My feedback, after testing it for several days, is:
I think placing tabs on top is a very bad idea. Here's why:
1. The Tabs are more *related to the web page* than to the user
interface. Separating the tabs from the web pages feels awkward and
2. Placing Tabs on top of the UI seems to pretend that there could be
other UI within a given tab. I don't think this will be a common
scenario at all, and it has the unrealistic and undesirable hope that
the Firefox browser become a *replacement for the operating system*.
Nice dream - Bad idea.
3. The argument that Tabs on top is more logical hierarchically is only
true from one or a very few limited, abstract, and less relevant
perspectives. In reality, it *breaks the hierarchical logic* of
switching/manipulating/viewing tabs in relation to the web page (e.g.,
I'm looking and pointing at the web page and thinking "I'd like to
switch to another page", "what's the title of this page", "is this page
done loading?"). It forces the user to traverse (visually and
physically/mouseually) all the UI in order to interact with pages+Tabs.
4. It does not save any vertical space. (And vertical space saving
doesn't trump all else. What is it with that obsession anyhow?)
5. Having the UI inside the Tabs seems like you're lugging around all
the UI numerous times, for each Tab. It subjectively makes Firefox feel
bloated and sluggish.
6. In order to have a (questionable) benefit from Fitt's Law, the window
will have to be maximized (many don't, and it's silly on the ever more
common wide-screen monitors), and the page's title in the title menu
will need to eliminated (a really crappy idea). So, there are more
disadvantages than advantages.
7. I especially don't like Tabs-on-top on my *netbook* because of the
*extra distance I have to traverse with my track-pad* to change/close Tabs.
8. Having the Tabs on top feels awkward and seems wrong. Probably due to
the *aggregate* of all the reasons stated above.
BTW: Having the Menu Bar under the Tabs is ridiculous. And I am doubtful
the planned "Firefox Menu" will ever materialize, let alone completely
eliminate the need for the (hidden-by-default) Menu Bar.
So, are there any significant advantages? Not really.
1. There are the initial coolness and "wow new" factors. They will wear
off quickly - and be replaced with annoyance over the things listed above.
2. There may be a *few* pixels of vertical space saved. Woop-dee-doo.
3. The Fitt's Law benefit is limited to only full-screen windows and is
achieved over the loss of having the full and consistently placed title
in the title menu.
New is not always better. A little vertical space saved doesn't justify
making so much else worse.
I suggest to make Tabs-on-top an *option* with the *default being off*.
Users that are obsessed with minor vertical space saving (see all the
ugly tiny-and-ugly-and-userUnfriendly themes floating about) are
typically advanced users (with hawk eyes), and they will know how to
switch to Tabs-on-top. Users who prefer *looks over usability* will also
have the option to switch. Regular users OTOH would be forced to suffer
the consequences of poor usability.
PS. Preemptive request: Please spare us the old and tired "you just need
to get used to it" and "some people just don't like change". I have been
using trunk builds since 10+ years. I love change. I yearn for change.
But I also recognize bling-bling crap when I see it.
I'd love to hear other pros and cons...
Bugs I think should be fixed ASAP:
Church of the Flying Spaghetti Monster: http://www.venganza.org/
Anthropogenic Global Warming skepsis: http://tinyurl.com/AGW-Skepsis
> Tabs-on-Top has been available in nightly builds for several days now (via "View / Toolbars / Tabs on top"). This thread is intended to gather feedback on people's impressions, and to discuss the pros and cons of placing the Tabs above the UI.
I think that's a great idea. I'd like to suggest some ground rules:
1. Please do NOT reply to someone else's list of pros and cons in order to try and prove a pro or con "wrong" - this just causes us to rathole on arguments.
2. Please DO feel free to repeat pros and cons that others have forwarded; build your list, even if it's repeating something that's been said!
3. Please DO feel free to forward observations from watching people interact with tabs in various places. Don't use words like "everyone I know" or "only the nerds" unless you've confirmed that those are actually the samples you're referring to, but do let us know if you've seen people run into big problems so we can figure out if there are better designs.
4. Please DO feel free to categorize some of the feedback as how it makes you feel. Emotions in design are important.
I'll submit my own pros and cons a little later today. Please, please, please adhere to rules 1 and 2!
> 1. Please do NOT reply to someone else's list of pros and cons in order to try and prove a pro or con "wrong" - this just causes us to rathole on arguments.
(Basically, don't quote-back someone else's list of pros and cons - build your own!)
It's very logical and makes perfect sense, because the tab always
switches what's below it, and the URLbar and back/forward buttons
definitely belong to the page. In fact, I proposed exactly that when
tabs were introduced:
I only miss the menu bar on the screenshots.
Personally I find it hard to even experiment with tabs-on-top right now
because it looks really strange visually. The tab color doesn't blend in to
the toolbar color.
