Breaking of Core FF Functionality
Jon. L.
fact that about:config no longer allows you to change the
"browser.tabs.tabMinWidth" with any effect to the UI is incredibly
thoughtless--this has an immediate usability impact for a feature that
has been around for as long as I can remember and more or less sets in
stone the massive tabs and scrolling behavior that I find highly
undesirable (as do many power users, I'm sure).
Now, when reading through bug reports (https://bugzilla.mozilla.org/
show_bug.cgi?id=574654) it seems that something changed and this
functionality was moved from being configurable in the front-end to
only being configurable in the userChrome.css file. This is
unacceptably opaque--end users aren't aware of this file and the
configurations required to implement this functionality. Further, a
casual search of the web yields solutions that actually break the UI,
leaving holes when tabs are deleted, for example. In fact, even a
cursory Google search yields many results where people are asking
about how to fix this functionality.
So, if there were truly changes that made this functionality break,
then fine, but a simple add-on fixes the issue. This seems like an
acceptable solution, except that it hardly seems reasonable, fair, or
smart to require users to go through the hoops of installing a very
limited functionality add-on, that may or may not be supported in the
future. In fact, the very existence of such an add-on makes it
incomprehensible why this configuration cannot be done in the main
U.I.--surely if a user could implement this functionality the
application should be able to.
Honestly, as Chrome begins to catch up, and IE releases a new, much
better version, do you really want end users frustrated immediately
upon upgrading when they are most likely to be willing to spend the
time and energy to explore these other options?
Fix this simple issue and don't let the end user desires get lost in
whatever highly technical, highly irrelevant reasons the core
development team has for eliminating this simple functionality.
-A New Chrome User
Dao
> So, if there were truly changes that made this functionality break,
> then fine, but a simple add-on fixes the issue. This seems like an
> acceptable solution, except that it hardly seems reasonable, fair, or
> smart to require users to go through the hoops of installing a very
> limited functionality add-on, that may or may not be supported in the
> future. In fact, the very existence of such an add-on makes it
> incomprehensible why this configuration cannot be done in the main
> U.I.--surely if a user could implement this functionality the
> application should be able to.
Sure, stock Firefox could do the same. This is true for any add-on out
there. The reason we usually don't do it is that we don't want the code
and/or UI overhead in stock Firefox.
Jon. L.
mozilla-central/rev/181a0b76dc4a), the code changes don't change
anything in the rendering, they just no longer have a place in the UI
for altering them anymore. So. I'm not at all sure what bloat you're
referring to. In fact, the "UI Overhead" seems to only be
"about:config" and that still exists. In fact, the attributes that
used to set the tab widths are still there to, they just have no
effect.
Honestly, removing this functionality under the banner of "Remove
browser.tabs.tabMinWidth and browser.tabs.tabMaxWidth in favor of
CSS" (title of the bug report) and having the first post being "Users
can override this using userChrome.css if they absolutely want it. I
don't think the prefs are worth it." is a big middle finger in the
face of end users.
Moreover, claiming this change isn't a big deal is simply ignoring
users. Fine, if that's the route you want to take, but just because
the devs don't want it in there doesn't make it reality. In fact,
looking on the Web shows that some users are pretty displeased with
the direction FF is going in, but that this change actually degrades a
core functionality.
In any event, users like myself are first finding this atrocious
decision in the new version and, likely, it's too late to retain the
users that will switch because this changes simply breaks their
browser, plain and simple. If you're lucky, some will just downgrade,
but my suspicion is that most will move to Chrome, which doesn't have
this stupid behavior and the 140 pixel tabs that make it impossible to
work with more than 10 open tabs.
I'm sure the <10 lines and however many bytes of overhead this saves
are completely worth the tradeoff, in thousands of users, though.
Dao
> Well, unless I'm reading this incorrectly (http://hg.mozilla.org/
> mozilla-central/rev/181a0b76dc4a), the code changes don't change
> anything in the rendering, they just no longer have a place in the UI
> for altering them anymore.
You're reading it incorrectly. The patch didn't just remove the hidden
pref, it simplified the code handling with minimum and maximum tab widths.
Millwood
documentation. I can easily search the about:config document on mozilla
on the hunch that something might be customizable. But I have no idea
how to do that for chrome stuff. The tabs issue raised here is a good
example.
If there is proper documentation, please point me to it.
Jon. L.
you're talking about. Further, if the patch is the extent of the
change, then spending time worrying about this sort of minutia shows
how unfocused the devs are.
In short, the devs can build the browser they want or the browser that
users want. Seems like the choice has already been made and, with all
the condescending language on the bug report ("you can stop spamming
the bug now" and devs deciding what users should and should not be
okay with doing) it shows a deeply negative and disturbing trend. Is
it any wonder there were 3 million less downloads on day 1 for FF4?
