FF 3.5 Clear Recent History is not an improvement
John J. Barton
Clear Recent History. I think it is the wrong direction.
The 3.5 version provides a small UI and hides more information. Why?
That is exactly the wrong direction. The UI give the strong impression
that the control clears browsing history, but it does a lot of other
things as well. In 3.0 all of the things were listed; in 3.5 none are
listed and none are explained.
In addition the 3.5 control is significantly more difficult to use. To
see the hidden information you have to drop down a list, navigate a
scroll bar on the left while you click boxes on the right, all in a tiny
box that will not expand. The drop down serves no purpose: there is
plenty of screen to show the info.
Reverting to the 3.0 control would be a big improvement. Presumably the
3.5 design was trying to solve some problem. There has to be a better
solution.
jjb
belt...@mozilla.com
available in nightlies, and is now closed to major changes, such as
any that would require string alterations.
While I understand your concerns, I do not think they apply to the
common user who just wants to be able to clear the history Firefox has
recenty collected. There are some corner cases (authenticated http
sessions, site specific page annotations) where we are trying to do
the right thing by default without overloading the dialog with
meaningless choices. I think we get these mostly right.
The small listbox size is something we might be able to fix, and
should consider. The design is meant to highlight the most common
history items, hiding the others from view. Not sure if it's working,
though, as the common user is unlikely to want to investigate that
drop-down, and it is obviously frustrating to the advanced user.
cheers,
mike
On 2009-05-05, at 12:20 PM, "John J. Barton" <johnj...@johnjbarton.com
> wrote:
> _______________________________________________
> dev-apps-firefox mailing list
> dev-apps...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-apps-firefox
Alex Faaborg
Why does the user have to scroll to see the full list of items?
-We are planning on fixing this, and it doesn't involve any strings so
the change isn't disruptive. The original design had a reduced height
to accommodate a large timeline control of the time range of history
being cleared, but that was removed (with plans for us to add it
back), so height isn't an issue any more.
Why does the user have to expand the details of items being cleared?
-The goal here is to obfuscate and encapsulate a variety of
implementation centric options so that mainstream users don't have to
wonder about arcane terms like "cache" or "authenticated sessions,"
and can focus more on their goal (selecting a specific time range and
obliterating everything).
Isn't this harder for users who need to regularly clear a specific
type of information?
-It is, but the interface is designed for users with a primary goal of
being focused on time, as opposed to the type of thing they need to
clear. For instance, some developers might need to regularly clear
something like authenticated sessions, but overall this type of
repetitive action isn't the intended primary use of the dialog.
Why does the interface talk about "history" when it deals with more
than just the list of URLs you have visited?
-Across the application we are standardizing our terminology so that
history means "the full set of things implicitly collected by Firefox
as you browse the Web." For the small subset of users that are
extremely familiar with strict definition of history as solely a list
of URLs, this is a bit of an adjustment, but the change has some
advantages. This allows us to say things like "in private browsing
mode history is not collected" or "Clear Recent History is like
private browsing mode, but in reverse." Also the privacy prefpane has
been simplified down to a top level choice of remember history or
don't remember history (with the addition of "use custom options" for
people who are wondering what history includes).
Cheers,
-Alex
On May 5, 2009, at 9:33 AM, belt...@mozilla.com wrote:
> Thanks for your input. The design was discussed here and in the bug,
> available in nightlies, and is now closed to major changes, such as
> any that would require string alterations.
>
> While I understand your concerns, I do not think they apply to the
> common user who just wants to be able to clear the history Firefox
> has recenty collected. There are some corner cases (authenticated
> http sessions, site specific page annotations) where we are trying
> to do the right thing by default without overloading the dialog with
> meaningless choices. I think we get these mostly right.
>
> The small listbox size is something we might be able to fix, and
> should consider. The design is meant to highlight the most common
> history items, hiding the others from view. Not sure if it's
> working, though, as the common user is unlikely to want to
> investigate that drop-down, and it is obviously frustrating to the
> advanced user.