Benjamin points out something important: the rest of the UI hasn't
finished being updated yet, so it's to be expected that it's not as
comfortable or polished as we expect it to be. When evaluating the
initial phase of the implementation, we should be sure to keep that in
> Tabs-on-Top has been available in nightly builds for several days now
> (via "View / Toolbars / Tabs on top"). This thread is intended to gather
> feedback on people's impressions, and to discuss the pros and cons of
> placing the Tabs above the UI.
(Using WinXP + WinXP Standard (non-classic) skin)
"Tabs on top" is great. But, when I enable it, the menu bar goes between
the tabs and the address bar, which looks really weird. If I disable the
menu bar and re-enable it, then the menu bar will be above the tabs, which
looks awesome. In other words, tabs directly above the navigation bar
looks great. Anything between them does not.
Also if I keep the menu bar disabled, things look even better and there's
more vertical space. But, there's no drop-down anywhere to get to
preferences and it's not very clear how to get the menu bar back.
Right-clicking on the navigation bar fixes that. But, it looks like
there's a way to disable all toolbars and not be able to get them back.
Well, there's the alt key, but that's not very newbie-friendly.
Of course, I understand that things are not finished.
In short though, "Tabs on top" (when completely finished) will be awesome!
BTW: I've hidden the Menu Bar during my testing (via "View / Toolbars /
Menu Bar"). Having the Menu Bar turned on (and below the Tabs) makes
Tabs-on-top even worse. But that is irrelevant *if* the Mozilla
developers manage (beyond likelihood) to make the Firefox Menu not suck.
That point is valid, but minor. Please note that none of my long list of
"disadvantages" are a result of the tabs' colors being different from
the toolbar's (I didn't even notice it until Benjamin just pointed it out).
I wonder how valuable a mere opinion is (without giving any rationale).
That's like deciding who will be president based on only 100 votes (out
of a population of 400,000,000).
I'm not just talking about colour; a lot of the UI pieces aren't
finished, including those related to how the tabs appear, how we
handle menus, etc.
(When you say things like "if the people doing this work can manage to
not suck, which I highly doubt", I'm pretty tempted to killfile you
and encourage others to do the same. If your points are strong, or
even if you just *think* your points are strong, you shouldn't need to
resort to ad hominems or other rudeness to make them. It's not how I
think we should behave in this community. And, more tactically, if
you are looking to have your points taken into consideration by
people, you would probably do best to not make them unavoidably
defensive by insulting them in the process. If you just want to
"score points" and rant, then I would prefer that you just get a
LiveJournal account and write there.)
> On Fri. 09.04.2010 16:16, Michael A. Puls II wrote:
>> "Tabs on top" is great.
> I wonder how valuable a mere opinion is (without giving any rationale). That's like deciding who will be president based on only 100 votes (out of a population of 400,000,000).
I don't know what makes you think that this decision will be made by a poll of this newsgroup.
Please respect the ground rules I set out for this conversation. It's the only way that it will end up being of any value to the user experience team and product drivers who will end up making this decision.
> That point is valid, but minor. Please note that none of my long list of "disadvantages" are a result of the tabs' colors being different from the toolbar's (I didn't even notice it until Benjamin just pointed it out).
Please allow everyone to make their own points without discrediting them. Disagreement is allowed and valid, but I don't think we profit from trying to convince everyone or reach a consensus. Respect the fact that those making the decisions are reviewing this all as input, with a critical mind.
If some people feel like they're unable to make a pros and cons list at this time because the design is in flux and incomplete, they should feel free to say so.
I don't. I was trying to avoid the thread drowning in posts with "it
sucks" and "it's great" (without supporting rationale).
> Please respect the ground rules I set out for this conversation. It's
> the only way that it will end up being of any value to the user
> experience team and product drivers who will end up making this
I'm not sure which ground rule you are referring to. The post I
responded to had no "pros and cons", so rule 1 doesn't apply. The poster
didn't say how the feedback made him feel, so rule 4 doesn't apply
(unless "it" in rule 4 didn't refer to "feedback" but to tabs-on-top
(bad grammar); and "it is great" is somehow useful feedback on how one
feels about tabs-on-top. Do you really want a bunch of posts with
nothing more than "it's great", it rocks", "love it", "hate it", ad
nauseam? If that's the kind of feedback you meant and want, then OK).
I'm still wondering how valuable a mere opinion is (especially since I
do not think "this decision will be made by a poll of this newsgroup").
Michael gave much more than a mere opinion. He talked about how the current interaction with the menuBar is odd, and how removing it requires us to think carefully about the discoverability of the functions therein. Some of that work is already in flight with the new Firefox button, but it's still valuable information.
I found a good amount of value in his post, and would rather you not dissuade others from posting their thoughts without fear of you telling them that they are of questionable value.
Thanks for being respectful of the quality of the discussion.
