Open Bug 944415 Opened 11 years ago Updated 2 years ago

[Australis] add distinct navigation buttons to palette

Categories

(Firefox :: Theme, defect)

Product:
Firefox
Component:
Theme
Type:
defect

Tracking

()

People

(Reporter: fb+mozdev, Unassigned)

References

(Blocks 2 open bugs)

Details

(Whiteboard: [Australis:P-])

Attachments

(1 file)

combined-stop-reload-hidden-state.png
534 bytes, image/png
Details 
Australis merged the navigation items (back/forward, stop/reload, location) into one widget, and that was good (not only due to removed magic). 

However, there have been numerous requests to allow moving of the stop/reload button (because they like to have navigation buttons next to each other) and for separate back/forward buttons (at least some dislike the conditional forward button). I personally really don't care and am happy with the current setup, but in the interest of customization, I'd like to propose the following solution (has already been raised on the firefox-dev mailing list): 

1. Add one (combined) stop/reload button. (The icon can be half-half when not in the toolbar/menu, see attachment for a synthesized button from the current buttons in release, just for illustration.)
2. Add one (combined) back/forward (keyhole) button with constantly visible (i.e. no conditional) forward button.
3. Both buttons are "hidden" by default (i.e. in the customization palette, not in the toolbar or menu), so users need to drag them to the toolbar to use them. 

4. MAGIC. When one or both of these buttons are in the (any?) toolbar, their respective counterparts in the navigation widget are hidden thus preventing visual duplication (and associated loss of space). 

Magic is bad, but IMHO the magic I proposed here is way less complicated (actually not complicated at all and intuitive) while fixing one of most commonly requested complaints of Australis. AFAICS, the solution does not introduce a big amount of maintenance work and adds only minimal duplicate code/data (I don't know if it's possible to fully reuse the code of the widget without copying it which would null any duplication and additional maintenance cost), so that's also not much of a concern.
I'm usually not one to suggest this, but this really sounds like a good idea for an addon. Personally, I'd like it to be an addon written by the UX team themselves, but this probably would work better as an addon rather than a built-in feature.
Well, yeah, this would be a prime candidate for an addon, though: 

1. interaction with the navigation widget
2. code duplication

… are both reasons why it would make sense to have this built into Firefox, apart from the not-marginal political reason (i.e. counter OMG-change or Australis-removes-my-customization).
Madhava, what do you think of Florian's proposal? Is this something we'd consider for the palette, or is it better to definitively label this as add-on material?
Flags: needinfo?(madhava)
Whiteboard: [Australis:P-]
mbrubeck - darktrojans addon WFM - \o/ https://github.com/darktrojan/buttonsback/releases
The UI change had my down vote so thanks for pointing there :)
After a few weeks using almost exclusively Australis, I am back to the release channel for this exact usability problem for me. Page navigation (back, home, reload) is just harder now with buttons so far apart from each other and I realize that those are buttons I use a lot more than I thought. I really need to have them grouped together and I really need the reload button to be bigger than this tiny widget in the url bar, it's hard to focus and it's not even obvious if I actually clicked it because my mouse pointer covers the whole widget...
Use this addon to back buttons: https://github.com/darktrojan/buttonsback
This is the first thing I customize when I install Firefox. The small reload button is both difficult to hit, and in the wrong place, making it real usability headache.

Obviously Australis is currently a pre-release, but if the ability to customize the buttons isn't returned to what we currently have in Firefox by the time it is generally released, then I will be switching browser.
(In reply to grahamtriggs from comment #8)
> This is the first thing I customize when I install Firefox. The small reload
> button is both difficult to hit, and in the wrong place, making it real
> usability headache.

I agree with you completely.

My biggest fear is that a user is being redirected to (or already on) a malicious URL, and the user cannot find the tiny button fast enough to stop it. On my 1920x1080 HD monitor, the button is less than a centimeter high!

I tried filing this problem alone as a security bug, but within 3 minutes an administrator cancelled the bug and said it was not a security risk. I completely and wholly disagree.
Version: 28 Branch → Trunk
(In reply to grahamtriggs from comment #8)
> This is the first thing I customize when I install Firefox. The small reload
> button is both difficult to hit, and in the wrong place, making it real
> usability headache.
>

The problem with it is Mozilla dont see this as an UX problem. I wish they could change it back. After forcing myself using Australis for months i still cant get it to stick. 