>
> cheers,
> mike
>
> On 2009-05-05, at 12:20 PM, "John J. Barton" <johnj...@johnjbarton.com
> > wrote:
>
Chris Hofmann
<snip>
> Why does the interface talk about "history" when it deals with more
> than just the list of URLs you have visited?
> -Across the application we are standardizing our terminology so that
> history means "the full set of things implicitly collected by Firefox
> as you browse the Web." For the small subset of users that are
> extremely familiar with strict definition of history as solely a list
> of URLs, this is a bit of an adjustment, but the change has some
> advantages.
The idea of simpification is a great one and we should pursue
relentlessly. That is a core reason that set Firefox on a path to
becoming a better browser. It would be interest to know how we arrived
at the idea that its a small subset of current users that understand the
new defintition of history. In some cases I'm sure we will have to
apply intuition, but in this case there are a few indicators that would
seem to buck the conventional wisdom that "browser cache" is a mis
understood term.
google search results for "browser cache" shows 695,000 links, but
"browser history" show only 462,000 links.
http://www.google.com/search?q=%22browser+cache%22&btnG=Search
http://www.google.com/search?q=%22browser+history%22&btnG=Search
deeper investigation of numbers like this might indicate that a lot of
people actually have written, and others may have actually read and know
more deeply about a seemingly obscure term like "browser cache". Are
there other sources of data like this that we could use to test the
level of user understanding about these terms? Do we have surveys and
other studies that might confirm or validate the understanding of the
terms in use? Is it worthwhile looking deeper at this?
> This allows us to say things like "in private browsing mode history is
> not collected" or "Clear Recent History is like private browsing mode,
> but in reverse." Also the privacy prefpane has been simplified down
> to a top level choice of remember history or don't remember history
> (with the addition of "use custom options" for people who are
> wondering what history includes).
Again, this kind of simpification is great, but you have pointed out the
confusion it presents to those that have learned a dfferent definition
of "history." What if we considered a new term for this new
diffinition of history instead of hijacking the old term. Maybe
"history" plus some modifier. "Complete History", "Comprehensive
History" or something along those lines. Its too late to add that
extra modifier for 3.5, but we should watch the reaction to figure out
if it should be something to consider for the next release.
-chofmann
John J. Barton
>...
> Why does the user have to expand the details of items being cleared?
> -The goal here is to obfuscate and encapsulate a variety of
> implementation centric options so that mainstream users don't have to
> wonder about arcane terms like "cache" or "authenticated sessions," and
> can focus more on their goal (selecting a specific time range and
> obliterating everything).
But that is not what will happen. According to the design, the control
will select a specific time range an obliterate only those things with
check boxes. That you can't see.
...
>
> Why does the interface talk about "history" when it deals with more than
> just the list of URLs you have visited?
> -Across the application we are standardizing our terminology so that
> history means "the full set of things implicitly collected by Firefox as
> you browse the Web."
So why not put that in the UI so folks know?
Time Range that will be cleared
The kinds of history that will cleared:
[] ... (recommended) <learn more>
[] ... (not recommended) <learn more>
jjb
Chris Hofmann
> Chris Hofmann wrote:
>> The idea of simpification is a great one and we should pursue
>> relentlessly.
> all together. That will be simpler.
>
sure, that extreme of removing all the funtionality would make things
simple, but not useful.
> Shouldn't the goal should be to provide users with a powerful tool
> that they can learn to use as quickly and easily as possible? Hiding
> complexity because we cannot figure out how to help users understand
> it does not create a powerful, easy to learn tool or for that matter a
> simpler tool. It creates a tool with hidden complexity.
>
as you took my suggestion to extreme, taking your suggestion here to
extreme might mean moving towards enumerating every about:config pref in
UI and trying to make each easier to use. That would be powerful, but
I'm also hoping you'll agree that tons of UI to deal with hundreds or
thousands of options creates unwanted complexity. Thats what was at
the core of the initial firefox design. There were a lot of sessions
to throw stuff overboard from the mozilla suite and historical artifacts
from early Netscape browsers, and I'd say that that effort struck a good
balance over time. I'd still argue that process has continued value.