Mike: You have misquoted an important part of my quote! I didn't say
what you said I said. Not even "like" it. I said "if the Mozilla
developers manage (beyond likelihood) to *make* the Firefox Menu not
suck" (not that *they* can manage not to suck!). Also, you are taking
something personally when it was not intended at all. I meant that it
will be unlikely to get the Firefox Menu to not suck because the
*subject itself* is very difficult to do in a non-sucking way -
regardless of who tries to fix it.
OTOH, you implying that I all I want to do is "score points and rant" is
an ad hominem attack - and particularly painful after I spent
considerable effort and time to formulate a list of rationales
concerning an import part of the UI. It's amazing how many people with
stones don't realize they live in glass houses...
If you're getting value out of it, then that's great!
And strategically, it's better to be too lenient than too strict on
what's on and what's off topic.
con: there is no visual dividing line between the bookmarks toolbar and the
page content any more. When using a white persona (currently using "Spring
Blossoms Again"), it just blends in completely, which is disconcerting.
And Michael is very right about the menu bar: at least in the current
version it should be on top, and the trick of disabling/reenabling it make
me much happier with tabs-on-top.
In my current configuration:
The bookmarks toolbar seems very out-of-place. It feels like the navigation
bar should be right up against the page content, and the bookmarks should
either be up somewhere, or even down near the status bar. I don't know
exactly why I feel this way.
I apologized for the misinterpretation; I should have been giving more
benefit of the doubt here, clearly.
> OTOH, you implying that I all I want to do is "score points and rant" is an
> ad hominem attack - and particularly painful after I spent considerable
> effort and time to formulate a list of rationales concerning an import part
> of the UI. It's amazing how many people with stones don't realize they live
> in glass houses...
I didn't mean to imply that it was you, and edited that chunk a bunch
of times, apparently poorly indeed. (I was in fact typing that you
had misinterpreted *my* post, when I went back and read it. :-( )
My mail as written was unproductive, and unfair. That it wasn't what
I meant it to say is not an excuse, but perhaps a reasonable basis for
I do indeed appreciate that you took the time to give your feedback,
and that I sometimes find your tone provocative is my problem and not
All gone. No problem!
PS. I do want to "provoke" - in the senses of: to stir up, call forth
(feelings, desires, or activity), to stimulate to action, to bring
Anyway, I'll offer what I see as some of the pros and cons:
1. Placing the tabs on top of the URL bar, and other UI that deal w/ the
page is exactly where they always should have been! Some may argue that
its bad to separate it from the page, but I would disagree and say that
all its doing is including *more* in the page. It drives home the point
that the URL bar, refresh, home, etc. apply to the current tab you're
working in, instead of being disconnected UI elements.
2. It really didn't take any getting used to at all, it seemed a natural
to have the tab encompass the extra UI elements.
3. There is more vertical space (especially when hiding the Menu Bar),
although vertical scrolling doesn't bother me a ton, it is /nice/ to
have more space for the actual *content*.
4. Removing the Menu Bar and reworking the UI is long over due. The
"File Menu" is obsolete (and doesn't really fit in a web browser, but
that's not my point). I'm relatively confident everything from the Menu
Bar could be fit into a variety of other places (not entirely in a
Firefox menu, but over a variety of places). I agree that "change for
change" is not good and that "new is not always better", but there is
much more going on then just saving a little vertical space saving!
1. I really don't see any disadvantages, there may be an initial shock
of users to having their tabs moved to a different place, but I feel
since Chrome has been doing it and it won't feel entirely out of place
(for people that have ever used Chrome on someone else's computer/tried
it out). Note that I'm *not* saying Firefox should feature parity with
Chrome or that tabs should be moved on top because of Chrome. I'm just
saying many users may have seen a /similar/ UI before allowing the
transition to go smoother.
2. The UI is kind of funky with the Menu Bar on top still, I may try
1. I don't care about any extra time (I believe this is all the stuff
about Fitt's Law in the first post? First time I've heard of that.) it
may take to travel to the tab bar with my mouse -- I change tabs
exclusively via CTRL+TAB on the keyboard. (And with my dual monitors my
mouse cursor usually has way more distance to travel just to get back to
the Firefox window that an extra 40 pixels of vertical space doesn't
matter.) Understandably, most "casual" users will use a mouse to point
to each tab, but personally it doesn't affect me.
2. The Bookmark Toolbar and how it interacts with tabs. (Although if
there is no more bookmarks menu...a drop down next to the tabs list the
brings up the bookmarks menu might be good -- it connects it to the tabs
[which is what the menu acts on] and still has it easily accessibly
instead in some giant Firefox menu.)
Overall, I'm quite happy with the changes that are going on and if
anything it will have me suggest Firefox even more to my friends,
relatives, acquaintances, etc. I'll reserve my total judgment until
more changes land (the Firefox menu and Aero integration to be
specific), but like where the UI team is driving Firefox.
Thanks for all the great work!
