How can the interface for choosing canned comments when opting to delete in Stack Exchange’s low-quality queue be improved to make reviewers more aware of the actual effects of their choice.
The current interface faced by reviewers who opt to delete a post in the low-quality queue on Stack Exchange is this:
They are shown the post and then can make their actual choice, i.e., whether they want it to be deleted, want to edit it, etc:
Reviewers who opt to delete, are presented with a choice of canned comments to be left on the post on their behalf:
Many reviewers never or almost never choose the no comment needed option, even though there already is a comment expressing the same thing on the post (thus making the new comment redundant) or none of the comments applies. Please assume this problem as given for the purpose of this question. If you want to debate this, here is a question on Meta SE where it would probably be well placed.
As far as I can tell, the reason for this is a mix of:
A variation of banner blindness that makes reviewers never (re)read the header of the canned-comment dialogue.
Users think that their choice in the canned-comment dialogue does more than just leaving a comment such as contributing to some statistics or affecting what happens to the post (e.g., this is a commentary actually effecting comment conversion).
Users think that no comment needed is not the right thing to do and may be held against them. (This is is not completely unjustified, because it is actually true, when one of the canned comments is spot-on and there is no equivalent comment yet.)
The interface focusses on the reason for deletion (e.g., “This is a ‘thank you’ comment.”), not on the comment that is actually chosen.
How could the interface be improved such that it is more clear that the canned comments are just canned comments and that leaving no comment can be a valid choice?
What I considered so far
The following options do not convince me though at least the first one would be better than nothing:
One could add more explanation to the no comment needed option such as a text explaining that “This is a good choice if an existing comment already addresses the post’s issues.” However, I fear that this may be affected by the same banner blindness and will be ignored.
One could add another level of dialogue: After opting for deletion, users will be presented with a dialogue that allows them only to choose between no comment, canned comment, and possibly custom comment (explaining these choices) and only presents them with a selection of canned comments if they choose canned comment. The disadvantage of this is that it is yet another layer of dialogue and it is not clear what the choices of canned comments are in the first layer.
Remove the bold headings displaying the reason for deletion (e.g., “This is a ‘thank you’ comment.”). This should considerably increase the effort for finding the right comment, when one is appropriate.
I’m trying to find an android phone app that will function similar to Google’s pre-sceening functionality that asks the user who they are. For example when someone calls me they are prompted to “Dial 5 to continue” and this would theoretically prevent bot calls.
Any thoughts or answers would be highly appreciated, Thanks
I am challenged with presenting 2+ applications & their user flows to my product partners as a bird’s eye view. Some of the steps will be common & some unique to these products. The purpose is to highlight a common step- user validation that occurs across a few applications but at different points during the experience. Any suggestions on an elegant way to solve this? Here’s a quick & dirty visual i came up with. Any thoughts on this representation will be helpful]1
We’re using outlined textfields from Material Design.
On some occasions, chrome(or any other browser I guess) can auto fill the data because the users has saved them. Chrome gives a hint to the user by setting the background color of the textfield to a light blue.
I find the part where the blue background crosses the label quite hard to read and very awkward.
I don’t want to force background color to white because I think overriding default browser behaviour is a bad practice that would not serve our users.
Couldn’t find any reference to that on MD docs.
Tried to force background color to white and label+outline to the “autofilled” color but I’m not sure users would take the hint.
Do you guys have any suggestions?
I need to use colour to indicate priority. Say, there are 20 priority levels. What is the best way to use colour to achieve this? My thoughts so far are:
- I’m leaning towards using shades of a single colour (say, red – from lightest to darkest); using different colours might make sense when the colours are shown in the order of priority, but when items of different priority in a list are mixed up, I think different colours wouldn’t work well…
- Is there a sound mathematical approach to arriving at these colours? I prefer to get the colours in hex values
I understand that 20 priority levels normally is too much, but 20 here is just a proprietary number, it can be 10 as well. The thing is, I have a priority score in the range of 1 to 100. And since I have a constraint of not being able to show the numeric value, I want to help the user gauge (not necessarily identify very clearly) relative priorities. Hence the 20 levels (with increments of 5 if it’s an arithmetic progression, or variable if it’s some geometric progression) help me provide this information to the user.
Also, thanks to everyone for the responses, many are very useful and interesting.
Our website uses a system where the username is the user’s email address. We have a function which permits the user to change his email.
What is the best way to go about this?
- Send an email only to the new email address just to confirm the change.
- Send an email to both the new and the old email addresses for confirmation.
- Send an email to the new address, asking the user to click a button for activation.
- I’m open to any different suggestions as well.
I’m concerned about functionality, security and usability. There could be some problems with activation; for example, activating an email address that has been created between the change request and the changed email activation would certainly cause some problems. I’m probably forgetting other corner cases…
I’m being told by an experienced BA (> 25 years) that “Numeric numbers are usually totaled therefore should be right justified.” And that text should always be left justified.
I have never heard this before now. Is this a true “standard”?
I have to code a widget displaying a table on a touch screen and I’m wondering what is commonly used to do so? I thought about holding click down between two rows to insert and holding click down on a row to remove it, just like this:
Since it’s my first time in the touch screen world, how can I make it intuitive and help the user to “guess” the way he can insert and remove? Because on a regular computer I’d pop some kind of tooltip, if the key stops to move, but touch screens are different because finger is not held when moving.
I know there are always exceptions to a design system.
But as a general rule, should you avoid mixing Selection Controls and Text Fields?
I tend to think Selection Controls belong more in a settings screen, and Text Fields are more for form builders and other data inputs.