking at the lab testing, it came as no surprise that some of the fanciest names and prices in the business did not rank very well at all.
Non of this directly addresses Perdo's point on sheer strength of physical protective power if that is what you are using the filter to do and Mongo cannot say that the Hoya fusion is anything like the strength of the ceramic filters he has mentioned. In the overall scheme of things, most of us would weigh the various aspects of a filter eg, strength, clarity, etc to decide what might be best if we had decided that a filter should be used on any particular lens. The answer to this question remains relatively unclear and that is why the same question seems to continue. One day , perhaps the perfect filter which addresses all the relevant issue will be developed and this enigma will be no more. In the interim, we are left to the best available research and technology at the time to help us decide and choose.