Ryan Nichols wrote:
The reason you would want to usa a 'popup' is for contextual information. Usually this is in more of a web application scenario than a website per-se. So you have to think more broadly in the term of website than serving documents with content in them (ala 'surfing'). Contextual information has been used for a looong time in user interfaces. Just think of the numerous desktop applications you use on a daily basis and how they use pop-up windows to either prompt for more data, or provide other useful information.
Computers are incredibly complex things to use and it generally takes a long time to learn how to use one properly... even double clicking is a skill that must be learnt. Not too many web sites have the luxury of making people learn how to use 'em. The only pop-up windows I have encountered contained advertising, or content better suited as inline content... not really helpful stuff like preventing me from overwriting a file in a save dialog. Also, with desktop applications you usually cannot proceed with a task until you have completed whatever subtask a dialog (pop-up) requires, which is not the case with pop-ups in a web browser.
If it's not directly related to the task enough to include in the page content how is opening a pop-up window any more conducive to task completion or any less confusing than inline content to someone who hasn't used your web application before?If you think about these scenarios, it is when the application needs to present something new to you, but do so in a way that you don't lose your context. You don't want to alter the layout of the 'page' for this, because the content may not be directly related, and it can confuse the user from accomplishing their task.
But wait theres more? At what point in making a sale do you want to interrupt that process? If it's related, and important enough to the checkout process, then include it inline. If not, then it's not really about task completion... it's about something else entirely. There are entire careers to be had in discovering why people abandon shopping carts.
Think of complex scenarios such as when your funneling a user through a shopping cart checkout. In these scenarios, you do not want to distract the user in any way from the task, you don't want to confuse them. However, often you might need to collect data on a particular topic related to the checkout process. This is a fantastic time to use contextual windows.
They allow the user to answer the question in a way
By this reasoning it's ok that each time you save a file your OS asks you something related to files, but not the file you are saving. For example:that they can still 'see' or be aware of what they were originally doing, rather than going to another page and losing context. Now the reason 'it's up to the user' bit doesn't apply is because this is an application.
Alert: I've notice that you haven't saved anything in "Projects December 2002" for a while. You have a file called "notes2003.txt" will I make a copy in "Projects December 2002"? [OK][CANCEL]" (wait a minute... that's XP)
The user is already using the application and they are
I completely agree that applications should help in task completion. However, as stated earlier the examples given don't seem to do that.actively engaged in it. Applications are geared to helping the user accomplish a task. Passive browsing is different, and most of the arguments expressed here are great points for that user context.
except with either the back button (which doesn't break when you don't open a window) or a good ol' hypertext link.
To use another example, imagine a long article on a webpage. You funneled through the navigation and selected your article you are going to invest time to read. The article has a lot of diagrams. This is a good place to use contextual information. Think of a physical book. Ever been anoyed at having to skip ahead or back to find that diagram they referenced? Your annoyed because you lost your context, you lost your place in the text and had to go somewhere else to find the extra information. In a web article, those same diagrams can 'popup' in small windows, and you can view and close them without having to lose your place. This is not possible by sending the user to a whole new webpage.
I am a dial up user at home. Content must still load regardless of which window it loads into. Generally user testing finds against the use of pop-ups.Trust me, user testing would find the contextual scenario much more pleasing. Also remember dial-up users. Loading and reloading that page takes time, even with 'cache'. Viewing the contextual information is much faster if it appears in it's own window.
Yes it is about usability, and some thing's don't carry to a new medium: the web is not print, television, or a desktop application.Now you could use DHTML for this. This is a pretty feasible alternative. But it has drawbacks all its own. The issue is not the mechanism for 'popping up', it's the usability of contextual information that is the issue. Remember there are years of history in user experience design for applications, and those tried and true methods don't fall off the face of the planet with a new medium :)
./tdw ****************************************************** The discussion list for http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list & getting help ******************************************************
