McFilter

Usability For The Masses

Today, Jon pointed to this rant about how Linux usability is doomed to failure by John Gruber.

I think the underlying point is that usability is difficult, and that it is much better to design it in from the start rather than try to tack it on as an afterthought. On this I totally agree with him. Unfortunately, he tries to back this up with a number of points which really grate with me.

The argument seems to boil down to "normal developers are just code monkeys, whereas us UI designers are artists and deserve lots of money and so can't do open source."

• "Talented programmers who work long full-time hours crafting software need to be paid. That means selling software." What? None of the guys who've written open source software are talented?!?! I'm sure they aren't all talented, but in my experience neither are paid programmers. Or is there something special about UI programmers that they are purely driven by money, rather than any of the other factors which motivate the rest of us? Are the UIs in John's own open source projects deliberately worse than in any paid work he's done?
• "The distributed, collaborative nature of open source software works for developer-level software, but works against user-level software... Movies are collaborative art, but require strong direction. So it is with end user software." This implies that "developer-level software" doesn't require strong direction. Good software design is a creative process best driven by a small core team of architects. The distributed, collaborative model of development will only work well if there's either enough code surrounding the new module that it's obvious how it will fit in; or there is effective communication of the designers' ideas. I fail to see how UI design is different, but if it truly is, then surely that is an obstacle to be overcome, rather than an immovable barrier?
• "Most programmers don’t have any aptitude for UI design whatsoever... no amount of practice or education is going to make them good at it. Improved, yes; good, no." So there's no point in complaining about it, we might as well all go home now, no? Sorry, but I'd take improved over bad if that were the only option; but it's not: improved visibility of usability is a good thing on it's own. Maybe some of the coders who really can't do UI will realise, and seek input from those who can. Maybe some of those coders have got an aptitude for UI, but haven't bothered to exercise it as it didn't seem important. And I don't buy this "It’s an art, and like any art, it requires innate ability." bullshit. That's like saying that because I may not become a concert pianist, there's no value in learning the piano, or that piano cannot be taught.

A much more productive, and constructive, response to the Eric Raymond article is the one I've been following by Jon Udell, from his initial discovery that OS X isn't any better through more recent articles promoting the development of UI and usability guideline "standards" and the issues with RSS going mainstream.

As long as usability and UI design is kept sacred for the "UI specialists", it will be given short shrift by those implementing the systems, because it's not their job and they don't have to worry about it.

It is the responsibility of all engineers to worry about, and aim to maximise, usability.