Can-o-worms!
The problem as I see it is that I have my own concept of what these kinds of words mean when I'm dancing about architecture and I think I often presume that others have the same basic idea structure behind the concepts these words are meant to convey. From time to time, that presumption is blown out of the water and I realize that others are thinking something entirely different.. like in this thread. So we can talk about our own definitions, but in the end I see the primary value of this thread as a general warning about subjectivism in descriptive terms about sound.
I like what Todd is saying. I also tend to lean the same way when it comes to personal sound preference (I think!) For me the terms bright and warm are tonal modifiers, they are analogous to bass and treble tone controls. You can have too much or too little of either, but in the absence of other things, they are aspects that careful eq adjustment can correct. I see them as covering a relatively wide range of frequencies at each end of the spectrum (low Q, in eq terms), the ‘meat and potato’ zones on either side of the midrange, not the extreme bass and treble. The terms thin, and dull seem to me to be subtractive aspects of warm and bright respectively.
Although I'm sensitive to the balance of sound across the frequency range, I’m most critical of the upper midrange. A recording can be too bright or not bright enough. If it's bright then it has too much energy somewhere in the upper midrange or treble. I don't prefer it, but I can more easily accept less than the ideal amount of energy in the bass regions (a good thing, as it makes the possibility of listening on anything with less than a 10" speaker enjoyable), where as too much bass energy at a certain frequency is always too much. Maybe that is why warm is more problematic of a term.. it always sounds like it should be a good thing.
There are other words that may also fall in the tonal category such as bloated, muddy, boomy, hollow, nasal, tinny, spitty, piercing, etc (arranged generally from low to high frequency), but those terms are much less precise for me, even though they are always used in a negative sense, so at least that part should be clear. At the very least they seem to refer to a less broad portion of the spectrum, describing a higher-Q frequency things, resonances, ringing, room modes, but they might also be used to convey non-tonal ideas as well.
Loose, tight, clean, saturated, fast, etc. seem to describe dynamic and transient aspects. There seems to be a definite polarity built into these terms themselves: good vs bad, as well as the dynamic thing it is describing. Maybe that makes them easier to agree on than warm and bright.
I like the term 'transparent', partially because it conveys no tonal idea for me at all, and so it carries less personal preference baggage. For me it conveys more of an ideal, a sense of clarity, being able to hear a sense of depth and detail (a few more idealized terms). Personally I value transparency, clarity, depth and detail so highly that i sometimes make choices that others might consider slightly bright or a sacrifice of warmth in an effort to enhance those aspects, using tonal tricks that trade on what might be my preferred choice of perfect tone in a better recording.
Now I’ve thrown out a bunch of terms with only partial definitions. Sometimes I use the same word to mean to different things. Does depth relate to being able to hear a sense of distance, a sense of clarity in space, or does it refer to how low in frequency the bass range goes? Depends.
Worms.