Well, Jobs and Raskin agreed on one thing: They were both utterly opposed to people being able to customize their environments to be more pleasant (or, in fact, possible) for them to work in. That’s what I call the Design Disease: Thinking you know better than your users, and therefore forcing them to do things your way, even when they know a way that’s better for them, given the totality of their environment and workflow. It fits with the Cult of Mac, which can claim things “Just Work” for values of “Just Work” which necessitate building your entire environment out of parts sourced from a single company.
When that philosophy works, it works well. When it doesn’t, you’re the one holding it wrong.
Hmm… literally all of the commercially successful platforms I can think of that target non-specialist users have adopted this philosophy to some extent. I mean MacOS (from before it was called that), PalmOS, Windows, iOS, Android… and now ChromeOS, which takes it to an extreme, and is what I usually recommend for young kids and grandparents, even though I really dislike Google the company. It’s just so much less pain for a computer non-specialist to use. In my experience, most people don’t want to “build an environment” at all. They just want a tool that they can understand enough to have it get out of their way while they get on with their work.
Meanwhile, the year of the linux desktop never came. The closest we have are big slick distros like Ubuntu, which are still way too complex, have too many breakable moving parts, for “regular folks” to be comfortable with – unless they’re especially motivated to learn the arcana. If they do climb up that learning curve, like you and I did, they inevitably accommodate to a whole bunch of idiosyncratic design decisions that went into things like shells and editors and network config: things that, in practice, nobody can redo.
The conclusion that I draw is that it’s a far greater failure to require your users to customize their environment before they can get on with whatever they want to be doing. For most people, computers have always been a means, not an end. I think it’s a pity how easily technical people lose sight of that.
Your bigotry against the disabled is unwelcome here.
Your ableism is unwelcome.
[Comment removed by author]
I have re-read the commend you are replying to and cannot find any bigotry. Can you please highlight the quote where you see it?
It’s ableism to imply that being unable to configure software to be able to work in multiple ways is a good thing.
Yeah. Unfortunately, the Jobs/Raskin school of design paternalism has become completely mainstream in proprietary software (and even in parts of open source). I identify it pretty strongly with corporate-capitalism: the need to standardize interfaces in order to make the workers fairly interchangable, even if the lack of configurability ultimately makes doing real work harder on average & makes doing any work at all impossible for many people.
I didn’t intend to glorify this part of their perspective. Mostly, I dislike the way that Raskin’s UI ideas are totally unrepresented in modern UI design. We can & should have as large a widget toolkit as is available, and have the flexibility to use unusual designs when it makes sense – and, assuming we completely ditched proprietary software and completely ditched proprietary-oriented structures for non-proprietary software (like distribution of binaries), this anti-standardization could be made to improve accessibility across the board instead of making it worse (as it would in a shrink-wrapped context).
Jef Raskin is not exactly an unsung genius anymore, but I’d still say “undersung”. Something of a tragic hero in the classical sense.
The Canon Cat is a little rare and expensive when you can find it, but a Swyftcard replica for your vintage Apple II is pretty affordable still. Also the Cat software (written in Forth!) is fully emulated in the MAME suite. If you find this stuff interesting, I’d strongly suggest at least reading Raskin’s book The Humane Interface. Wikipedia’s page on Archy has some interesting tidbits too.
Apart from some good ideas on human interface design, there is a broader lesson to be learned about the real political and economic reasons why technical projects succeed or fail. Smalltalk, Oberon, Lisp machines, BeOS, NeXT, the Newton… all worth study.
Lobsters, what are your favorite coulda-been systems? I’d especially love to hear from the old-timers among us.
I’m the author of the above piece – which I intended to be part of a whole series on coulda-been systems (before Sonya got a real job / hit crunch time & Exolymph essentially shut down). I specifically wanted to cover Jot, xu88, & xu92, along with HyperCard & HyperTIES, at the time. (As I’m learning more about NeWS, I’m getting more excited to cover it as well.)
I probably gave the Lisa a short shrift in this, since my primary source for that was Folklore.org & period documentation. I got to use a Lisa, briefly, last weekend & it was not as bad as I had heard. So, maybe if I do the rest of the series, I’ll cover the Lisa in more detail, since I now know somebody with fairly deep knowledge.
Other candidates: Plan9, the Alto, OpenDoc, the Cambridge Z88.
I recently submitted the intended second entry in the series, about the personal robot market of the mid 1980s (with a special focus on the Arctec Gemini), to Tedium, & once it’s published I’ll post it.
Mainstream PowerPC Amiga. :’(
Making something impossible to configure makes it impossible to use if you’re disabled (or even less able) in some key way which hampers your ability to use the defaults. This is a form of bigotry, and must be stamped out wherever possible.