Honestly, it's unbelievable that a simple change like this has devs
fighting tooth and nail against a change that one could find hundreds
of instances of complaints across the web (most of these dating very
far back, so this are the beta and pre-release FF users who are
technical enough to figure things out and find the place to complain--
there are probably hundreds if not thousands of users for every one of
these). Message received: "Users, shut up."
This is exactly like the debate I used to see where zealots would
argue that Thunderbird shouldn't support top posting at all and should
force the user to post their response BELOW the prior email--
thankfully, that didn't get implemented, but it showed how utterly
tone deaf to users devs could be.
It's simple--fix these sorts of things, or watch the user base errode
further.
Shawn Wilsher
> I fail to see how the source of the number affects the simplifcation
> you're talking about. Further, if the patch is the extent of the
> change, then spending time worrying about this sort of minutia shows
> how unfocused the devs are.
worry about a large and obscure set of preferences and how they interact
with every patch, we end up focusing our attention all over the place.
For each preference that is included there are at least two possible
behaviors we have to worry about which results in a minimum of 2^N
number of possibilities a reviewer has to worry about (N being the
number of preferences). The bigger N is, the bigger the problem.
> In short, the devs can build the browser they want or the browser that
> users want. Seems like the choice has already been made and, with all
> the condescending language on the bug report ("you can stop spamming
> the bug now" and devs deciding what users should and should not be
> okay with doing) it shows a deeply negative and disturbing trend. Is
> it any wonder there were 3 million less downloads on day 1 for FF4?
There's a tradeoff here which you are just glazing over. We can't build
the browser that *every* user wants, which means that some things that
some people want simply won't make the cut.
> Honestly, it's unbelievable that a simple change like this has devs
> fighting tooth and nail against a change that one could find hundreds
> of instances of complaints across the web (most of these dating very
> far back, so this are the beta and pre-release FF users who are
> technical enough to figure things out and find the place to complain--
> there are probably hundreds if not thousands of users for every one of
> these). Message received: "Users, shut up."
Again, see above regarding 2^N number of permutations that we then have
to worry about.
With all this being said, it shouldn't be too difficult for someone to
write an add-on to do what you want. Perhaps you should invest your
efforts in recruiting someone to do that instead of telling us we don't
know how to do our jobs.
Cheers,
Shawn
Dao
> With all this being said, it shouldn't be too difficult for someone to
> write an add-on to do what you want.
True: https://addons.mozilla.org/en-US/firefox/addon/custom-tab-width/
Jon. L.
> On 3/23/2011 10:21 AM, Jon. L. wrote:> I fail to see how the source of the number affects the simplifcation
> > you're talking about. Further, if the patch is the extent of the
> > change, then spending time worrying about this sort of minutia shows
> > how unfocused the devs are.
>
> It actually demonstrates the opposite; if we are constantly having to
> worry about a large and obscure set of preferences and how they interact
> with every patch, we end up focusing our attention all over the place.
> For each preference that is included there are at least two possible
> behaviors we have to worry about which results in a minimum of 2^N
> number of possibilities a reviewer has to worry about (N being the
> number of preferences). The bigger N is, the bigger the problem.
Completely missing the point. When you break the UI, it's a large
impact on the user experience. And, not a single person has yet
explained why THIS particular preference needed to be altered. The
first line in the bugzilla entry that changed this behavior was,
"Users can override this using userChrome.css if they absolutely want
it. I don't think the prefs are worth it." Think about what that means
for a second... a dev decided that for HIM it was completely
unnecessary to have that there. No user study, no survey, no analysis
at all, just "I don't think this functionality is necessary, so I'm
removing it." In and of itself, that's disturbing ...
Feel free to remove the preferences for default link color, blinking
text, "bookmark_added_static_root" or any other number of user prefs
that users truly do not use or that would actually make sense in a CSS
file.
> > In short, the devs can build the browser they want or the browser that
> > users want. Seems like the choice has already been made and, with all
> > the condescending language on the bug report ("you can stop spamming
> > the bug now" and devs deciding what users should and should not be
> > okay with doing) it shows a deeply negative and disturbing trend. Is
> > it any wonder there were 3 million less downloads on day 1 for FF4?
>
> There's a tradeoff here which you are just glazing over. We can't build
> the browser that *every* user wants, which means that some things that
> some people want simply won't make the cut.
Okay, so how about we start with not breaking the UI/UX? And, again, I
see no reason that this functionality (which wasn't removed, but
merely pushed down to an even more base layer of the browser that uses
an even more opaque method of changing it) had to get cut versus any
other number of preferences. And, correct me if I'm wrong, but doesn't
the code for the browser and UI take these parameters in as inputs
anyway? Don't you scale tabs according to content within a certain
range currently? All that you stopped is users easily defining these
values. flying in the face of what other browsers are doing might I
add.