It is just plain wrong.
(In reply to Dan Q (RedBlade7) from comment #12)
> Do the people at Mozilla have extreme vision and super 31337 mouse skillz?
> ONE SQUARE CENTIMETER WITH NO VISIBLE BORDER! I still don't get it.

I understand that this is a problem for you, but I doubt most people would consider what you just described as something outside of reasonable expectations. If you are not capable of using a pointing device with the necessary precision, I suggest either installing a theme with larger buttons or using keyboard shortcuts.

https://addons.mozilla.org/en-US/firefox/search/?q=large&cat=2%2C0&appver=29.0
https://support.mozilla.org/en-US/kb/keyboard-shortcuts-perform-firefox-tasks-quickly#w_navigation

It might be nice if Mozilla would put out official alternate themes for high-contrast and/or large buttons for users with different needs.
(In reply to Dave Garrett from comment #13)
> (In reply to Dan Q (RedBlade7) from comment #12)
> > Do the people at Mozilla have extreme vision and super 31337 mouse skillz?
> > ONE SQUARE CENTIMETER WITH NO VISIBLE BORDER! I still don't get it.
> 
> I understand that this is a problem for you, but I doubt most people would
> consider what you just described as something outside of reasonable
> expectations. If you are not capable of using a pointing device with the
> necessary precision, I suggest either installing a theme with larger buttons
> or using keyboard shortcuts.
> 
> https://addons.mozilla.org/en-US/firefox/search/?q=large&cat=2%2C0&appver=29.
> 0
> https://support.mozilla.org/en-US/kb/keyboard-shortcuts-perform-firefox-
> tasks-quickly#w_navigation
> 
> It might be nice if Mozilla would put out official alternate themes for
> high-contrast and/or large buttons for users with different needs.


It's a pretty standard optical mouse on a pretty standard desk...
I guess I can take out the battery-guzzling expensive gaming stuff from the storage room if I had to

Still there is no possible justification for a borderless gray-on-white square-centimeter-size stop/refresh.
Since people here think it's all in my head, I have shown Australis to more people:

* A 56 year old farsighted female could see the button. Admitted button was small, but familiarity with Chrome. No problems using the mouse, even with injured hand. No other complaints (other than the size of my 23" monitor).

* A 56 year old half blind male could not see the button until I pointed to it. Asked me where the Stop button was. He attempted to use the mouse to access it, but slipped frequently.

I can provide more tests for individuals under the age of 35 within the following month or two.
NOTE: No hand injuries in the half-blind male, unlike the farsighted female.
This is a problem for me as well. What baffles me the most about this UI decision is that Australis is being trumpeted by Mozilla as an interface that's easier to customize. This UI regression contradicts that marketing assertion.
For my workflow, I need the reload button to be on the left side. Another windows are over the right side and since I can not change the position of the reload button, I need two clicks to reload the main page (one to take the window to the front, so I can see the right side of the url bar, and another one to reload the page), when I could put it on the left side, I needed just one click to do both, take the main window to the front, and at the same time, reload the page.
Buttonsback addon WFM ... I'm on nightly (31.01a) so I use version 5.

https://github.com/darktrojan/buttonsback/releases
For people that might be interested, I've created a simple 'stylish' user style that puts the reload button at the left side of the URL bar:
http://userstyles.org/styles/101143/left-reload-button
My problem with this is the button flips back and forth between "stop" and "reload." I do not like the risk of reloading a page I am intending to stop loading, potentially just as it loads. I want them as broken out single purpose buttons. 

Being immovable is silly for a "highly customizable interface."
Hi,

I just want to thank all those who developed the new right-click navigation buttons in Firefox 32 Beta. It is so much easier for people with large HD monitors since Australis created the despised-by-most #944415. 

This will help me a lot with my Linux desktop and 23" monitor, and we were discussing disabled users on this thread, such as my half-blind father who was also struggling with the tiny refresh/cancel forced since Australis. I haven't asked him about it yet, but I'm sure this will help most users who suffer from blindness or just the average RedBlade7 who wears glasses/contacts.

