Firefox 2 Alpha Ramblings

Firefox 2 Alpha, branded Bon Echo, was just released. As usual with Firefox releases, it was announced on a number of big websites before it was officially ready. I downloaded it a day early too, but I knew what I was getting. This is an Alpha release meaning it is for testing, not for most users. It is not near Beta quality as previous Firefox Alpha releases have been so be prepared if you try it.

I know it is an alpha, but the new stuff is a mess and the old stuff isn’t improved much if any. If Mozilla is hoping to get the final 2.0 out before or around the same time IE7 and Vista are released I fear it is going to be pretty rough and not going to stand up to IE7’s advances.

I read a comment that compared this to a version 1.6 Alpha rather than 2.0 Alpha because of the few improvements over 1.5. He was optimistic that by the final release it would be worthy of a .5 version increase. The currently planned features would do it for me, but are they going to be done in time or will features be thrown out to make the release date?

So far it appears to use less memory than FF 1.5 but I haven’t used it heavily. If it does, that is a really good thing since that was a frequent complaint about 1.5. It still uses a lot, but at least it doesn’t go wild eating up whatever free memory it can get.

I thought part of the point of moving Bookmarks and history into an SQLite database with Places was so having a lot of Bookmarks wouldn’t slow down the Bookmarks menu. It is still takes several seconds to display the menu after I click on Bookmarks just as FF 1.5. I have a lot of bookmarks, over 2000 bookmarks which is still more than 600 KB after removing embedded favicons and some other meta data.

But so far Places is very rough, it currently doesn’t even have as much functionality as the old Bookmark system they are replacing. But supposedly it only landed on this branch two weeks ago. For two weeks work it seems pretty good, but that isn’t how things work I am sure. It has surely gone through a lot of development before it landed. I have high hopes for Places though. I liked having my Bookmarks in HTML format, but that just isn’t efficient.

A simplified version of Places should open up in a sidebar by default. Users of IE7 are going to like their Favorites Center dropdown menu that can turn into a sidebar. I have seem several older users that like to have their bookmarks or history sidebar open all or most of the time in both IE and FF.

How about the biggest mistake of all? Close buttons on every tab. There are two major problems with that.

First, you can accidentally close a tab while switching to it. I could live with the close buttons on tabs if the buttons weren’t active when the tab isn’t active. Currently even though the buttons are still active when the tab is not, they are grayed as if they are not active. I know you will be able to undo closing tabs, but you shouldn’t have to. Opera 8.5 does the exact same thing and it is pretty annoying in the little I used Opera. I closed a bunch of tabs by accident while switching. Opera already has the undo function built in as the usual Edit, Undo. That is easily discoverable and something Firefox developers should think about copying. Current FF extensions that provide the undo function I think are too complex for most users. They are great for advanced users, but for most people a whole lot of undo close choices such as window or tab and which ones or all going back several actions is too much.

My other complaint about having close buttons on tabs is because each takes up valuable horizontal real estate. The more windows you have open the worse it will be. Currently the solution is to hide the close button on tabs when you reach a certain number of tabs (or maybe it is by tab width but it doesn’t appear that way). For my setup, after 7 tabs the close buttons hide on inactive tabs. I am sure it is just a bug, but currently when you close the 8th tab they don’t come back though, you have to get back down to 6 tabs. Not considering that, this is an inconsistent interface. Either the close buttons should always be on inactive tabs (which is not a good idea) or they should always be hidden. It is an inconsistent UI feature otherwise.

There is all kinds of discussion on handling tab overflow but there just is no good method. So preventing the overflow condition as long as possible is important. Clearly that is why the inconsistent UI is meant to solve, but a much better solution would be to only show the X on the active tab.

It is hard to believe, but Microsoft does pretty much exactly what I want with tab close buttons in IE7. In addition to them only being shown on the active tab, they stand out less. Firefox’s are big and bright red so along with the favicon the page title is surrounded by graphics. I wonder how many people will make their site’s favicon into FF Xs just to throw people off.

I discovered hitting “Use Current Pages” as Homepage adds all the open tabs. Clearly the button text says that but last time I did it in 1.5 I didn’t have more than one tab open. As currently displayed all on a single line, a user may think FF is adding only the first tab in the window as the homepage. I did and I even knew FF could have multiple homepages.