Further, what benefit is it to have this same configuration option
require editing text files, cutting and pasting CSS code from the web,
and restarting the browser to reload a hard to find configuration
file? Don't the devs want to abstract away the details of the tab
animations and display? And what happens when this functionality is
removed or the behavior of these entries in the userChrome.css file
changed? It all breaks all over again and then even fewer people will
understand how to fix it. This is why these options need a frontend,
to abstract the low-level details away from the end users.
This is so obvious, I'm amazed I have to keep restating it. What's
next? Finding the address bar and back button are gone? The Web is a
strongly connected component after all, starting at some well-placed
start page should mean you can get to wherever you need to go.
> > Honestly, it's unbelievable that a simple change like this has devs
> > fighting tooth and nail against a change that one could find hundreds
> > of instances of complaints across the web (most of these dating very
> > far back, so this are the beta and pre-release FF users who are
> > technical enough to figure things out and find the place to complain--
> > there are probably hundreds if not thousands of users for every one of
> > these). Message received: "Users, shut up."
>
> Again, see above regarding 2^N number of permutations that we then have
> to worry about.
>
> With all this being said, it shouldn't be too difficult for someone to
> write an add-on to do what you want. Perhaps you should invest your
> efforts in recruiting someone to do that instead of telling us we don't
> know how to do our jobs.
Your jobs? From the Mozilla.org site, it definitely seems like you
don't know your job ... to wit "Mozilla is a global non-profit
dedicated to putting you in control of your online experience." Or how
about from the Manifesto, "Transparent community-based processes
promote participation, accountability, and trust."
After all these exchanges, the open hostility towards users, and the
defiance from the devs, I haven't found much evidence that any of the
devs have taken the mission of MOzilla to heart at all, or even know
it. Maybe if your job was to code, then fine, but it's actually more
complex than that and it involves delivering a product.
I suppose we could just have add-ons that do everything and you could
just deliver a rendering engine and nothing else? I bet it would be
coded very efficiently.
> Cheers,
>
> Shawn
Annoyingly, I suspect all this time is wasted on trying to explain
what's wrong with the decisions and the decision making process. I
have a dual degree in computer science and electrical engineering,
worked on building analytics and trading systems for years, and have
dealt with building products that were very end-user focused. I often
had to deal with the much more monolithic IT departments, and once
they stopped caring about what the end users cared about, it became a
useless exercise to deal with them. Like Is aid above, all my
interactions with the devs on this issue has shown me that this is a
product being built, users be damned.
Good luck on shipping the next few versions of Firefox--after being a
user of Firefox and predecessor products for over a decade, you've
finally given me reason enough to go elsewhere.
Asa Dotzler
> Good luck on shipping the next few versions of Firefox--after being a
> user of Firefox and predecessor products for over a decade, you've
> finally given me reason enough to go elsewhere.
Thanks for the best wishes and I hope you enjoy the other browsers.
- A
ff4-test
> to worry about.
>
> With all this being said, it shouldn't be too difficult for someone to
> write an add-on to do what you want. Perhaps you should invest your
> efforts in recruiting someone to do that instead of telling us we don't
> know how to do our jobs.
>
> Cheers,
>
> Shawn
So you'd rather have everybody have a gazillion add-ons installed for
everything under the sun to make it easier for you to test rather than
simply not remove stuff that is already there and people use? Testing
is never the most fun part of coding, but it's necessary. (Unless you
work at M$ - from their track record, it appears that they don't
believe in software testing.) It sounds to me like you don't need
anybody to tell you that you don't know how to do your jobs - that's
apparent. Paying attention to what the people using your software
want is the number one job of any software engineer; my brother was
fired last year for ignoring what the person he was writing the
program for wanted and telling her what she wanted instead, much the
same as the FF devs are doing here and on the bug report.
Here's a CC of what I just posted on the bugzilla bug for those
interested. How many loyal, long-term users do you have to lose
before you start paying attention to what people want?
I created a temporary e-mail account and registered for the sole
reason of commenting on this defect in Firefox 4.
Seriously, what is wrong with the developers? I have two college
degrees, both in programming, and have been a Mozilla/Firefox user
since Mozilla 0.7. I'd estimate that between 800 and 900 people have
switched to Firefox over the years because of me. The lack of
consideration for the users is getting worse instead of better - you
guys are approaching M$ in terms of how you view/treat users! A
feature is added in one version, then jerked away like Lucy pulling
the football away from Charlie Brown. Worse than that even - to my
knowledge Lucy never didn't pull the football away, whereas the
Mozilla team adds features, leaves them in for a few versions so
people will get used to them, and then yanks them away without warning
or even semi-logical reasons.