> The sub-functions for Clear History (BTW what is the difference
> between Recent History and History?) are complex and technical no
> doubt. To me the solution is to explain the consequences of clearing
> these sub functions rather than putting user interface barriers in the
> way.
>
> A simple and effective way to help users is to add a Learn More link
> to each item that points to wikipedia for that item. We can assert
> that wikipedia has mastered the art of explaining to user all manner
> of complex issues. We can participate by adding information but let
> users speak in their own words about the subject. And if wikipedia's
> explanation is not adequate, users will do what users do in these
> cases: retreat or boldly push some button and discover for themselves.
> It's ok.
>
rather than using wikipedia I'd suggest using support.mozilla.org. it
has the same kind of community envolvement, is more specific to firefox
topics, and we can look at lots of extra data that helps us to
understand use patterns and understanding of features.
-chofmann
> jjb
Alex Faaborg
> the level of user understanding about these terms?
We could deploy a mechanical turk study to test comprehension of each
of these topics (would only take a moment), but overall I'm not sure
the results would really impact the overall design decision that
encapsulation and obfuscation of implementation centric details is
generally a good thing. For instance, the http error code 404 has
entered mainstream vernacular, but it's still bad form to show it to
users.
> What if we considered a new term for this new diffinition of history
> instead of hijacking the old term.
Something to consider is that in some cases the concept of history as
a more global thing isn't just baked into terminology, but also baked
into interactions in the interface. For instance, if you are viewing
your history in the library window, and you delete a site, we actually
clear all [1] of the sub-components of history (cache, cookies,
authenticated sessions, etc.) as part of that deletion from "history."
-Alex
[1] well, should be all, I think we are currently clearing some of the
sub-components but not others
On May 5, 2009, at 10:45 AM, Chris Hofmann wrote:
> Alex Faaborg wrote:,
>
> <snip>
>> Why does the interface talk about "history" when it deals with more
>> than just the list of URLs you have visited?
>> -Across the application we are standardizing our terminology so
>> that history means "the full set of things implicitly collected by
>> Firefox as you browse the Web." For the small subset of users that
>> are extremely familiar with strict definition of history as solely
>> a list of URLs, this is a bit of an adjustment, but the change has
>> some advantages.
> The idea of simpification is a great one and we should pursue
Mike Beltzner
> Something to consider is that in some cases the concept of history
> as a more global thing isn't just baked into terminology, but also
> baked into interactions in the interface. For instance, if you are
> viewing your history in the library window, and you delete a site,
> we actually clear all [1] of the sub-components of history (cache,
> cookies, authenticated sessions, etc.) as part of that deletion from
> "history."
>
> -Alex
>
> [1] well, should be all, I think we are currently clearing some of
> the sub-components but not others
Point of fact: we do that only if the user asks to "Forget" the site.
Deleting a single entry deletes only that visit record from the table.
cheers,
mike
Alex Faaborg
John J Barton
> John J. Barton wrote:
>> Chris Hofmann wrote:
>>> The idea of simpification is a great one and we should pursue
>>> relentlessly.
>> Sorry I disagree completely. If this is your goal, remove the function
>> all together. That will be simpler.
>>
> sure, that extreme of removing all the funtionality would make things
> simple, but not useful.
>
>> Shouldn't the goal should be to provide users with a powerful tool
>> that they can learn to use as quickly and easily as possible? Hiding
>> complexity because we cannot figure out how to help users understand
>> it does not create a powerful, easy to learn tool or for that matter a
>> simpler tool. It creates a tool with hidden complexity.
>>
> as you took my suggestion to extreme, taking your suggestion here to
> extreme might mean moving towards enumerating every about:config pref in
> UI and trying to make each easier to use. That would be powerful, but
> I'm also hoping you'll agree that tons of UI to deal with hundreds or
> thousands of options creates unwanted complexity.
Yes I would agree that having options creates complexity which is why I
suggested that removing options increases simplicity.
But given a set of options how shall the user be allowed to use them?
Hiding them behind UI barriers because you think they are too complex
does not make them simpler, it just hides them.
