Remove Refresh button

I believe I had mentioned previously that I could see having an "auto refresh" option (on by default?) and that if this was disabled then we could show the manual refresh button. It sounds like the general appetite is to err on the side of simplicity, which I can understand. I'm inclined to say we should proceed with the removal and if anyone can make a compelling case to reinstate it, reversing the removal should be fairly straightforward.

Created a patch for removing the Refresh button, please review.
Thank You.

There are two schools of thought here: The one that guided our early decisions and the one we've heard from UX feedback.

Early Perspective

We should not lose any functionality between our same-page-preview and Drupal Core's preview functionality. We should be able to implement all the things that already exist. That way people won't be confused or upset about missing functionliaty.

UX Review perspective

Users want simplicity. They largely don't know Drupalisms. They want a quick check on their work that is unencumbered by complex decision making, specialization and options.

And an extra perspective:

Accessibility Review perspective

All users want to use a preview feature, as long as it works for them. And means a number of things:

• preview needs the ability to be navigated by tools / screen readers.
• The visuals needs to support low contrast.
• It need to be useable with just a keyboard.
• I'm likely forgetting a number of takeaways from that convo.

All that said, I'm getting on board with the idea that our preview does not need to be the same as Drupal core's preview. Add features like view mode selection and a manual refresh button make the accessibility support harder. So if we have a path to removing them, I'm in favor of doing that.

We could potentially remove the entire PreviewControlsForm. That would simplify our PreviewPaneController by removing the form injection and our need to pass the node into it. Removing the controls removes quite a bit of functionality for our preview but I justify that in my mind by thinking those specializations of the preview are best suited for a stand-alone preview experience.

(and we just assumed people would want these things, we don't really know)

That's where my head is at right now at least. Let's make some cuts of features and lock in on delivering a "reactive" preview that responds to field updates.

In general, if we can really nail the auto refresh having a manual refresh button shouldn't be necessary. But I'm not sure if that can be the only option for a11y reasons. It really depends on what the experience is like with auto refresh enabled. Would auto refresh result in frequent announcements that content has been updated (not really sure what that best practice is here?) If so, I could see that being disruptive, and then requiring that we have a way to disable auto refresh, which might bring us back to needing the button again.

Also, I'm not in favor of removing the view mode selection (even though that adds complexity.) I think that is a very useful feature and makes it possible to 'same page preview' in contexts that wouldn't be possible otherwise.

I guess it'll depend a lot on how often the content is updated or when that action is triggered. If the goal is to remove that option the UI should give enough confidence to editors by communicating when is the last time the content has been updated. This could be a "last updated 4 seconds ago" kind of text/message or a "notification" or state that shows when the preview is "up to date". So whatever is decided, there will be the need to save some space for this visual communication.

We're a go on this work. At Midcamp, we got together and discussed how with the work we've committed in Phase 2, the refresh button isn't really doing anything. It can go.

Wrapped up the proposed changes in a MR.

I will review this issue

Patch is applied,
its working properly Looking good to me. so moving to RTBC please refer the attached screenshots

Merged, moving on

Automatically closed - issue fixed for 2 weeks with no activity.