The example screen-captures in the article appear to be of something subtly different: involuntary promotion of new features or recent changes. There is probably a more established term for this UX tactic, but I know it as a "guided demo." These guided demos are often employed when orienting a new user with key elements of the application's user interface.
1. Tooltips are in response to voluntary exploration by the user. Guided demos are usually involuntary.
2. Tooltips are available at will and infinitely repeatable; they always appear when the user is "inspecting" an element. Guided demos usually are once-and-done (and are notoriously difficult for users to revisit later if they elected to skip them).
3. Because tooltips are voluntary, they occur at the user's area of attention. Guided demos are typically attempting to draw user attention to somewhere it's not already. Hence, guided demos are often coupled with some form of obfuscation of other UI elements (e.g., a dimming mask layer over the rest of the UI).
4. Tooltips are immediate both in and out—they appear on hover and disappear on exit. A tell-tale of guided demos is that the pop-ups usually include some form of dismiss button ("OK" button or close button), because they are involuntary and therefore need to confirm they've successfully captured user attention.
5. Tooltips are individual and paced by the user's interest. Guided demos are commonly linked into a "wizard-like" sequence (e.g., let's walk you through the three most important elements of this UI.)
If it's not clear from the above, I feel traditional tooltips are generally good because they reward exploration while remaining entirely voluntary. Meanwhile, guided demos are a mixed bag: sometimes a necessary evil to clear an especially difficult UI hurdle, but easily used to excess and disruptive to users.
I might be in the minority but guided demos today seem to achieve very little of these desired effects: they usually start at inconvenient times (I might still be configuring my profile or registration settings), as bhauer mentioned they're hard to re-activate and re-visit, they're not comprehensive (usually only cover the top 5-10 actions or functions), and generally a little overwhelming when getting accustomed to a new interface.
I will acknowledge today's user interfaces are different to late 1990's software where everything had an explicit button or menu action, particularly with touch screen interfaces now. But given how often I'm walking friends or family through common actions on popular interfaces I would say today's interfaces are doing a terrible job of teaching users about what feature are available and how to use them.
At the same time, I find the forced, “guided” tooltips to be invasive and annoying.
I love traditional tooltips. They don't work so well for touch devices, but if you've got a mouse they're great and can almost be seen as a progressive enhancement.
I'm also a fan of the even older tooltip cousin where a bottom status bar of a window has info about the object in the window that currently has focus. This has disappeared from many apps, but is especially useful to keyboard users.
But the examples the OP are pointing out are a bit different. Rather than a "guided demo" which sounds like a sequence of these -- what's being covered here is usually a single solitary one, usually for a newly released feature, that appears on screen load, even if you are a long time user of the app, and without asking for any kind of guidance or demo, until you dismiss it (or in some cases they can time out on their own).
I have been noticing this phenomenon more and more, and think it is worth pointing out and commenting upon. Which is more interesting than debating whether "tooltip" is the right term for it; or talking about traditional tooltips which might be interesting, but isn't what the OP is about at all, someone can write a different article about that.
(They are "tips about tools", and _look_ like traditional tooltips, so I can see why the OP author called them tooltips. But this is clearly confusing many people. I guess I'd call them "feature disclosure callouts" or something like that).
For example, on Android, long pressing an icon in the top bar of an app will show its label.
Tap on a bubble and it expands into a larger one or navigates to a separate help page.
It is an effective way also to learn an app - by doing, not reading.
If you need tooltip or guides like those, your UI has other serious issues.
But rather "callout bubbles" that appear on initial screen display, to point out/explain a new feature (or a feature they want to 'advertise' for some other reason). They usually either stay there until you dismiss them, or until they time out.
I have been noticing these more and more, and I think they are worth commenting upon, I'm not sure this article does a great job of it, starting with calling them "tooltips" that is misleading some people not reading the article carefully to miss what it's actually about, as evidenced in the HN comments here. (To be fair to the OP, they are "tips about tools"...).
One thing I wonder is how effective they are, or if there's any public research on it. Myself, I'd say ~75% of the time I dismiss them without even reading them, as being there on page load (and often covering up other parts of the page), they're getting in the way of whatever I came to site/page/app to do, and have in mind to get done when I see them.
Coach marks/app tours can be great, however apps that have them, but don't have an easy method to skip make me rage!
The OP examples seem to be more the latter.
"Coach mark" is a good term, thanks.
they are usually beyond obvious, further rage inducing.
when they are actually useful, again still rage inducing because I skipped it and there is generally no way to get the coach tip back.
sorry, ad pop-ins have ruined what would otherwise be fairly useful.