I think a reasonable solution is to remove the sub-options altogether.
Then Clear History just has a confirm/cancel button. "Do you want to
clear all forms of history (includes pages visited, ,,,)"?
jjb
Chris Hofmann
>> Are there other sources of data like this that we could use to test
>> the level of user understanding about these terms?
>
> We could deploy a mechanical turk study to test comprehension of each
> of these topics (would only take a moment), but overall I'm not sure
> the results would really impact the overall design decision that
> encapsulation and obfuscation of implementation centric details is
> generally a good thing. For instance, the http error code 404 has
> entered mainstream vernacular, but it's still bad form to show it to
> users.
>
codes is bad form and in the case of 404's sites should replace with a
better explaination, 'guess' the correct web page, or offer search to
help get the user to where they want to go.
It doesn't seem like that is what we are doing in this case. In the
case it seems like we are not trying to rewrite a single term to provide
a more clear understanding, but we are combining and redefining and
hijacking existing terms. I agree that that line is fuzzy here; but it
seems like when we combine and redefine terms it might be worth the
extra study.
I've been talking to djst about ways we can really figure out how much
people know about the browsers, the web, and the internet, with the
idea that more study of that might help us to provide better support.
Maybe we can "browser history" as one additional area to figure out what
people really do, and don't know, about the several (and growing list)
of browsing history components.
we should also think about ways to propogate this new definition of
"browser history" across important sites.
Do the SUMO definitions of browser history match up with the new
combined definition, or will they add confusion?
http://support.mozilla.com/tiki-searchresults.php?locale=en-US&q=browser+history&sa=
-chofmann
John Bird
To better balance the simple folk who just want to clear stuff, and the
power user tinkerers who want to see the details I suggest:
Its a trivial change to change the Details button to enlarge the window
somewhat to show all the options that will be cleared. The details buttons
already does expand the window, just not enough to show all the checkboxes,
so just stretch it a bit more so the scroll bar then is not needed.
I reckon if you do that you can make everyone happy - and thats not a bad
thing!
John Bird
Robert Kaiser
> The 3.5 version provides a small UI and hides more information. Why?
Because Firefox users are dumbed-down creatures, 1337 people are
using... erm, no, I don't want to get into that again ;-)
In fact, it's true that the vast majority of Firefox users don't
understand the details of all the items displayed there and the new UI
tries to take that into account and make them not bother with stuff they
don't understand. Sure, that will keep them from learning about all
those details, but that's not _necessarily_ bad.
The Firefox philosophy is apparently to make things as easy as possible
for an as large as possible group of people and leave more detailed and
geeky control knobs to add-ons and to other Mozilla-based browsers that
target more advanced users (like SeaMonkey). And I think that's great
because there's a good "market" for all of those.
Robert Kaiser
Peter Lairo
> Its a trivial change to change the Details button to enlarge the window
> somewhat to show all the options that will be cleared. The details
> buttons already does expand the window, just not enough to show all the
> checkboxes, so just stretch it a bit more so the scroll bar then is not
> needed.
I fully agree. It seems silly to have the window just a tiny bit too
small. It's annoying on several levels (e.g., small scroll view-ports
are more disorienting, forced to remember hidden choices).
--
Regards,
Peter Lairo
The browser you can trust: www.Firefox.com
Reclaim Your Inbox: www.GetThunderbird.com
Islam: http://www.jihadwatch.org/islam101/
Israel: http://www.jewishvirtuallibrary.org/jsource/myths2/
Church of the Flying Spaghetti Monster: http://www.venganza.org/
"So, why don't we ever talk about the sun's contribution to global
warming? Well, because we can't regulate it, tax it, or make it feel
guilty for what it's doing" (www.WhatYouOughtToKnow.com)
Drew Willcoxon
Bug 489958 - New Clear recent history dialog shows unnecessary
scrollbars with details expanded
https://bugzilla.mozilla.org/show_bug.cgi?id=491638
Bug 491638 - Clear recent history dialog should persist details expansion
Drew
Alex Faaborg
> small.
yep, no one disagrees on that point, everyone agrees, and we need to
change the size.