It is now three months since Apple seeded the final version of Mac OS X (build 4K78) to its developers. At that time there were mixed opinions about the interface, and a great deal of anxiety about whether Apple could have the product ready in time for the March 24th shipping date. We've now lived with that interface for a quarter of a year. Some of us have lived still longer with the interfaces of the Public Beta release and the developer seeds. So what does the human interface of Mac OS X look like now, three months down the track? What can be said about the controversial dock, the menus, the Finder? These are the questions that I will address in this column. I'm not concerned here with technical interface issues, such as the lack of speed and the lack of threading in the Finder; rather I'm interested in discussing the interface paradigm shifts that Apple's new OS has brought with it. There's a lot that could be said about this topic but today I'll confine myself to a discussion of the dock. I do this because in many ways it is emblematic of the interface issues that run across the entire operating system.
Criticisms of the dock
The dock is arguably the single most noticeable departure from Mac OS 9's well-known and well-loved Platinum interface. The ability to hide the dock or reduce it in size notwithstanding, the dock is hard to miss. Most detractors of the dock criticized the amount of space it occupied on the desktop, the way the dock could inadvertently obscure window controls (such as the window resize handle), and the way that it presented a "moving target". Others claimed that it was trying to do too much when compared with the applications menu that it replaced. Another significant criticism was that icons in the dock could easily be confused, particularly when they were alike or even identical (as in the case of folders or drive icons or minimized windows).
Answering the criticisms
Three months down the track we can be a little more circumspect about the dock. Have these problems been the "show-stoppers" that it was claimed they would be? Some would say no; others yes. Here, however, my aim is to open up discussion on the issue rather than give a definitive answer. (Needless to say, we welcome Observer comments about this!)
It is true that the dock occupies a large amount of space on the desktop. One can minimize the dock to a tiny size or even hide it from the screen altogether, but both of these detract significantly from the dock's central strength: the ability to convey at a glance a variety of information and to provide easy access to various objects (windows, folders, dockling functions and so forth) and targets. But rather than dismiss the dock because of the space it requires to be used to its full potential, we should be asking the question, "Is it worth the investment of desktop real estate that it requires?" In other words, if an object is valuable enough, then it deserves and indeed demands desktop space. The question here is whether the dock is valuable enough to justify the real estate that it occupies?
My personal view on this one is that yes it does indeed justify itself. It can change the way we work, and for the better. I remember that during and after Mac OS X's Public Beta period people were wishing that the dock could be moved to alternative sides of the screen and that it could be anchored at the beginning or end as opposed to being centrally aligned. I was one of the people who was distraught when Apple removed from one of the late developer seeds the ability to hack the dock and move it. I was also one of those who rejoiced when they learned that the hack was back in Mac OS X Final.
After a week or so of using the dock pinned to the right hand side of my screen I realized that it wasn't so bad down the bottom after all. I eventually started leaving it down there and it remains there to this day. Despite the fact that most of the documents I read or create on my computer are best viewed in a "portrait" (upright) orientation, I've found that the dock's functions are so important that they deserve the central position along the bottom of the screen. Thanks to TinkerTool the dock is able to tell me which app is in the foreground and which ones are hidden (ghosted) -- and of course the dock can also tell me how many new mails I've received, show me a moving graph of my CPU usage and of my CPU temperature, show which apps I am running, tell me the battery level in my PowerBook, and provide hierarchical menu access to each of my drives. It also allows me to drop files onto apps, to bring all the windows of a particular app to the front or hide all the windows belonging to other apps (see Kyle's Hot Cocoa column from last week for these and a list of other tips and shortcuts), and much more. Sounds like a valuable use of space to me.
Perhaps Steve knows best after all...
The criticisms that the dock presents a moving target and that this is somehow unacceptable are unfounded. A few minutes spent launching apps, dragging objects to the trash and minimizing and maximizing windows will convince even the most neophyte user that hitting dock items isn't at all difficult. The scale of movement here is only an inch or two most times -- but users still know that the trash can will always be the right most item. It can be hard to hit a dock icon when magnification is turned on, but if we want to get the most out of the dock we'll leave it large enough that magnification shouldn't be necessary anyway.
But there is still work to do before the dock will be a perfect piece of human interface design. Happily for Mac users, they can rest assured that Apple is working all the time on improving Mac OS X, the dock included, with the goal of making it the best OS the world has ever seen. As such we can expect lots of goodies in the future: better differentiation between similar looking icons, the ability of applications to customize their pop-up dock menus (much like docklings already do), more extensive use of dock icons to display application information (as "Mail" currently does with it's new mail indicator), the ability to break the dock into multiple segments (putting applications and windows on entirely different edges of the screen, if the user desires it, for instance), dock icons of variable widths (a volume dockling, for example, need only be a few pixels wide; a stock ticker might be several times wider than today's dock icons), more docklings like the control strip items of old, and the dock could be made to behave like the NeXT "shelf" (as a temporary storage area for files to be used in copying and dragging -- note this is distinct from the current use of the dock where files dragged to the dock only really exist there as shortcuts). These are only some of the possibilities that Apple is working on. Rest assured that there are many more that we haven't even imagined yet...
"Iconic" versus "Photographic"
The dock is also the place where Apple's decision to abandon "iconic" icons and switch to "photographic" ones is at the fore. Apple attracted some criticism for this move from certain interface experts. These experts argued that the virtue of iconic representations (in other words, pictures which eliminated detail as far as possible without destroying the recognizability of an object) was that they could be apprehended almost instantly. Take the example of a simple folder icon: in Mac OS 9 it was composed of a simple linear shape with minimal shading and uncomplicated edges. The argument is that such an object can be recognized for what it is with extreme rapidity because their is less material for the eyes and brain to scan and process. Compare this with the near-photographic drive icons we have in Mac OS X; the argument is that these icons are more complex both tonally and structurally and hence require more mental processing to be recognized.
Once again, the issue shouldn't be one in which we dogmatically cling to long-held beliefs about The Right Way To Do Things. Rather, we should ask what the relative benefits and costs of the new method are. In this case, we have lost visual simplicity. Perhaps we've made it a little harder to quickly scan a set of objects and locate the one we want. But what have we gained? I would argue that we've gained an operating environment of vastly improved aesthetic quality, with a corresponding increase in the pleasantness of our user experience. More pragmatically, we've gained the ability to display huge amounts of information in icons -- gone are the mere alias arrows of the icons of old; here we have icons which can convey all sorts of things: incoming mail, waiting instant messages, the progress of our SETI@home client, and so forth. One need look no further than xicons.com to see the awesome opportunities for customization and differentiation that these new icon frameworks provide, and the popularity they have rapidly attained.
As with so many other things in Mac OS X, whether you love the icons or hate them at first, if you continue to use the OS for long enough there is an excellent chance that you will start to warm to them. Even the most conservative Classic Mac OS fan can have his or her dogmatism gradually eroded by time spent with the new OS. My argument here is not that Mac OS X is perfect -- there is undoubtedly a long way to go yet -- but rather that many of the paradigm shifts it brings with it are actually exciting opportunities for the future and not scary departures from the past.