Windows 7, 1440x900, menubar hidden accessible with alt key, windows
usually maximized, bookmark toolbar hidden but accessible with a toolbar
1. tab title is closer to where my 15 years of browser usage tells me to
look for the page title. There's some small joy in returning to the
habit of looking near the top of the window for the page titles. Totally
subjective, and probably only for a few old codgers like me, but it just
2. fewer accidental bookmarking (dropping on bm toolbar) or popping out
new window (dropping on content area) when I meant to reorder a tab.
3. fewer accidental mix-ups between tab overflow button and bookmark
toolbar overflow button. even the tiny cognitive hiccup of trying to
pick the right overflow button is gone and that's nice.
4. completely subjective, but in combination with hidden menubar and
hidden bookmarks toolbar, the tabs just look better to me on the top.
Except for small padding, gradient, and other theme issues that I'm sure
will be cleaned up, no cons so far.
It seems to me that the titlebar should remain just that: a place where
you can see the title of the page you are reading. Putting tabs there
would obscure the title of the current tab. If people do not like the
amount of vertical space taken up, they can get different themes. There
are compact themes which make the various bars very thin.
Why are people complaining so much about the menu bar? I would hate to
have it hidden. I use the menu bar all the time. It is the most
convenient place to access the bookmarks, after all.
Why would you want the stuff in the menu bar, which is now organized and
easy to find, scattered all over the place? I like the menu bar because
it is easy to find things there. The interface is straightforward with
a menu bar in plain sight. Scatter all the settings all over the place,
and you are likely to discourage new users (and some old users as well).
Perhaps you expect people to memorize a huge list of keyboard shortcuts?
Please observe the ground rules I set out at the start of this thread, and allow people to state their pros and cons without them being questioned or challenged. We're looking for broad feedback.
If you have questions of clarification, feel free to ask. These seem like disagreements, though, not requests for clarification information.
-Its sort of logical if we look at the UI as: what belongs to the Tab is
-I do like still having the Title. Now the title on top logically does not
make sense, but it is very useful. (So what are the purists going to say
-Title on top has a special logic of its own to it - it can be long and
shares the window title with the close/minimise buttons. Its the best
candidate to have there
-Chrome does not have a title bar - that seems daft to me - and I cannot
turn it on either!
-Tabs next to title bar is good.
-Like most people I am used to the other way.
-It is a bit further to move the mouse.
-If we look at it the UI that: the further up stuff is stuff that I use
less often (rather than what belongs to what), then tabs on top is probably
-Tabs immediately above the actual web page is visually useful when looking
to see which tab I am on
I really don't think you should get very metaphysical about what should be
where - it comes also to ergonomics, how far the mouse has to move, and
where the eye is used to falling, and to good design - which is not always
strictly logical but is usually elegant.
And - its a good idea to cater for the main preferred flavours.
I think the questions to consider are more:
With main stuff that is more just to look at it should be visually very easy
to spot. Tabs sitting just above the web page makes sense as I am often
looking at the combination to see "Which tab am I currently on?". Title
bar is also very useful for this.
1 - Is the "Tabs on Top" option going to stay? Or is it just for testing?
2 - Whats the future for the Title bar? Is it still going to be there, or
at least as a settable option?
I reckon you have to keep the "Tabs on Top" setting in.
If you are going to leave the option to tick "Tabs on Top" in the menu then
most of the argument is irrelevant - people can set it the way they like.
This is well in accord with the Firefox ethos of being customizable. Will
save an awful lot of arguments about what people are used to and are
expecting. If the option stays (my vote) then the only argument is really
over which option is the better to have as the default installation (And I
reckon going by what I have seen here even that could get heated).
So I cut the quote down to what I think you're replying to?
Anyway, I'm not asking users to memorize a lot of keyboard short cuts.
I use a handful of shortcuts, but I certainly do not know and use all of
them. So I still definitely want them to be available. Your reply
implies that I want them to be randomly distributed throughout the UI:
of course I wouldn't want that. I'm just saying its about time the UI
What I was kind of getting at is that there are a 7 menus in the menubar:
Parts of File, parts of View, Tools and Help deal with the application
as a whole and I would like to see them in a "Firefox" button of some sort.
Bookmarks and History I would like to see combined into the tab bar in
some way, yes I know this would take away some of the horizontal space
of the tab bar, but as the tab bar is now the "outermost" (I can't think
of a better word, I hope you know what I mean by this?) UI that
encompasses pages, I feel this would be the appropriate place for this.
I.e. a quick idea of what could be done, right now we have a place to
add a tab, there could be a "Bookmarks +" that opens a drop down list
and lets you open a new page from a bookmark (or even better...perhaps a
Bookmarks tab that when clicked opens the bookmarks manager in a tab, on
double clicking a bookmark opens it in a new tab). I think History is
also associated with the pages themselves (as opposed to the application).
And honestly...I'm not sure what I'd do with Edit (I actually hide my
Edit menu via a Style), it all deals with the pages themselves so it
/shouldn't/ be in the "Firefox" menu (in my mind), it probably could be
just in the right click menu.
