First of all, let me say that I appreciate you holding up the discussion with the consistent respect to my arguments. I may have been too insistent on my being right in the past without doing much to prove it, which does not contribute to a discussion in any way. I've realized today that holding up a discussion with the point of advancing knowledge is all that really matters in communication.

The first is that when I made the distinction between web design and web development you said that's not the point.

So, your argument is that because Bootstrap (as I keep emphasizing, though there are others) is a web development tool and not a web design tool, it doesn't deserve the criticism that I've expressed earlier.

After giving it some thought, I'd say that a) Bootstrap, indeed, doesn't deserve that kind of criticism, b) no CSS framework does unless it explicitly states to retain the best qualities and design practices available while not holding up that standard. Moreover, it was unfair of me to pick on a single specimen of a wide array of CSS frameworks simply because it's the most well-known - something I've blamed other people for on Hubski recently.

My blame of Bootstrap was motivated by the fact that I felt like it should somehow promote the idea that it's only a baseline for design while encouraging the user to move past it and into the territory of more elaborate, educated design - not simply learning more CSS, but also color theory, the holy grail design and other such things. It's not a fair premise, however: though it is not a bad practice to encourage others to learn and educate themselves for a better result, it's not a responsibility for those who promote design/development frameworks (and Bootstrap is a design framework, but only in part) available and aimed for the mass usage. Would have been cool if it were, but it isn't and it doesn't have to be. It doesn't kill you to have bad design, unlike, say, smoking cigarettes and drinking alcohol.

I don't think Bootstrap is "heavily" customizable. I think it's versatile by its nature and that one could use their creativity to produce amazing and unorthodox results (I was going to prove my assessment with screenshots from the web, but Google has malfunctioned on me right as I tried to find some). In this respect, it's similar to LEGO: both are massive toolkits that could be flexible enough to produce a wide array of good results. That being said...

The second issue is that i'm unsure what you're constituting as an opinionated framework? <...> You could just have the grid if you wanted, with no CSS, pre-made components, or JavaScript at all.

Have you actually tried to download only the Grid component? I have. It includes normalize.css by default, with no option to opt out. That's "opinionated". Just how opinionated? Of the ~1000 lines of the main development .CSS file, something like 250 are normalize.css alone. To add insult to injury, it contains default classes I didn't ask for (like .img-rounded and .hidden), which adds up to about 75 lines. A third of Grid-only development .CSS file is something I didn't ask for and had no choice to opt out of.

At least the theme .CSS is a separate file.

Long story short: I wasn't justified in beating Bootstrap's drum as far as design practices are concerned. You're right in saying that it's on the user, not the toolmaker. However, I don't appreciate its bundleware status. No framework should tell user "You must use this with our product". Recommend? Knock yourself out - but don't put it into the package anyway without asking whether I want or need it first.