I dunno about others, but I always change my tag colors from the default (my user color). I want it to behave like how connection colors are random. In my observations, most (if not all) of the tags I see in the wild are also different from a user’s color.
I also agree
the regression/wrinkle with this is that while you’re typing in a tag, the use of the default color lets you identify whether your current typing is a new tag and how close you are to an existing tag (which would likely have a custom color in this scenario)
Maybe you can use a different indicator. Because as it is, it’s going to still match other tags that are still your avatar color.
No color at all:
Using a label:
If these assumptions are correct :
1- you’ll want to choose your own label colors anyways
2- autogenerated random colors are something you’ll probably want to tweak
3- you’re reusing existing tags often and making new tags less often
Then having to explicitly label tags as new or existing feels like a regression
For me, all three of these assumptions are wrong
- Almost always, the only thing I care about is that the tag color is different. So, not my avatar color, and not an existing tag on the space.
- As long as the random color is different enough, I’m good (see above).
- I’m 50/50 here, but I still think reducing the steps in this flow is good.
I don’t see how this is a regression in behavior. It’s a change, but it doesn’t get in the way of the normal flow. As others in this thread have said, and also in observation of tag colors out there, it seems like people never want the tag color to be their avatar color. So why not make it random and remove the tedium of tweaking the tag color for most users?
Explicitly labeling tags as new is additional and helpful information—I don’t see this as a regression from the user’s perspective. As it is, since the default color has been the avatar color, it potentially ambiguous still whether the tag already exists or not because you already see other tags with the color.
i meant that if it’s less clear (or requires close reading to figure out) whether a tag is new or not, that’s the potential regression. That said I’ll think about how to do this
Just putting this out there (I know this isn’t how your other auto-completes work), if the search weren’t so fuzzy and got whittled down to nothing once you typed a non-matching tag, then that would be unambiguous…
update releasing soon:
- new tag color is random
- new tag is explicitly indicated w label
Love it. I think you’ve handled your initial concerns really well and avoids confusion, at least for me.