Sure it might take some getting used, but I think if there are locations
you go in the UI for particular tasks it would save time. So instead of
looking through menus to find the options I want you know that if it has
to do with pages its in location x, if it has to do with the application
as a whole its in the "Firefox" menu.
I mean a lot of people didn't like the drastic overhaul of Office 2007,
but personally I was all for it.
I hope that explains a bit more, if not feel free to ask more questions!
Being on a mac, I'm not really affected by the discussion itself, I
guess. Hope, looking at it briefly.
What I want to say, though, is: I like the thread. The rules you set
out, as hard to follow for us, make a thread like this one actually an
interesting read. Congrats to that.
Yes. I find it very odd to have anything about the menu bar.
It would also help with another problem I run into: the increased mouse
way I have to travel to switch between tabs. (I open tabs for almost
*every* link I click on, so I do that a *lot*)
Because it would make sense :). The urlbar applies directly to the page,
the bookmarks bar less so. So, it makes sense that you "feel" the urlbar
should be directly on top of the page.
The bookmarks bar is more like a wider bookmarks menu, and is in a way a
menu itself. So, it makes sense that it's above the urlbar and
underneath the menu bar.
I personally agree with those that find tabs on top a good thing, especially
the hierarchical logic and vertical space gain are the points I find more
appropriate, but I suspect this list is biased over geeks and lacks of
John Bird wrote:
> -Like most people I am used to the other way.
> -It is a bit further to move the mouse.
> -If we look at it the UI that: the further up stuff is stuff that I use
> less often (rather than what belongs to what), then tabs on top is probably
> a nuisance
I think these Cons especially apply to newbies, and I think we should really
take care of what newbies feel about their experience while browsing. Habits
are the most difficult things to change, and newbies don't see a good reason for
the change in this case. That's why I would make Tabs on top an option
disabled by default (and for me, as a geek, it will be the first option I'll
> On Fri. 09.04.2010 16:16, Michael A. Puls II wrote:
>> "Tabs on top" is great.
> I wonder how valuable a mere opinion is (without giving any rationale).
> That's like deciding who will be president based on only 100 votes (out
> of a population of 400,000,000).
Well, if you need even more:
It's great because it makes Firefox more visually appealing.
I also consider the address of a page to be a very important part of the
page. When the tabs are not on top, the focus of the address field feels
very very detached and in the wrong place. With tabs on top, the address
field feels like part of the tab and is where I naturally move my mouse.
It's the difference between a lone address field widget that changes to
match the tab you have focused and each tab having its own address field.
The latter is much more appealing and I'm more productive with it because
it matches where I instinctively move my mouse (and my eyes).
With tabs up top, it's like a weight is lifted off my shoulders.
But, surely, 100/400000000 is only a first stage of seeing how much people
like the new feature to see if it's even worth introducing to a larger
number. Then, it just comes down to whether it's well-accepted or not.
* Having the navigation toolbar more associated with the page content
seems more natural (note: you get this with tabs on the side too). I
think content-related doorhanger notifications would feel more natural
* Similarly, having the tab management close to the window management
controls (titlebar, minimize, maximize, close) feels surprisingly good
* Less distance to move mouse to URL bar, bookmarks bar (see
* With a lot of tabs, it moves the busy area away from the content -
which makes it less distracting. Although a busy bookmarks bar can take
away from this a bit.
* Further to move mouse to manage tabs.
* On Windows, the all-tabs dropdown and new-tab button are extremely
close to the window controls (specifically, the close window button) -
making a mis-click potentially quite destructive.
* Takes some getting used to (but this can be said of any change).
* It partially covers up the pretty birdies in the persona I use - there
would have been a clear space for this previously.
* This makes me want to experiment with having the bookmarks toolbar
above the navigation bar.
You can experiment with this easily, using userChrome.css and
- What kind of pros and cons could really affect this reasoning?
- Can we try harder to make tabs-on-bottom look sleek and modern?
> What irritates me about the discussion is that, based on a discussion
> with Alex in bug 544815, we already seem to have accepted that tabs on
> top aren't necessarily more logical or more ergonomic, and therefore we
> would avoid doing it by default on Win XP, OS X and Linux. On Win 7 we'd
I don't think that a decision has been made about the default location of tabs, and in fact comment 28 of that bug is where Alex suggests that discussion about the potential change be taken to the newsgroups.
Discussions about potential UI change are frustrating, challenging and often lead to bickering. I think the ground rules have helped us avoid the bickering, for what it's worth, and the feedback is helpful. It is *not* the sole criteria by which this decision will be made, just like feedback on the location of closebuttons on tabs was not the sole criteria that was used to make that decision.
> - Can we try harder to make tabs-on-bottom look sleek and modern?
I think this is an excellent question, but it presumes that "sleek and modern" is the sole advantage to tabs on top, and as the pros and cons of others illustrate, there are more advantages to be had. Disadvantages, too.
