It's because you are usually explicitly contrarian to whatever was posted, extremely sure that your own opinion is the only valid one, assuming a lot of things that are incorrect. It's tiring.
Well, for those areas where I think everything is already perfect, or as good as I have the capacity to recommend on, I don't tend to see any need to say something. If something is done right to that degree, it is presumably done so with intention; that is to say, the person who does it is aware of what was done well because they went out of their way to achieve that. Similarly, why would I say something if I didn't have a reasonable confidence that my opinion was justified in fact? If it's based on a lack of foundational information, then it would be helpful to clarify; if I don't know, I'm probably not the only one. Of course, if you're too tired to be bothered, that's understandable as well, especially the way things are these days. Certainly with all the fires I've been a bit stressed myself; it happens.
What's "restricts"? No matter what you pick, you need a color scheme for the UI where the components work well together. No matter what you pick, every color added narrows the range of what other colors can be used and look good. If you can make a stone/metal that looks good, by all means do so, but so far the color choices aren't nearly as versatile in terms of the range of screens that they look good on as it would be to stick with tried and true options. You haven't even really got control over whether the elements will be predominantly warm or cool tones.
That's your response. On my note that I personally favour rock/metal.
You assumed that I designed the UI.
You assumed I said anything against parchment.
You assumed I currently "haven't even really got control over whether the elements will be predominantly warm or cool tones." - What?
You start with a rhetorical question, then telling me " by all means do" what I like, followed by the obvious "but" here is what you should actually do.
I was using the general "You", speaking to the design principle. I'll try and avoid using second person in such abstract circumstances, if that's particularly bothersome to you. It's the kind of thing I can easily forget though, since we talk only relatively sporadically over months and years, so if it happens again, please bear with me. I can try to avoid rhetorical questions as well, if that's something you perceive as inherently rude, but I use them a lot and will definitely forget to check back and rephrase things at some point. I already caught two in this very post where I'm more likely to be aware of it than any.
Phrasing aside:
I didn't assume you designed the UI, but I did assume you had some degree of say in it, based on the imperfect understanding of the design process. I interpreted your statement about parchment as being against it, which I don't think is an unreasonable interpretation based on the uncaveated way that you phrased it, but if you merely meant to highlight a potential weakness, then I apologize for my misunderstanding. I'm sure I've been similarly unclear myself. With regards to the lack of control over warm or cool, I was referring again to the variety in how these colors can appear on different monitors; although the colors are very slightly warm, they are close enough to a purely even tone that given the tendency many screens have to emphasize blue light, they can wind up on the cool side of things. In other words, the colors chosen do not exert a very strong control over what the end user will perceive, because their position within the spectrum is near to a point that makes a substantial qualitative difference. That point falls within the range of variation in how the colors defined in the software are transmitted through the hardware to the wetware.
On the last point, I can see how my phrasing seemed petty and domineering, for which I apologize. What I meant is that I recognize that the stone/metal palette could look good, and I would by no means be opposed to it on principle, but it is more susceptible to a particular limitation which has not been overcome in the current (and, I know, definitely not final) implementation. I was also highlighting that there are alternatives which dodge the issue more effectively.