Regardless of what you may believe, people do NOT enjoy this.
Comments like "I don't think the prefs are worth it" and "stop
spamming this bug now" made by a developer turned up on the bugzilla
item when searching for it only make it worse, as the attitude toward
users and what they want is more that of an inconvenience instead of
somebody to make happy. (IMO, you've only begun to see the beginning
of people complaining about this... it was a dumb idea and most
anybody who has need to find the bug is going to think pretty much the
same thing.) I don't know what went on internally, but it sounds like
one single person didn't use this feature and therefore removed it on
his own personal whim. (Or more accurately, shoved it somewhere that
97+% of the population will not be able to access/edit.) I never
changed the default behavior in FF3, but the defaults were changed in
FF4 and then the user is required to jump through hoops to get it to
work the way it was. To be blunt, this design decision is freakin
retarded...
Something very similar happened a few years ago. Print Preview was
available as an icon in the toolbar and is something I use every
single day. Then in one version, without warning, whoosh - gone!
It's annoying enough to install an add-on for every box I put Firefox
on, but guess what? Then I get the fun of dealing with it during
EVERY significant upgrade to Firefox. Updated to Firefox 4 today and
big surprise, the Print/Print Preview add-on isn't compatible.
Thankfully, somebody (not the original author of the add-on) patched
it to work with FF4. Print Preview is already a menu item - would it
*really* be that much trouble to have it be an available icon like it
used to be???
Another absolutely fantastic design change in FF4 is the password
remembering options. (Noticed this when registering.) Instead of one
click to remember/not now/never in a temporary drop-down bar at the
top of the screen or even in a modal dialog box as in FF2, if you want
anything except remember, you have to click three times instead of
once. WTF sense does that make?!?!??? My second-year UI design class
had stuff like that as examples of BAD UI design - our grade on a
project would get dinged if we turned in a program with an interface
like that - and yet you actually expended the effort to change things
to INTENTIONALLY make it work like that! Isn't there ANY sort of
review process for changes like this???
I see that there is now an add-on to restore functionality that was
added in a previous version and then pulled away, just like Lucy and
that football. This is yet one more step when setting up FF on a
machine and more hassle during every update for the foreseeable
future. Truly a WONDERFUL experience to be certain...
Being lean really isn't a valid reason because having the 742,398 add-
ons loaded to do what used to exist natively almost certainly takes
more memory than if they were built-in (like they used to be). Not
wanting to test is just plain lazy - almost worse than the arrogance
of a single developer removing whatever he feels like for no apparent
reason. I could sell you a vehicle that would be excellent on fuel
and only cost $150. Available add-ons include an engine, steering
wheel, front wheels, electricity, heater, brakes, speedometer,
transmission, etc, etc, etc. Yup - you get the vehicle pictured at
http://waterfordhackney.com/Images/Funny-Taxi-Cab.jpg I can foresee
the time when Firefox is the same way - you get a rendering engine and
then require add-ons to have such 'extra features' like a menu bar,
address bar, tabs, back/forward buttons, the ability to print, etc.
FF would certainly be lean and easy to test, but would also be utterly
useless. (OH NO - I think I just gave them ideas!!! :( ;) What's
truly pathetic is that I really honestly thought of removing this for
fear one of the developers will decide adopt such a design model...)
I'm going back to FF3 as soon as I get done posting this (which is
really too bad since there are some neat features I think I'd like, at
least for however long they last). I will not upgrade any of the 11
other machines I use on a regular basis at home/work and will advise
everybody both friends and co-workers to use FF3 and explicitly
specify NOT 4 because they broke it. I'm also going to start looking
at other browsers. Incidentally, this is the first time I'll be doing
evaluating browsers in over 10 years. I sincerely hope that this fact
fills developers who make decisions like this (the ones which are best
classified as 'F-in dumb') with a feeling of great pride - you've
managed to make a loyal user and advocate of Firefox for what amounts
to forever in the computer industry jump ship. It's not the change
itself that's doing this, it's the underlying attitude it represents.
Firefox is heading down the same path as Micro$oft, only you don't
have enough of a monopolistic stranglehold on the market to survive
people getting POed at your product and using something else.
Hopefully you guys will gain some wisdom and go back to what made
people use FF in the first place (features, user friendliness,
customizability, security, stability) before a product I've used and
loved for many years dies a slow painful death. Somehow I doubt this
will happen... after all, why listen to what people like/don't like/
want? People have only been complaining about it for the 8 or so
months since it was introduced during the beta and the change still
made it into the final.