Tag Archives: user interface

The Tabtitlebar – Where Apple and Google went right, wrong or M.I.A.

This will be short.

Already, skirmishes of opinion over Apple’s interpretation of Google Chrome‘s tabtitlebar – or the positioning of the tabs in the title bar of the browser in order to utilize screen space for webpages – have erupted on various news sites and blogs. Even on OS X, the jury over Safari 4‘s user interface design is divided; on the Windows port, most who express favor with Safari 4’s user interface design are usually users of Windows Vista, so the main supposition is that those who have the most trouble with the tabtitlebar are Windows XP users; oh, and (on a cross-platform basis) people who simply like more screen real estate for pageviewing.

Of course, Google Chrome Beta (Google’s first browser, and currently Windows-only) was the first notable web browser to publicly promote the UI infusion, and it was not without its initial criticisms for that same reason (among others, including the option of sending anonymous usage and crash data to Google). Safari 4 Beta, on the other hand, marks a first for Apple in its legendary user interface history, and may also begin a user interface trend for Apple throughout its Mac OS X applications (Finder, Mac OS’s file navigation app, is hinted to have tabs in early alphas of Snow Leopard, OS X’s 6th iteration, so it may well take up Safari 4’s tabtitlebar feature).

Many have written about problems with this emerging user interface widget, which utilizes the operating system’s window manager by embedding navigation between tabs within the title bar (mostly on Windows). Personally, I think it is a good idea due to a month or so of using Google Chrome (I also find the process-per-tab idea to be a good idea), but problems still linger with both implementations as far as I’m concerned.

  • Speed: Chrome is far better in the speed department than Safari as far as opening, closing and navigating between tabs is concerned (again, maybe because of the process-per-tab idea). The Windows port of Safari 4, like its Mac port, does not allow an option (AFAIK) for using middle-click to close tabs; left-clicking to close tabs in Safari 4 on WinXP feels like chipping into brick. The responsiveness to clicking buttons to close and open tabs is horrendous on this platform.
  • Tab navigation and perception: In this area, I’ll say that Firefox 2.0+, despite its own speed and resource management issues, did a better tour-de-force at tab navigation than Opera, Chrome or Safari, by allowing one to scroll through as many tabs as one may please without the tabs being scrunched up in the tab area. All the other browsers, even Chrome and Safari, continue to maintain the metaphor of the "limited tab area", with Safari 4 doing worse by maximizing the view of the tab of the current page in view at the expense of other tabs-in-waiting.

Yes, Firefox actually did something that I’ve come to appreciate (and wish that they could extend into a much wider territory of tab navigation; getting rid of the tab-specfic close button in 3.0 remains questionable, IMO).

There is room for improvement for Safari and Chrome in this combination of two user interface metaphors. In fact, there’s room for improvement for most of Apple’s software ports to Windows, but I highly doubt that Apple is willing to consider such ideas as unscrunching the tabtitlebar for tabscrolling or giving separate processes to each tab, just as how I doubt that Microsoft will do the same with IE8 or IE9.

The Mobile Web, and why wikis aren’t “mobile applications”

The Mobile Web is the World Wide Web as seen and used on millions of mobile devices.

The best mobile web browser, arguably, is Safari on the iPhone OS, at least because you can surf the Web using up to 8 tabs.

However, I, like Google, thinks surfing the Web on a mobile device, even the “Mobile Web”, sucks. Caching web pages on Safari tends to be very short-lived, albeit because of the poor battery life of the iPhone and iPod touch, and copy-paste would be nice to use on the iPhone OS.

But most of all, using “made-for-iPhone” web applications on Safari tends to be sucky (although Facebook’s web app tends be a bit more usable, touchy-feely and animated than any of Google’s Web-based offerings). So with the opening on July 10th of the App Store after months of preparation and queuing by both development startups and Web-based companies who wanted to deliver a better mobile approach to their web services than would be reachable on their own mobile browser-based services, I suppose that a new rule of mobile Internet-dependent software development was realized:

Never create for a web browser what you can create as a native application that can pull and present data from off the Web or another Internet-based application.

Or: Apple may have intentionally let WebKit deal with “made-for-iPhone” web apps so horribly in order to pressure Web-based companies into making use of the iPhone SDK and the App Store to deliver far-better, but more proprietary, user interfaces for RIAs than could ever be hoped for on Safari for iPhone OS.

Either way, this means that the “apps” of the Web (social networking, blogging, email, pseudo-SMS, etc.), if they want to attract more flies, will have to be recreated as native apps for the iPhone OS, Symbian OS, and other current and future operating systems (with the exception of Palm. Their OS doesn’t deserve any recognition, LOL.). It may be nice for Apple to allow background processes in the future to allow for cross-platform development and widget design for the same-branded native apps on more than one OS.

However, after all the web apps for mobile phones have been turned into native apps, what will remain on the Mobile Web? Or at least what websites will remain that can’t be fully or successfully converted into native mobile apps (it feels like I’m asking about who will remain after the Apocalypse or something like that)?

I think one candidate for such a position would be any wiki website, especially Wikipedia.

Simply put, wikis are dynamic websites that allow one to create a page about anything (that’s notable or important); they also make it rather easy to link to other articles by typing “[[ ]]” around any word or group of words, although that means if a linked article doesn’t exist (hence the red text of a non-existent page link) then one can create the missing article by clicking on the red link, which presents a “Create this page” page. Rinse and repeat.

As you can see, an article on a wiki can fill up pretty fast with an array of links to, well, anything on the wiki that needs “wikilinking” (oh, and the slightly less-common, but necessary hyperlinking for external links and references).

So trying to create a hypothetical Wikipedia application for mobile touchscreen-input access on the iPhone OS is something that may very well run aground when one has to open up links of either the wiki- or hyper- type.

Plus, links on a webpage, especially on a wiki, aren’t designed as buttons. They’re designed as text links (and image links that direct to the original wikipage of the image). So hyperlinks or wikilinks can’t, at the moment, be considered as widgets that can be tapped in order to deliver or render animated transitions of the user interface from one page to another, except for the basic loading of another page. And let’s not forget about how, whenever one opens up a hyperlink in a native app, the webpage immediately and automatically opens up in Safari, thus bring us back to the Mobile Web.

One developer can partially remedy this for the creation of a native Wikipedia or wiki reader by simply stripping the text of any links, although that would take away the “fun” in navigating through Wiki articles. Another can remedy this by styling the wikilinks, reflinks and hyperlinks as touchy-feely widgets, although that would result in a wide interspersion of “embedded” but raised widgets within the text.

Finally, no web-based third-party Wikipedia-for-iPhone reader in existence has taken on how a wiki can be edited within the iPhone OS using a virtual touchscreen keyboard that takes up nearly half the screen’s real estate. Not even Wikipedia’s own official mobile interface.

Do I think that a mobile wiki that can do just as much as a web-based wiki can be created? I don’t discount the possibility.

But I don’t think that the current situation of mobile software development or mobile user interface design lends a great deal of resources to such an ideal as a mobile, iPhone-accessible wiki interface.