For people who are curious, my mind isn't made up at this point, and neither is the UX team's. I think many of us lean in different directions. Getting additional people's impressions of what's good and bad about the designs (again - without bickering or preventing people from holding their own opinions) really helps us.
The problem with consistency is that sometimes the best solution in
isolation isn't necessarily the best solution overall. For instance, it's
somewhat ridiculous for a Web browser to use the command structure of a 90s
word processor, but on Windows XP pretty much every application uses that
interface, so it probably does more harm to present the user with a Firefox
button instead of the traditional menu bar.
So this causes any interface discussion (tabs on top, firefox button, etc.)
to basically have two stages:
1) is the interface better
2) given what the user is going to expect, is it still a net positive
I believe that tabs on top passes point 1 (conceptually pure, allows for
chromeless application tabs, allows for tab notifications that are
associated with the site's identity). I also believe that giving current
users an interface that is different (albeit better) might be too
destructive on XP, Linux, and OS X for long time Firefox users (were are
seeing some of the complaints in this thread). However, it would be
incorrect to conclude from my views on point 2 that tabs on top doesn't pass
Instead of a long list of detailed pros and cons, I would like to offer a
simple thought experiment. To test the innate intuitiveness of Firefox, we
create a clone of one of the world's most intelligent people (you can pick
your own example, perhaps even yourself!) We then raise this clone in a
controlled environment, giving them access to the types of things people
encounter in the real world, but limit their access to computers to a few
very basic operations (closing a window, perhaps one demo of back/forward
navigation with a tabless window or iPhone). One day we present the hyper
intelligent test subject with an image of Firefox with several tabs open,
and we ask a few questions:
-if you were to hit the back button here, what do you think would happen
-this text in the long white bar, what does it relate to or describe?
-if you hit this X here, what do you expect to disappear?
I believe the test subject would get most of these questions wrong with our
current interface, and every question right if the visual grouping actually
matched the real behavior of Firefox. (third question is only wrong when
shown OS X: "the tab X closes the browser chrome and leaves the content area
and other tabs").
Furthermore, let's consider this data visualization that Asa created:
That is why I changed my thinking on what we should ship on new operating
systems, the growth curve of people who are new to computers, and new to the
Web is so large that the "hyper intelligent test subject in a controlled
environment being exposed to Firefox for the first time" evaluation isn't
happening just in a thought experiment, but hundreds of millions of times
(perhaps minus the hyper intelligent part, but that was just to give them
the best possible chance in the experiment). There are of course new users
showing up on older operating systems as well, so it's debatable where
exactly we might want to draw the line (although in general I think we
should design Firefox's default behavior to a particular platform, as
opposed to "if there is a new user profile" checks).
In general my thinking is that in situations where everything is new (the OS
interface and a large part of the user population) we really don't have much
to evaluate in terms of consistency, and can just focus on creating the best
interface in isolation.
On Sat, Apr 10, 2010 at 5:33 AM, Mike Beltzner <belt...@mozilla.com> wrote:
> On 2010-04-10, at 6:06 AM, Dao wrote:
> > What irritates me about the discussion is that, based on a discussion
> > with Alex in bug 544815, we already seem to have accepted that tabs on
> > top aren't necessarily more logical or more ergonomic, and therefore we
> > would avoid doing it by default on Win XP, OS X and Linux. On Win 7 we'd
> I don't think that a decision has been made about the default location of
> tabs, and in fact comment 28 of that bug is where Alex suggests that
> discussion about the potential change be taken to the newsgroups.
> Discussions about potential UI change are frustrating, challenging and
> often lead to bickering. I think the ground rules have helped us avoid the
> bickering, for what it's worth, and the feedback is helpful. It is *not* the
> sole criteria by which this decision will be made, just like feedback on the
> location of closebuttons on tabs was not the sole criteria that was used to
> make that decision.
> > - Can we try harder to make tabs-on-bottom look sleek and modern?
> I think this is an excellent question, but it presumes that "sleek and
> modern" is the sole advantage to tabs on top, and as the pros and cons of
> others illustrate, there are more advantages to be had. Disadvantages, too.
> For people who are curious, my mind isn't made up at this point, and
> neither is the UX team's. I think many of us lean in different directions.
> Getting additional people's impressions of what's good and bad about the
> designs (again - without bickering or preventing people from holding their
> own opinions) really helps us.
> dev-apps-firefox mailing list
I (respectfully) disagree with that reasoning. "We" are not a
representative sample. And AFAIK, in statistics, non-representative
samples are shunned as useless.
> "To the scientist, however, representative sampling is the only
> justified procedure for choosing individual objects for use as the
> basis of generalization, and is therefore usually the only acceptable
> basis for ascertaining truth."
So what "we" like should not be a large determining factor in what is
best for Firefox (lest we become as insignificant as Linux distros -
total devs' wants control).