While I still believe the address bar should be customizable, the right-click buttons are a great help from the aggravation Australis has caused me (that Vimperator couldn't make up for).

Thank you!
Any updates on when this bug might get fixed?
Any updates? This is an upgrade blocker.
This should probably be closed as WONTFIX at this point.

For those wanting detailed UI customization beyond what is built-in, we generally recommend using:
https://addons.mozilla.org/en-US/firefox/addon/ClassicThemeRestorer/

That addon includes the ability to do this as well as many other advanced UI customizations.
I feel the whole point of this bug is it is a regression, and against the core design values.

https://blog.mozilla.org/ux/2014/03/introducing-the-update-experience-for-australis/ Links to https://people.mozilla.org/%7Emadhava/FDV/FirefoxDesignValuesBooklet.pdf . 

Specifically:
1. Takes care of you - No surprises. A stop button that converts to a reload button based on the actions of a remote network and webserver over which I have minimal control.
2. You help make it - Start people with smart defaults + Implicit as well as explicit customization. No problem, we start with a button many will like, but remove functionality otherwise. 
3. Balances power and simplicity - 80/20/2. This seems like a clear winner for the "20" category. It's not a dangerous feature, nor does it add a significant burden to the browser to support. (High User-performance)
4. Makes sense of the web - Focus on real human tasks and contexts. I really do use the stop and reload buttons separately, and am frustrated when I go to click a button and it changes. This is akin to on a mobile device where you start reaching your finger to tap something, and asynchronous search results pop up where you intended to click something which pushes it down.
And that add-on will allow a user to keep the Australis user interface (which I like other than this design bug) while moving the navigation buttons where we want them?


If it doesn't, then this needs to be fixed, not closed as WONTFIX.

(In reply to Dave Garrett from comment #33)
> This should probably be closed as WONTFIX at this point.
> 
> For those wanting detailed UI customization beyond what is built-in, we
> generally recommend using:
> https://addons.mozilla.org/en-US/firefox/addon/ClassicThemeRestorer/
> 
> That addon includes the ability to do this as well as many other advanced UI
> customizations.
I agree with the sentiment that it goes against the whole point of the Australis configurability and customization palette not to offer something so trivial as back, forward and reload buttons for the users to place as they wish.

I feel that Classic Theme Restorer is not an answer because it does so many things, some of which may not be desirable, and it goes against the spirit of Australis (see previous paragraph).  Also, some people (such as myself) may wish to minimize the number of extensions they install for fear of (A)bloat, (B)undesirable interactions (or breakage when upgrading to more recent versions of FF) and/or (C)security concerns.  Installing an extension for something large and complex is fine.  For something very trivial that used to be part of the core Firefox functionality, it is disturbing.

Now my answer to this was to make my own extension, one that is much more trivial than Classic Theme Restorer, and *only* adds the back, forward and reload buttons so I can place them where I want (and even though they duplicate those in the URL bar, but I don't find this kind of duplication to be problematic).  This solves the above-mentioned problems (A), (B) (to some extent) and (C) (as far as I'm concerned, because I trust myself, but of course it would not apply to other people).

I'm willing to publish this extension if there is some demand for it (although it would be a non-trivial task for me, because I just installed it on my system by creating a subdirectory of my Firefox profile, I'd have to learn how to write an extension package).  I'd be even more willing to convert it to a patch on Firefox itself if there is a reasonable hope that this patch might land.

In other words, if this can help close this bug, I'm willing to submit a patch.  (But then, if I was able to do it, it should be fairly trivial to the FF maintainers, so I'm not sure whether this is of any value.)
(In reply to Matt from comment #35)
> And that add-on will allow a user to keep the Australis user interface
> (which I like other than this design bug) while moving the navigation
> buttons where we want them?

Yeah, sure.

It appears that since the initial Australis release, the developer of that addon has added loads of extra customizations. It starts with a different tab style on install, but turning it off and going to curvy tabs is easy. It kinda has too many options, which might make you take a bit of time to find everything you want, but it does seem to work quite well and cover lots of things people want to customize beyond the default stuff.

