Pollock in light of the Office 2007 GUI
I think Microsoft is on the right track with the changes made to Office 2007. Processing power and screen real estate have increased, and a more expressive GUI is a natural consequence. Why hide functionality in drop down menus when you have space to show it? Showing more information on the choices we give users, is probably a good idea.
Even if you do not want to design you application to look like Office, there are lessons to be leaned. Office 2007 shows how a GUI can be specialized to fit the characteristics of the application. The ribbon might not be a good idea for all types of applications, but for Office it is. Making you own framework for the GUI, instead of relying on the standard menu bar and the toolbar, might be worth considering. You can make something not looking exactly like all other applications, and get away with it. You might even end up with a better product. Personally I believe OpenOffice will look outdated when Office 2007 is released.
I wonder how easily the new GUI framework for VisualWorks, "Pollock" will allow customising your own GUI... My guess is that Pollock will be very good for making new widgets. Widgets look and feel are all implemented in Smalltalk, and you can easily customize an existing one, or make a totally new widget. I guess making something like the ribbon would take a lot of work, but it could be done. I also hope that I can easily customize menus to contain more detailed information.
Pollock does not fix all problems in the current VisualWorks GUI. For example, window styles are not enhanced by Pollock. Luckily, other projects following Pollock will address these issues.
Currently, performance of Pollock has been varying. This is certainly not a concern now, as Pollock is still in beta. However, having a responsive GUI is very important:
From the perspective of the Office 12 new user experience, performance is critical. The very visual nature of the Ribbon, with lots of galleries and Live Preview, lends itself to experimentation. (…) In fact, we've noticed a profound link between performance and usability; some usability tasks that totally fail when the interface is laggy and lethargic suddenly do great once the software is responsive enough.
Previously there has been focus on how Pollock will make sure the GUI looks good, by ensuring screen updates do cause flickering. Pollock definitely also need to have focus on drawing speed.
Comments
Make it run, Make it right, Make it fast
[James Robertson] March 29, 2006 16:03:03.262
Trackback from Smalltalk Tidbits, Industry Rants
Make it run, Make it right, Make it fast by James Robertson
In a post about the new MS Office, Runar mentions Pollock:
Pollock is not really even in beta yet. We are closing in on the first supported release (see the roadmap here) - and here's the thing - don't worry about the performance issues. Our developers have been focused on the "heavy lifting" thus far, and, even before they get to serious optimization (and optimization is a large part of the internal roadmap), they know of a number of things which, when optimized, will make performance improve a lot.
We don't follow the theory of premature optimization here - we want to get the big things right before we focus on the rest of it. Rest assured, Pollock will be fast when it's ready to walk out the door.
[] March 29, 2006 16:53:21.520
I guess he didn't catch the "This is certainly not a concern now, as Pollock is still in beta."
[Reinout Heeck] March 30, 2006 7:30:17.135
James is not that daft,
there has been some discussion about Pollock speed on the VisualWorks mailing lists recently where it became clear that some people have bet their company on Pollock being production ready before it goes beta....
I guess he simply found this a nice thread to comment on for the purpose of making a statement about the Pollock roadmap in the BlogoSphere.