What we should be doing IMO is use our advanced knowledge and our
reasoning abilities to analyze which features make sense.
PS. I liked your rational evaluation of the pros, as you see them, of
...and above the tabs?
I don't believe that tabs-on-bottom or tabs-on-top should either of
them be primary goals. They should be things we choose in service of
another goal, that of a great user experience. Trying extra hard to
make one of them great seems like misapplied effort, if we could make
the other choice great more easily. (I think we've really tried quite
a bit to make tabs-on-bottom modernized, but that's not to say that
you might not have more ideas that are worth experimenting.)
Further, I don't think that we should be trying to *convince* people
of anything in this discussion; the best favour we can do to our users
is to state our impressions of tabs-on-top, because that's the one we
need the most data on. If you wanted to start a thread collecting
feedback on tabs-on-bottom, I think that would OK, though I must
confess my belief that we have heard pretty much everything there is
to say about tabs-on-bottom in the last years of the product.
It's important to note that it's like that *for you*.
Visual appeal is a matter of taste. One thing of making such a
discussion fair is to make sure you don't insult other by telling them
their taste is wrong (which you do when you imply yours is right).
I haven't tried it yet myself, for one thing because I'm not using
Firefox in daily work (but I'd like to see the option - as a mere
*option* - in SeaMonkey as well), for the other because I'd need time to
play with it and surely won't cheer for changing the processes I'm used
to right away (I am very bound by my habits).
In any case, I've always seen tabs as an excuse for multiple windows not
working well enough in today's window managers (no matter what different
OSes call them). And so, having the tabs be a more general
application-level thing sounds pretty logical. Also the possibility of
changing toolbar contents or display of those at all dependent on what
tab is selected makes some good sense to me.
That experiment is surely interesting, I'll watch it closely (even if it
doesn't match my habits right now).
Yes, I would probably try to arrange it like this:
that probably looks better and also reduces the mouse way from page to tabs.
But I don't have a strong opinion on it, bookmarks bar might also go
underneath the tabs, above urlbar.
It would be great is someone(TM) (Alex F.?) could make a mock-up of
those two orders (always with the Menu Bar hidden) and put them up on
the wiki page:
My suspicion is that it might look odd having the bookmarks bar at the
very top. We'll see...
I fear that if the bookmarks bar is shown (which I prefer for it's
enormous convenience), tabs-on-top might look bad in every arrangement.
That depends on the styling. If it looks like the menu bar, why not?
(At least, people would start to actually notice the menu bar and maybe
use it, assuming they use the bookmarks bar. :) )
Placing the bookmarks bar on top is visually difficult since there are
usually two z-order levels in each design, the background window level
(glass, blue in XP Luna), and the foreground tab level (normal browser
chrome). So choices include placing the bookmarks bar on the background
level (text on glass is hard to pull off without at least adding a lot of
frosted area), or just giving up on the notion of having a window level and
a tab level and draw the bookmarks bar as normal browser chrome, except
above the tabs.
Generally, yes, except perhaps in cases where emotions or opinions are
likely to run high -- it can be good in such cases to reduce the scope
of the discussion a bit to help stay on point.
Just my two cents. (I haven't tried the nightlies, and am not a
usability expert, so have no real opinion yet. The screenshots look
pretty, though. :-))
Nathan Tuggy [:tuggyne]
Sorry, but I have no idea what you're talking about.
Just make it look like the menu, as I said.
> Yes, I would probably try to arrange it like this:
> bookmarks bar
In my case, my navigation, menu and url bar, are customized to be on the
same bar. I use "Compact Menu" to reduce the normal menu down to the
single blue button to the left of my URL bar. This works fine, as the menu
bar items don't get nearly enough use to justify the room they normally
occupy. I dumped the search bar, as I never used it either. So, my
combined nav/menu/url bar looks like this:
The overly large text is because I have low vision.
Having my tab bar immediately above or immediately below my nav/menu/url
bar would make little difference to me, though logic suggests that
navigation is per tab, and should probably be under each tab. The main
criteria I have is that the tabs, and their related navigation be very
close to each other in the UI.
The bookmark bar seems out of place as it currently is. I believe that
should be above the navigation and tabs, as it has little to do with what
is in the tabs, or with the navigation within those tabs.
Either of these two arrangements would be fine for me, but I have no strong
preference between the two.
If I were not using Compact Menu to make the menu into a button, I'd likely
prefer the menu bar to be above the bookmarks bar and below the title bar.
I'm actually surprised that something like Compact Menu wasn't built in
from the start, as it is the number one space-saving extension I can think
Ideally, tabs above or below the nav bar will be an option, and all bars
will be fully customizable, no matter what changes are made. Logical or
not, many of us have differing preferred setups - some of which might be
partly dictated by visual or other disabilities.
FoxWolfie / CubCoon
Here's how it looks like, on my system (Linux, GTK with ThinIce theme):
I think it looks good - just make the icons of the bookmarks toolbar
smaller. No idea about other systems, but it must be possible to make it
look like a menu there, too.