(In reply to David A. Madore from comment #36)
There are other addons, but they don't seem to be maintained much anymore because everyone seems to have gravitated towards Classic Theme Restorer. (addons that aren't needed by lots of people, like built-in features that aren't needed by lots of people, also get dropped at some point)

https://addons.mozilla.org/en-US/firefox/search/?q=buttons
(In reply to David A. Madore from comment #36)
> (B)undesirable interactions (or breakage when upgrading to more
> recent versions of FF)

This is my largest concern. If the functionality is not considered a guarantee in the core UX/UI code, then the plugin could break in 12 months. Meanwhile the codebase has moved so far past support that it is too large of a task to reasonably request as a release-day regression, thus it's a feature request with all of the deeper justification that requires.

Today I use "Better Stop" to add back esc-stops-animations, because that was a genuine bug which I happened to love. It does one thing, and does it well.
So the answer to customizing the "easiest to customize" UI of Firefox in a small way that was trivial in the previous UI versions of the browser is to install an extension that adds the aforementioned bloat/maintenance/security/incompatibility concerns?

That's a rubbish solution, and I can't believe that that Firefox team hasn't fixed this bug in Australis yet.

Requiring add-ons to get the same level of customization as the previous out-of-the-box browser makes this, by definition, the most difficult to customize UI for Firefox.

(In reply to Dave Garrett from comment #37)
> (In reply to Matt from comment #35)
> > And that add-on will allow a user to keep the Australis user interface
> > (which I like other than this design bug) while moving the navigation
> > buttons where we want them?
> 
> Yeah, sure.
> 
> It appears that since the initial Australis release, the developer of that
> addon has added loads of extra customizations. It starts with a different
> tab style on install, but turning it off and going to curvy tabs is easy. It
> kinda has too many options, which might make you take a bit of time to find
> everything you want, but it does seem to work quite well and cover lots of
> things people want to customize beyond the default stuff.
> 
> (In reply to David A. Madore from comment #36)
> There are other addons, but they don't seem to be maintained much anymore
> because everyone seems to have gravitated towards Classic Theme Restorer.
> (addons that aren't needed by lots of people, like built-in features that
> aren't needed by lots of people, also get dropped at some point)
> 
> https://addons.mozilla.org/en-US/firefox/search/?q=buttons
(In reply to Matt from comment #39)
The entire reason Firefox was created is because the previous Mozilla Suite was overloaded with everyone's pet features. The explicit strategy was to create a more minimal browser and allow people to pick and choose their other features with addons. If for any feature, even minor ones, most people are completely fine with the default, then the correct course of action is to use an addon to customize it. You must remember, there are MANY little things like this that add up to a mountain of bloat if they all are expected to be built-in.

Arguing here is not productive. This is a minor issue that, quite frankly, very few people care about, and there is an addon developer who has provided a solution for this general topic that is better than what would ever be considered to be built-into Firefox.
(In reply to Dave Garrett from comment #40)
> The entire reason Firefox was created is because the previous Mozilla Suite
> was overloaded with everyone's pet features. The explicit strategy was to
> create a more minimal browser and allow people to pick and choose their
> other features with addons.

But this argument would also demand that Australis, as a whole, take the form of an add-on, because it is certainly not essential in a minimal browser.  Given that Australis *has* been integrated into Firefox, apparently with the goal of providing the user with more configurability, it seems extremely weird that this configurability does not include "back", "forward", and "reload" buttons, which have always been part of the browser palette and which are very basic features (especially compared to the "share" and "social" buttons).  What is the logic here?  Either users don't want to configure where there browser buttons go, in which case Australis makes no sense, or they do, in which case it makes no sense not to allow it for back, forward and reload.  I just don't get it.  Is there any empirical evidence that users want to move certain buttons but not others?

Conversely, the problem with redirecting users towards add-ons whenever they have a problem is that (A)it adds a lot of bloat to their browser (hence, slowness), (B)add-ons keep breaking or being disabled when Firefox releases new versions (hence, unhappiness), (B′)add-ons aren't maintained for long (more unhappiness; see the fiasco in bug #1052480), and (C)given that Firefox, unless FFOS, has no permission systems to speak of, it means surrendering control over the user's computer to the add-on author, which can be frightening (or should be).  Now in many cases, add-ons make sense, but one can't just claim that they're the panacea to any problem.  Evidently there must be some trade-off between things that go inside FF and things that are left do add-ons.  And if there is a trade-off, it makes sense to discuss how the trade-off should be decided.

