Sundry ‘Spray-On’ Clarifications and Corrections

Tuesday, 6 April 2004

Parallel Ports

First, forget the crack about parallel ports. Please. You have no idea how much email I got about this.

To answer the most frequent question, no, I wouldn’t throw out a working, reliable laser printer just because it had a parallel port interface. What I would do is spend $50-60 on something like this so that I could just plug it into my network and be done with it. But hooking it up to one computer and then sharing it using software is perfectly reasonable.

Mozilla

Also, forget what I wrote about the Mozilla project being an example of an open source project that has produced some decent GUI software (Camino and Firefox) thanks to extensive corporate backing. Camino and Firefox are good apps, but Matthew “MPT” Thomas points out why they aren’t good examples of the point I was trying to make:

Yes, Camino and Firefox were begun by Netscape programmers. But in the early days of both projects (then known as Chimera and mozilla/browser, respectively), those programmers (Mike Pinkerton for Chimera, Ben Goodger and Blake Ross for mozilla/browser, and Dave Hyatt for both) were terrified that Netscape would shut them down.

Camino and Firefox are as well-designed as they are because:

  1. they’re not designed by Netscape’s incompetent designers (like the Mozilla suite, and especially Netscape 6/7, were),
  2. they have few primary developers, and
  3. those primary developers have become fairly good at design as well as programming.

Unfortunately the developers of most other Free Software projects aren’t as good.

Chicken vs. Egg

I argue:

  1. The user interface — the entire user experience — should be designed first, and the underlying implementation should be built to support the design.
  2. Thus, the traditional Unix/Linux development model, wherein back-end implementations are written first, and GUI “wrappers” are written around them later, is completely backwards.

A bunch of people seem to think this argument is at least partly undermined by the fact that Mac OS X wraps a GUI around CUPS, the same underlying print architecture that gave Eric Raymond fits. (MPT, for example, mentions this.)

But I don’t see this as a contradiction. For one thing, Mac OS X’s print architecture is a lot more than just a “wrapper” around CUPS. But even if I concede this point with regard to CUPS, it’d be an exception, not the norm.

Yes, Mac OS X is built with lots of open source software under the hood. But point #1 doesn’t imply that the entire implementation needs to be written from scratch. Yes, Mac OS X contains a lot of open source software under the hood — but these components are used to implement Apple’s designs, not the other way around.

Switchers to Mac OS X

What seemed to tie a lot of knickers into knots — especially in the Slashdot crowd — was my assertion that “Unix nerds who care about usability are switching to Mac OS X in droves,” and that those who remain are “either cheapskates or free-software political zealots”.

I heard from Linux users who claim to be neither cheapskates nor political zealots, but who have no intention of switching to Mac OS X, under any circumstances, ever. The reasons vary, but common ones include:

  • Mac OS X does not have an option for “focus follows mouse”.
  • Mac OS X does not allow you to use a different window manager.
  • You can’t change the way Mac OS X looks without resorting to unsupported hacks.

But the particular reasons don’t really matter. It all boils down to the fact that most aspects of Mac OS X are not designed to be configurable or replacable; they are designed to be usable, and to fit in with the design of the rest of the system.

They’re also designed to work specifically with Apple’s own hardware — which many of these “I’m not a cheapskate but I don’t want to pay for Apple hardware” types refuse to recognize as a huge usability advantage for Mac OS X.

I didn’t say “Unix nerds” are switching to Mac OS X in droves; I said “Unix nerds who care about usability”. People who want a Unix system that just works, so they can get on with their real work — those are the ones who are switching. As opposed to Unix nerds whose interest is the computer itself, and who want to tinker with it at any and every level — i.e. Unix nerds who do not care about usability.

Rootin’-Tootin’

Perhaps the biggest misconception is that I’m somehow “rooting against” desktop Linux. I really don’t see how anything I’ve written implies that, unless you subscribe to the “if you’re not with us, you’re against us” school of thought.

Regardless, it’s not the case. I’m not rooting against desktop Linux, nor have I ever claimed it can’t succeed. What I am saying is:

  1. It hasn’t succeeded yet.
  2. It’s unlikely to succeed without direction and substantial commercial support.

Where by “succeed” I mean “provide a terrific user experience”.

The commercial support is there. Companies like Novell and Red Hat (and others) are employing teams of open source developers, many of whom care very much about UI design. But the key word here is direction.

Good UI design will never be forged by community consensus. Bottom-up design won’t lead to a coherent and unified whole. Cf. John Siracusa’s comments about desktop Linux in his interview with Robb Beal:

Right now, the Linux community values “diversity” too highly to ever get a single, consistent GUI, let alone a good one. At the same time, it holds on doggedly to its (often ancient) Unix-rooted traditions and conventions. Finally, it’s hard to get a really large group of Linux developers to do much of anything beyond a single “project.” A GUI is not a “project”. It’s the whole OS from the user’s perspective, and it must be from the creators’ perspective too or it will fail.

I think Linux as an institution is allergic to a good, consistent GUI. Their priorities are reversed. They want to build a GUI on top of an OS. If they want to compete on the desktop, they should be building an OS to power their GUI. Of course, they don’t have a GUI, which is problem number one.

They need to think of what they want the user experience to be, and then design a system that provides that experience. Period. This is so basic that even Apple forgets it from time to time. If you don’t know where you want to go, you’re never going to get there.

Havoc Pennington — who works for Red Hat as a leading developer/designer for the Gnome desktop — gets it:

At Red Hat, we’re building the desktop team around a top-down design-first approach, driven by professional interaction designers. We’ll see how it works out. It will take some time before the first results are visible.

The key is that there’s never going to be a good desktop user interface for Linux that pleases the Linux nerds who don’t care about usability. If the reason you use Linux is that you value tweakability over usability, or if you get off on the fact that a normal person couldn’t sit down in front of your computer and figure out how to use it, you’re probably not going to like a system that doesn’t even have a replaceable “window manager”. Trying to create a cohesive GUI system that appeals to these guys is like trying to write music that appeals to the tone deaf.

The worst part is that if anyone succeeds at putting together a usable desktop for Linux, these anti-usability Linux advocates will piss all over it.

It’s not worth listening to the opinions of assholes; and but once you shut them off, they just get angrier.