During my three years of usability presence in Open Source projects, printing has twice been the stumbling block to extensive and hot-blooded discussions about the whole purpose of usability for Linux.
I much enjoyed the first one: Being relatively new to OSS usability, the public rant from the well-known Open Source evangelist Eric S. Raymond about the usability of printer installation - an essay he published under the title An Open Source Horror Story was some kind of a promoter for my own work. Suddenly everybody cried for usability =)
The second one, however, was less funny: Displeased by his impression that Gnome tried to reduce complexity through cutting down functionality, for example in its printing dialogs, Linus Torvalds publicly encouraged users to switch to KDE. In the upcoming discussions, usability people were blamed to overshoot the mark, to reduce complexity to a degree where applications are no longer usable for above-average users.
Why is it so hard to make printing dialogs that meet the users expectations? After all, it should just print, no? Famous last words...
In an attempt to find a solution that meets the expectations of all involved parties - Linux distributions and the major desktops, printer manufacturer, network admins, professional designers and home users - the Open Source Desktop Labs have formed up a workgroup that cares for the standardisation and simplification of printing efforts under Linux.
As technical reliability and a sane user interface design should go hand in hand, they invited OpenUsability to take part in the OSDL Printing Summit next week in Atlanta. Celeste, Jan and me will go there and design UI sketches for all printing related tasks - installation, configuration, maintenance, printing and job observation.
A tough job as I learned last Friday when I visited Kurt at Danka, a print service provider, to get a private lesson that let me into the secrets of printing: Apart from the different positions how much functionality should be available, general security issues, permission handling, driver reliability, mixed-network support, adequate error detection and handling, particularities of different distros and printer manufacturers need to be considered.
However, before racking our branes about that, we'll pass a weekend at Celeste's place in Washington D.C. to do some KDE usability stuff. And visit the cherry blossom festival. And the White House. And everything.
... my first time in the U.S., I'm curious =)
Syndicate Blogs
False premise
> the mentality to reduce complexity through cutting down functionality in Gnome printing dialog
This is a false premise. It is not true. The work just hadn't been done yet, and nobody was claiming otherwise. One person guessed wrongly that that might be the reason, and the arguments sprang from the assumption that it was a correct guess.
rephrasing
hm, it was what he said, though. That does not mean it's my opinion, I'm just rephrasing his comment. I've changed the wording to make that more clear.
As Linus himself pointed
As Linus himself pointed out, Gnome is a special case. It's not that they are really, consciously trying to make simple, non-confusing print interfaces, it's the fact that they simply don't have the resources to produce the depth and quality of interface required. Whether this is because of a lack of developers, or development tools and frameworks that aren't quite good enough, one can only guess. The difference between Gnome and other projects, as Linus pointed out, is that they won't admit this.
I can tell you that KDE's print management GUIs, although they could be organised better, have been a godsend for me on many occasions especially when managing print queues and other settings. I, and other people using them, would not wish any features to be cut out because they might confuse me. With careful development I don't see how people cannot have a simple set if interfaces for printing, but with the more complex stuff behind the scenes that administrators and many users, at one time or another, will want.
Where do you draw the line? Do you purposefully discourage graphical management tools for stuff like LDAP, Apache or CUPS because it might confuse people (different group of users, but important none the less)? I would call those things essential, and is exactly what any GUI environment should strive to provide.
Where do you draw the line?
Where do you draw the line?
With regard to printing, that's exactly what we'll try to figure out in Atlanta. The major challenge will be to find a solution that makes everybody happy: The network admin, graphics artist, office worker, home user, and both KDE and Gnome.