> Arguing here is not productive.

I agree that arguing in a bug report is generally a poor idea, but what I don't know is: where would be the right place to discuss this instead?  Where is the right place to send feedback, and, more importantly, constructive criticism and logical arguments (such as the previous paragraph), about what should or should not be in Firefox?
(In reply to David A. Madore from comment #41)
> But this argument would also demand that Australis, as a whole, take the
> form of an add-on, because it is certainly not essential in a minimal
> browser.

Every browser has a base theme and minimal UI customizability. Australis is just the current version for Firefox.

> I just don't get it.  Is there any empirical
> evidence that users want to move certain buttons but not others?

Yes. Studies have shown that, by far, most users keep defaults and many users have moved certain things around in ways that prevent them from properly using their browser. Mozilla produced extensive heatmaps of all UI usage prior to designing Australis and based things heavily on how people actually interact with their browser.

The changes greatly improve the chances of a user not shooting themself in the foot through their desire to customize wrongly. I have had to deal with people who somehow had moved away their address bar but not other navigation buttons before and couldn't understand why none of the support given to them made sense. Yes, it's annoying that we have to make the standard for the lowest common denominator, but we nonetheless do.

> > Arguing here is not productive.
> 
> I agree that arguing in a bug report is generally a poor idea, but what I
> don't know is: where would be the right place to discuss this instead? 

To be blunt, the right place is this and similar bugs, but two years ago.

I too have lost arguments against UI changes (e.g. I think the RSS button could've stayed in the address bar), but years later, it's not something worth discussing further. There are addons to deal with things if desired, and just getting used to the new UI is often enough to stop caring. (e.g. I stopped using the RSS button and just use the subscribe menu in the bookmarks menu)
Flags: needinfo?(madhava)
(In reply to Dave Garrett from comment #40)
> (In reply to Matt from comment #39)
> The entire reason Firefox was created is because the previous Mozilla Suite
> was overloaded with everyone's pet features. The explicit strategy was to
> create a more minimal browser and allow people to pick and choose their
> other features with addons. If for any feature, even minor ones, most people
> are completely fine with the default, then the correct course of action is
> to use an addon to customize it. You must remember, there are MANY little
> things like this that add up to a mountain of bloat if they all are expected
> to be built-in.

This is a petty argument: Why not have the new communications application "Hello" be a addon? Netscape Conference comes back, but integrated. Why do we need cut/copy/paste buttons? It's a really weird line in the sand.
If this is the wrong time to be having this discussion for Australis, then I'll ask a more timely question.

How soon will a new UI for Firefox be released that actually increases the customization of the stock browser instead of decreasing it like Australis did?

(In reply to Dave Garrett from comment #42)
> (In reply to David A. Madore from comment #41)
> > But this argument would also demand that Australis, as a whole, take the
> > form of an add-on, because it is certainly not essential in a minimal
> > browser.
> 
> Every browser has a base theme and minimal UI customizability. Australis is
> just the current version for Firefox.
> 
> > I just don't get it.  Is there any empirical
> > evidence that users want to move certain buttons but not others?
> 
> Yes. Studies have shown that, by far, most users keep defaults and many
> users have moved certain things around in ways that prevent them from
> properly using their browser. Mozilla produced extensive heatmaps of all UI
> usage prior to designing Australis and based things heavily on how people
> actually interact with their browser.
> 
> The changes greatly improve the chances of a user not shooting themself in
> the foot through their desire to customize wrongly. I have had to deal with
> people who somehow had moved away their address bar but not other navigation
> buttons before and couldn't understand why none of the support given to them
> made sense. Yes, it's annoying that we have to make the standard for the
> lowest common denominator, but we nonetheless do.
> 
> > > Arguing here is not productive.
> > 
> > I agree that arguing in a bug report is generally a poor idea, but what I
> > don't know is: where would be the right place to discuss this instead? 
> 
> To be blunt, the right place is this and similar bugs, but two years ago.
> 
> I too have lost arguments against UI changes (e.g. I think the RSS button
> could've stayed in the address bar), but years later, it's not something
> worth discussing further. There are addons to deal with things if desired,
> and just getting used to the new UI is often enough to stop caring. (e.g. I
> stopped using the RSS button and just use the subscribe menu in the
> bookmarks menu)
(In reply to Dave Garrett from comment #42)
>
> > I just don't get it.  Is there any empirical
> > evidence that users want to move certain buttons but not others?
> 
> Yes. Studies have shown that, by far, most users keep defaults and many
> users have moved certain things around in ways that prevent them from
> properly using their browser. Mozilla produced extensive heatmaps of all UI
> usage prior to designing Australis and based things heavily on how people
> actually interact with their browser.
> 

