As you can see, the fixed navbar can take up a disturbing amount of screen real estate, even on desktops. I’ve never liked the fixed navbars for this one reason. I consider this broken and I’d hereby like to propose a fix.
The navigation bar should work as follows:
The navbar is hidden once you start scrolling down. As soon as you scroll up a few pixels it is shown. After that, when you scroll down again it gets hidden. When you have reached the bottom of the page the navigation bar reappears.
This way, the fixed navbar would be perfect, even for mobile devices.
Wow, this is like the third independent thread about this. The disappearing header is certainly a trend (I’d argue that it’s a somewhat thoughtless one)… but some issues as mentioned before: you miss notifications as they happen, you have no immediate frame of location reference (especially if you have multiple topics open), you have to scroll up to access basic functionality like search, etc…
Makes way more sense on mobile than desktop based on the previous conversations. I’m not sure you can sell me on the idea that removing accessible features to gain 150px of vertical space is making the experience better for anyone on a desktop machine.
What exactly does a fixed header break other than eating into the absolutely massive wealth of available screen real estate that most desktop machines have?
As mentioned in my post, you don’t have to scroll all the way up. Scroll up just a bit, and the navbar would be visible. Please take a look at the examples and projects I linked to in my earlier post.
I, like many others, find it distracting and inconvenient, which is why I am trying to suggest an improvement.
It doesn’t matter how far you need to scroll up — If I get a notification as I’m scrolling down I don’t know I have a notification until I scroll up. I might not even scroll up again on this visit to see that I have a notification.
That’s a broken experience when you’re talking about a platform that has real-time notifications as a primary feature.
I get it, it’s a popular trend that’s popping up that has been used on some really elegant looking websites – but we’re not talking about hiding pages of content on Discourse, we’ve talking about hiding core features.
It’s also a functionality that would be easy to implement on a personal install of Discourse.
I’d scroll down if I am reading something, and not when I am waiting for notifications.
When I am done reading or when I want to check for notifications, I simply scroll up a little (which brings the fixed navbar back in view) and I can see the notifications right there.
If I am going to close the website I am signed-into and usually expect notifications from, I’d at least do one last check for notifications before closing it (which, if I am at the end of the page, would be to scroll up a little).
This is how, I presume, most people would do it. So I find your excuse unconvincing.
Please stop saying that. I don’t suggest features/improvements because they are tending or popular.
Consider what’d happen if most people kept their ideas and thoughts to themselves just because they are easier to implement.
I wouldn’t even bother modding my forum to implement this feature. I setup the forum as a comments platform for my blog.
As I can’t code, I am doing my bid by reporting bugs and suggesting improvements. Whether you implement it or not, I leave it for the team to decide.
These statements are really starting to become a stretch.
You sit around on a page waiting for notifications to happen? No - you see them happen and you choose whether or not now is a good time to respond to them. I don’t know about you, but I want to see notifications when they happen because they’re notifications.
The other two points you made are taking a passive behavior (being notified) and making it active (looking for notifications). I no longer get notifications, I have to look for them. It’s adding a barrier to a fundamental feature. It’s also adding a new behavior for a user to learn. There’s a lot a new user needs to learn here, and I don’t think adding something that obscures that process is helpful.
I appreciate the idea and it’s come up and been debated a few other times. I don’t think it’s appropriate for the desktop experience of the platform, and I haven’t seen a compelling argument aside from vague statements like “it saves screen real estate” or “the header is distracting”.
It’s been mentioned in the threads I previously linked that this is not in consideration to be added as a feature because of these, and other, reasons.