What is with swapping the Stop and Reload buttons? Neither are likely frequently used when compared to the back button. And if you are putting them in order of most use, I suspect the forward button is used less by common users. But that would clearly be dumb. My point is, don’t swap buttons just for the sake of moving them. Consistency in placement is also important (just look at the huge mistake IE has made with IE7’s toolbars). Every time I go to hit stop now I end up almost hitting Reload. I am currently using large icons which I don’t normally do so that proves to me that my brain looks for the stop button in Firefox as the third button from the right, not the distance and clearly not first looking at the image.

I have also read that the Home button may be on the way out (by default). Maybe that was just discussion or a planned “feature” since it is still here in 2.0 Alpha. Based on usability tests the home button may never get used, but does that really reflect actual usage? I don’t think so. I know there are a lot of people that probably never use it, but a lot of people don’t use Bookmarks/Favorites either and that isn’t being thrown out. When I am using one of my computers I use the Home button pretty frequently. I have a custom homepage with most of the links that I go to frequently. The default homepage is just a Google search page. It is not necessary to go Home to run a search since we have the search box. IE7 hasn’t gotten rid of the Home button though they did move it in their scattering around the toolbar buttons.

The Home page is where a portal site makes sense. When I setup someone’s browser for them, I usually choose Google News or Yahoo News. My Yahoo or Google’s Personalized Home are great choices even if the user doesn’t have an account. The page non logged in users get already have a good selection of information (more so on My Yahoo).

Back to the Bookmarks and History menu, why do we have two options in each that do basically the same thing. Bookmarks has “Search in Bookmarks” and “Organize Bookmarks” which both bring up Places just the same. History does exactly the same with “Search in History” and “View All History.” Not only is this pointless, it takes up space and is are just more menu options you must read through. Even if they are intended to do a slightly different thing eventually it seems a waste to include both. The idea of places is to have everything in one place. Searching and managing. No need for menu items for both.

One reason the wasted space in the Bookmarks menu bothers me is because the overflow scrolling sucks a whole lot. If you have a ton of bookmarks and don’t have them in a well organized hierarchy using the scroll up and down “buttons” is horrible. It is very slow to reach where you want, but too fast to actually see where you are going. Clearly those are contradictory needs, but that is where the problem lies with this implementation. It doesn’t meet either need because it tries to meet both at the same time.

I don’t have a good suggestion how to fix it though. Something like Microsoft’s Personalized Menus would certainly be possible now that Bookmarks and History are in the DB, but I hate that in UIs. A scroll bar would be a bit more usable, but not consistent with what users expect from dropdown menus. K-Meleon uses submenus to deal with this. At the bottom of what Bookmarks could be displayed on the screen it has a [more] “folder” that expands into another row of Bookmark items. It is an easier to use solution than what is currently in Firefox, but it is not a great one.

And a longtime annoyance is the Add Bookmark menu items scroll with the bookmarks. It is a major pain to scroll back up to the top to bookmark a site. I know Ctrl+D bookmarks the page, but I had to look that up, I don’t remember it. Why would I, they letter ‘D’ has nothing to make me think bookmark, favorite, or add. Too bad ‘B’, ‘F’, and ‘A’ already have pretty standard uses.

I have read that Bookmarks/Favorites aren’t used by many people. My observations of others show the same thing. As for myself, I bookmark a lot of stuff but I don’t often go back to them. I bookmark them mostly just in case. Of course, part of the reason I don’t use them is the UI sucks and is slow when handling lots of bookmarks. The other reason is because I built my own custom homepage with most of the links I go to regularly.

They also badly need to get Native Theme Rendering done. FF menus just look horrible on Windows other than those using the Luna theme on XP. It is listed as one of the “Nice to Have” features of 2.0. Meaning we still probably won’t see it till 3.0 if ever.

I also decided to give a Trunk build with Cairo a try (Trunk is where work on the road to Firefox 3.0 is done). Cairo is much better than I imagined. I figured it would still be pretty slow but it wasn’t bad. There are display bugs with it, but my main concern was the speed. Don’t know about slow machines, but it runs fine on my laptop which isn’t the fastest anymore. Other than Cairo there aren’t many visible differences from the Firefox 2 Alpha.

Update: Mike Beltzner, user experience lead for Mozilla, posted a bunch of info on the direction they are planning for Firefox 2. This comes from open discussion with users in their newsgroups. This is a very good sign and I like most of what is planned. But we will see how much actually makes it into Firefox 2.

Leave a Reply