Sometimes powerful interfaces for doing complicated/powerful things need to be learned.
(And an airplane cockpit can still be designed poorly or well at varying degrees, certainly! A well-designed one will be safer and easier to use. But nobody thinks an untrained person should be able to step into even a well-designed one and fly a plane. That doesn't mean it's a poorly designed interface).
This is me, too, although I don't read them more than 75% of the time. I find them disruptive. I've even developed muscle memory so any time I see a "got it" button, I just immediately click it so I can get on with whatever I was trying to do.
Now I need tutorials to figure out simple programs - even web apps - frustrating.
And while we're at it, "Tips of the day" on opening a software dripped important knowledge.
How is anyone supposed to figure this stuff out?
Also, like anything else in the app, settings are context-dependent. Some are system-wise, whereas others might be user-specific. Therefore, it makes sense to segregate them, IMHO.
Also within Settings, there should be a tab for "system" and "user".
Compare this with real tooltips, like the one you get in MS Word when hovering the floppy icon! It appears consistently on hover; tells you the button’s functionality (Save) and its hotkey too (Ctrl + S).
It's hard for me to think of them as anything but "fuck off boxes" after cursing them dozens and dozens of times.
It also doesn't block the actual thing _I care about_. I'm incredibly likely to just slap that tooltip out of the way so I can get back to what I'm doing instinctively, and then wonder what it was trying to tell me.
Any sort of tooltip that obscures the thing I'm trying to accomplish that I didn't request, or that I can't bring back on demand, is a shitty user experience.
They instead seem to be the kind that appear on the page initially, pointing to a new device and explaining a new feature, until the user dismisses them. I don't believe they involve hover at all.
I'm still not personally sure how workable this is on a small screen (because I personally don't have experience with it), and could use some discussion or examples of how this has been done (if it has been done) on mobile. It seems odd to me too to write an article about UX these days without mentioning mobile.
But it is worth commenting on the trend of "initially visible tooltips pointing out new or advanced features." Personally, I almost always dismiss them without reading them, they are usually getting in the way of whatever I came to that site/app to do. I wonder if there is public research on their effectiveness.
__To me__ those might be tooltips in terms of implementation (i.e., code) but I'm not so sure I would consider them TTs in the traditional sense. They're more like "feature flags".
They are tooltips in the plain language meaning of the word, they are tips about tools. But I agree the term "tooltip" initially made me think they were talking about the appear-on-hover kind.
"feature flag" also means something very different in software discussions, I wouldn't try to re-use it for these. Words are hard. https://en.wikipedia.org/wiki/Feature_toggle
I might call them "feature disclosure callouts" or "feature announcement callouts" or something like that, but that's a mouthful. "callout" is one graphic design term for that bubble with a pointer annotating the the thing pointed to. https://en.wikipedia.org/wiki/Callout
For me personally it would be nice if there was a reliable way to detect a quick jab (tool tip), vs a press, vs a long press, but I could see this does not working well for people with mobility issues etc.
Tap the balloon and it expands with more detailed explanation.
The help mode could be triggered by a tap on an always present (?) icon somewhere in the corner.
I think one of the big innovations Google Search brought to the table is the Google doodle. The main features of their interface have remained the same for over a decade: type what you want to search into the box, click "Google Search" and a list of results appears with links and brief intro text. There's new stuff on the search results page that wasn't there ten years ago, but it doesn't really interfere with the functionality. Such a simple interface would feel stale after 10 years for most applications, but having the Doodle means it changes over time, keeping it fresh in a way that doesn't interfere with the functionality. This allows them to keep users from getting bored with their product, while also not forcing users to learn a new interface with every refresh.
I have a lot of problems with Google and don't use their search (I use DuckDuckGo). But I think we can still learn some valuable things from them.
Tooltips are as old as the hills from my perspective, but they've been uncool for years, only BLOBs use 'em, too confusing for modern users, etc. I use 'em anyways.
Suddenly someone writes an article and throws up some material design clipart and poof, tooltips have been re-invented and are cool again.
Is this what it is like for the old timers seeing SOA and ESB coming back as microservices and 58323213455233999 layers of container/docker/k8s/rancher abstractions?
By "tooltips", I'm referring to those canned pieces of text in a balloon that show up when the pointer is hovered over a UI element.
The article is stupidly calling something else "tooltip"; namely "tip of the day" type dialogs that pop up and must be dismissed, which teach the user about features or usage.
Such events are not called "tooltips".
Please don't contribute to confusion in naming.
How do you provide the same impact of tooltips on a touch-only based interface?