I don't use the bookmarks toolbar, and I like it enough that I'm going
to run with this.
In case anybody wants it (patch is trivial):
Hmm, that's surely true in some way, but then, without wanting to
critize you, shouldn't we first think about how it affects usability and
only then find a design to match the usability?
I think defining something like the order of the toolbar stack by some
design choices we have in mind right now is potentially harming good
ideas for usability improvements.
I'm not saying that Ben's idea necessarily results in better usability,
but I think it's worth a try as much as tabs-on-top itself, and we
should only think very hard about how to make design really slick once
we have figured out what matches the usability we want to have best.
In terms of usability, I think a case could be made that placing the
bookmarks toolbar items on the browser level would make it appear as if they
either act on all tabs, or open in a new tab, while placing them at the tab
level makes it more obvious that they act on the current tab (which is in
fact the case). Either way the bookmarks toolbar isn't as clear cut as
something like Back or Reload, which obviously acts on the active tab.
earlier requested mockups forthcoming :)
These are of course all stephen's visual design work (TM?), I just moved
some sections around and labeled stuff. Anyway, you can see that having
the bookmarks bar created out of glass is a little problematic with the text
(purely a visual thing, not interactive). This could be solved by removing
transparency to that area and sort of having three z-order levels.
As for which of these two options makes the most conceptual sense, really
you could make arguments either way. Earlier I said that since the
bookmarks navigate the content area so they act on the tab level (unlike say
an identity button that logs you in and out of the browser, or the Window
The inverse argument is that the bookmarks are different destinations from
the tabs, and especially if you consider that they have favicons, they don't
really belong in the visual hierarchy under the active tab's favicon.
Additionally, the bookmarks toolbar does not change based on what tab is
selected (like the location bar does). If the tab is producing a
notification, it is also more clear which site is "speaking" if the
bookmarks are placed at the browser level instead of the tab level (since
otherwise there are sort of sites all over the place).
I'm personally split on which position makes the most sense for the
bookmarks toolbar, but I think I'm leaning towards browser level being more
On Sat, Apr 10, 2010 at 7:27 PM, Alex Faaborg <faa...@mozilla.com> wrote:
> Right, I totally agree. I was just referring to the comment "that depends
> on the styling."
> In terms of usability, I think a case could be made that placing the
> bookmarks toolbar items on the browser level would make it appear as if they
> either act on all tabs, or open in a new tab, while placing them at the tab
> level makes it more obvious that they act on the current tab (which is in
> fact the case). Either way the bookmarks toolbar isn't as clear cut as
> something like Back or Reload, which obviously acts on the active tab.
> earlier requested mockups forthcoming :)
> On Sat, Apr 10, 2010 at 6:57 PM, Robert Kaiser <ka...@kairo.at> wrote:
* Bookmarks toolbar on top of tabs feels very wrong. I think because it
create a mis-match of what's content-related and what's app-related -
bookmarks feel more content-related (for usage related to the toolbar,
* I think I prefer having the bookmarks toolbar above the navigation bar
(but this isn't a strong feeling). Below feels a lot more familiar, but
on top moves clutter away from content, and the navigation bar feels
more directly related to content (so benefits from being next to content).
And I was just replying to "it might look odd" from Peter Lairo. So, if
you quote me, please quote me correctly with sufficient context, please.
I realized you can do that with userChrome.css without any patches to
source or installation:
Ops, sorry, that was not supposed to go out here.
(Very nice thinking an explanation. That's basically what I meant with
> Furthermore, let's consider this data visualization that Asa created:
> the growth curve of people who are new to computers
While true, the graph is deceptive: It suggests that the new users are
Firefox users. Draw Firefox at the bottom and MSIE on top and it looks
very different and you draw entirely different conclusions - that your
users are those longer-time computer users and want consistency. Firefox
is used by all kinds of users, from novice to geeks to highly
intelligent, but in tendency, I think users in growing markets just use
Windows XP with MSIE, see e.g. China. So, be careful with such conclusions.
> users are those longer-time computer users and want consistency. Firefox
> is used by all kinds of users, from novice to geeks to highly
> intelligent, but in tendency, I think users in growing markets just use
> Windows XP with MSIE, see e.g. China. So, be careful with such conclusions.
It doesn't matter much how you split it; if you're designing to be
relevant in a few years, you had better take that new user growth into
consideration (and the PC upgrade cycle as well.)
It might also be helpful to remind yourself every so often that there
are more people who are new to the Web since Firefox 0.8 shipped than
were ever online before 0.8. That is, more than half the people online
today have no experience of a pre-Firefox Web. (Further, more than 10%
of all the people online today arrived after Chrome launched.)
With 100M new arrivals to the Web each year, and another 200M people
upgrading to new PCs each year, it's be silly to rigidly bind ourselves
to the past.