Can you please provide a source for these studies?
(In reply to Dan Q (RedBlade7) from comment #45)
> Can you please provide a source for these studies?

https://blog.mozilla.org/ux/2012/06/firefox-heatmap-study-2012-results-are-in/

Mozilla telemetry, and generally all telemetry for all software, shows that most features are left at defaults for most things. This is not really a surprising thing, but understanding it lets you make better decisions on what areas to focus on.

I don't particularly have the time to go Googling for more ancient UI/UX research. (yeah, 2 years is ancient in this area :p )

(In reply to Kelly Kane from comment #43)
> Why not have the new communications application
> "Hello" be a addon?

Very much the counter-argument. Mozilla has been going back down this road with some major features. A balance needs to be made. I think the largest area where they've added niche features that previous strategy would leave out is the developer tools, but their stats showed that WAY more developers used them if they were built-in, even though installing an addon for it would be trivial. These things need to be taken on a case-by-case basis.

> It's a really weird line in the sand.

You really need to stop thinking in absolutes here. You have developers, and many volunteers, attempting to come up with the best combination of stuff possible for the majority of people. It's not going to please everyone by default.

(In reply to Matt from comment #44)
> How soon will a new UI for Firefox be released that actually increases the
> customization of the stock browser instead of decreasing it like Australis
> did?

You're probably not going to get what you want built-in. I know; shocking. You are perfectly capable of getting what you want via an addon that comes highly recommended. Please realize that this is a trivial issue with a simple resolution, and that posting walls of text in a long abandoned bug tracking a two year old feature overhaul is not going to get you anywhere.

To restate: General policy is against people WONTFIXing others' bugs, so I'm not going to close this personally. (if someone else wants to, be my guest) This feature suggestion, however, has clearly not been accepted. Please move on and just use addons for extra customizability.

I can only attempt to explain the situation to you so much, so I'm going to stop replying here further. Please understand that what I'm doing is just that, and I'm not personally determining policy here. Attempting to argue to me, or into the aether, will not get you anything.
What is the situation, or what is the "official" Mozilla policy (assuming this makes sense and there is one), concerning this bug in Firefox 57 and later?

As the earlier discussion shows, the answer to people demanding "back", "forward" and "stop"/"reload" buttons in Firefox has thus far been "there's an addon for that" (viz., "classic theme restorer").  But with Firefox 57 removing support for XUL+XPCOM addons, these addons have all stopped working.

I'm not sure whether it's even possible to write a WebExtension to add "back", "forward" and "stop"/"reload" buttons (I couldn't find WebExtension API equivalents to the XULBrowserWindow and nsIWebProgressListener that are used in the XUL extension I'm looking at; nor even exact equivalents of the back, forward, stop and reload actions; but I'm new to this and maybe I don't know where to look).

So what would be the best hope to get navigation buttons working on Firefox 57 and later?
Severity: normal → S3

The severity field for this bug is relatively low, S3. However, the bug has 6 duplicates and 14 votes.
:dao, could you consider increasing the bug severity?

For more information, please visit auto_nag documentation.

Flags: needinfo?(dao+bmo)

The last needinfo from me was triggered in error by recent activity on the bug. I'm clearing the needinfo since this is a very old bug and I don't know if it's still relevant.

Flags: needinfo?(dao+bmo)
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: