I hate word processors. In my opinion, they are an excellent example of something, that is really natural and easy for humans to do with pen and paper, done so hard that it baffles me every single time I need to do something I previously didn't need. Or indeed, do something I do rarely. Sure, the people who've been using these unnatural abominations for years and years know how to operate them, but if the casual user wants to do something as radical as putting an image somewhere in the document and wrap text around it they can be in for a world of pain. Or, let's say I want page numbers, starting from the very first page but not displaying them for the first ten pages.
It's all in the help of course, but knowing where to look is not always obvious. Open up your word processor's help and look at the size of that thing for a quick reference of what I'm talking about. This is basically understandable, as word processors have (and I guess they need to have) a plethora of features. A lot of these features are stuff needed only by selected few of us - each user group needs only a fraction of the entire feature set. The problem is, with this many features, finding the one you're looking for is difficult. Everything is organized, but before you can make any use of that, understanding the categorization logic is necessary. Simply put, they are not context-aware.
What games (good ones anyway) do effectively and word processors do not, is providing information on how to use them when you actually need that information. Sure, Microsoft tried with the Paper Clip of Mighty Annoyance, and that didn't turn out well, but is also no reason to stop trying. One problem of course is that it's quite hard for the application to understand what the user is doing, due to its monolithic design. Still, the application should at least provide a help shortcut on each and every one of its dialog windows, to open up a list of help topic relevant to that particular dialog. Help should also be action-oriented. When I want those page numbers, I don't want to know what each and every gadget on the dialog goes, I just want to know how to get my page numbers exactly where I want them.
Of course, games have the definite advantage of much better understanding about what the player is about to do, due to their predetermined nature. Games with linear progression are obvious, but also sandbox-games where the player can go where he pleases can still usually have some idea what the player might need to know next. But it's not like we can't make word processors context-aware. Perhaps it'd be best to actually break down their monolithic design, and try an object oriented approach, something programmers should be very familiar with, but which strangely doesn't find its way to user interfaces too often. I think it's better to provide an example scenario.
I start writing a document, and at the moment the word processor only provides the very basic features for writing text and positioning it intelligently. This is all the word processor itself does: manages positioning and text input. Then I realize I want a picture, so I paste in a picture component. When I click on this component, I only see tools for picture management, which is again something I don't need to see when I'm not setting properties for my picture. I write again for a while, until I need to paste in a table component. Lo and behold, there's a table, and when I select it, I have all the table-editing tools at my fingertip, and nothing else. Finally, when I'm done I realize I want to alter the positioning. So I summon up a grid, tell my components and text paragraphs to snap to it (to keep them neatly lined) and drag them around until I'm happy with the results.
See? Context-aware word processor. I actually quite recently found out that Word 2007 (been using Open Office in Linux) does a lot of this, so perhaps there is still hope for word processors as well. The last part of my example scenario on the other hand is something I can do in Excel or similar programs, but not in word processors. There could be a very good reason for this, but it's also a real possibility that there isn't, because software generally tends to follow old conventions. Of course, it is nice that all the new software works somewhat like the one's we've been using before, but if no one breaks this cycle, we're stuck with usability models that have been outdated for a long time, and bringing new people, with little experience using computers, becomes harder and harder as time goes by and feature numbers creep up.
So, perhaps it is indeed time to rethink everything from scratch.
Dota 2 Next Level Meta: Support Broodmother
2 years ago