Jef Raskin, who helped design Apple’s classic user interface, is working on a new system, THE, that could be a big improvement. BusinessWeek features an article introducing the technology. Jef also replied to our forums here and here a month ago, explaining his system.
THE Key to User-Friendly Computers?
2003-01-23 Graphics 24 Comments
I’ve read Raskin’s book “The Humane Interface” as well as a lot of negative criticism of Raskin from people who don’t understand the nature of what he is trying to accomplish. Having once been a graphic designer, I remember the scores of conflicts I had with superiors regarding design matters; inevitably, their argument was that good design is a matter of perception and there was no one design that was best for everyone. Science, on the other hand, has often proven otherwise. Virtually every human mind perceives information in much the same way. The reason serif style fonts are used for large amounts of body text is because they “lock” the eye onto the page due to the fact that the hooks in them makes them slightly harder to read. Why is this important? Because making something slightly harder to read means people must concentrate more intently, thus improving retention. Many people hate serif fonts, but that is because the fonts are doing what they are supposed to do, increasing concentration, and many people experience this as discomfort. Then it becomes a matter of perception. Jef Raskin understands that there is a BASELINE set of techniques that would make using computers drastically easier but many people accuse him of trying to stifle customizability (is that a REAL word ;-)). Raskin only wants us to start at a strong common point; the bottom line is the concept of applications is grossly outdated. I’ve discovered through my own research that about thirteen unique interfaces are required to perform virtually every known task for which computers are used. Yet, the main selling point of MS Windows is that there are THOUSANDS of unique programs for the environment; regardless of whether you accept it or not, that situation makes computers HARDER to use. Many of the tasks in those programs are duplicates; Raskins wants to seperate data from command sets so that people only have to learn ONE command set that would apply to the majority of tasks. Any additional or specialized commands could be added AS NEEDED thereby practically eliminating “command clutter.” It’s a proven fact that typing is the single most efficient means of input for the computer so he has developed a means of activating most general commands via the keyboard. However, commands that are best done with the mouse are still available and the two different methods have no overlap. This simplifies computing TREMENDOUSLY with little to no true disadvantage to the user. I’m greatly simplifying Raskin’s work but that is the gist. If people open their minds, they’ll see that the time has come to change the way we work. I think Raskin is on to something.
My understanding was that serif fonts are used because sans-serif fonts were just too hard to read on early printing machines and typewriters.
It seems to me that this THE has a lot in common with VI. If I understand it right, his ‘Leap’ functionality is there in vi by typing slash, characters to find (as in ‘less’).
Furthermore, Mach (above) suggests that keyboard input is the most efficient way of input for the computer. Vi exactly follows this principle and has a lot smart key combinations (well, they are hard to remember, but if you manage to learn them it works fast.)
I am not sure if THE is for everyone. I think that if you do not work much with a computer, that the WIMP interface has the advantage that you can see (in the menus) which actions are available and that you do not have to remember that.
For programmers and other people who work a lot with computers, the THE philosophy is quite good I think.
“I’ve discovered through my own research that about thirteen unique interfaces are required to perform virtually every known task for which computers are used. Yet, the main selling point of MS Windows is that there are THOUSANDS of unique programs for the environment; regardless of whether you accept it or not, that situation makes computers HARDER to use.”
So you want to take away my right to choose my ideal environment for accomplishing my tasks? “You must work in the following manner.”
Next you’ll be dictating what type of clothes to wear as they are ‘the only clothes you’ll need to be most efficient’.
I don’t buy it. At all. People are different, they’ve lead different lives and endured different experiences. Some people like Vi’s modal nature as it allows rapid editing. Some people like Emacs’ complex keyset as it gives them control. Each man to his own, I say. THE will NEVER be mainstream. I’d bet my house on it. (Not that I have a house.)
THE is not like Vi. After a cursory reading of the literature, I jumped to the same conclusion, but there are radical differences: Vi is quite unapologetically modal, whereas THE is specifically designed not to have modes (rather pseudo- or “quasi-modes”). The searching in THE is always incremental (a bit like Emacs but with some additions like leaving the cursor on the first character of the search string rather than the last) and so on.
Many criticise THE as being “just” a text editor: THE is way, way more than that, and I would recommed reading the book to understand the reasons behind the implementation. As a developer developing “familiar” style applications, it has already given me some great clues about how to improve my work and make it better to use.
Charlie: THE isn’t telling how you should work any more than Microsoft, RedHat or Apple are.
As a programmer I spend most of my life in a text editor and I am still shopping for “The RIGHT” one…
I dont have access to a mac at the moment so its not really an option to try it on a mac…
Has anybody heard of a porting effort?
I was always taught in my graphic design and page layout classes that serif fonts are better for body text. The serifs help the letters connect and flow together smoothly, thus making large blocks of text easier to read.
As I understand, all of the developers use Macs but there should be people working on ports to the major platforms (Win & Lin).
For the same reason why some people who have had a stroke can’t recognise faces, some people are neurologically better than others at the skills required to design things well from a human-usability point of view. People who go into the computer industry probably demographically are neurologically the worst in this department of just about career area, yet they’re the ones designing the software. You can have scientists developing the theory of how to make things easy to use, but unless you have individuals with the neuropsychological skillset and intuition to implement every little step on a day to day and on-the-fly basis, it isn’t going to go anywhere.
I believe *the* way to make computing user-friendly is by looking at the mass of individuals who produce the computing experience. There are *reasons* why in the now two+ decades of graphical computing that the interface hasn’t been progressed very far.
I don’t think the problem is in the programs avaible, they simply act as needed – cumbersome or not.
No, I speculate that the complete design of the common computer (PC) makes it hard to use. Ordinary people expect to use computers as just another tool, since that’s what it is. But, ahem, they need to deal with spurious IRQ interrupts (?), memory faults (?), ip addresses (?), defrag (?), harddrive/ram size (?), this document is not compitable… (?), and alot of other things that is completely irrelevant to their work or plessure.
IMO this is pure bs, and I can’t see why some gifted people get to fix all these stupid problems and makes it avaible for a fair price.
from my email signature:
the CLI *is* user-friendly, just happens to be selective on who it makes friendship with…
“I was always taught in my graphic design and page layout classes that serif fonts are better for body text.
The serifs help the letters connect and flow together smoothly, thus making large blocks of text easier to
Serif fonts are easier for most people in printed body text because
they emphasise the shape of the whole word, and most readers recognise
words by their overall shape. It’s doubtful whether that applies to
the very low resolution displays on computer monitors (72 dpi as
opposed to 600dpi).
Dyslexic readers do not recognise whole words but work through the
text one letter at a time. They find sans serif fonts easier because
the letters are clearly separated.
This suggests that, for instance, web designers should not specify the
font used in text on web pages, but should leave it to the user to
select a preferred font. This applies even more to email and text
editors or word processors.
(A word processor is a text editor that has a single key to give a
word count and puts line-endings only at the end of each paragraph.)
“This suggests that, for instance, web designers should not specify the
font used in text on web pages, but should leave it to the user to
select a preferred font. This applies even more to email and text
editors or word processors.”
As a graphic designer, the concept of design itself is thrown out the window when you let people “roll their own”, so to speak. The point of design is to create an overall effect of unity and cohesion… specification of typeface is important in that respect…
From the article:
“The cursor is represented by a flashing blue block. Within the
blue block sits a single letter or text command such as space
(indicated by a black dot) or a tab (indicated by an arrow).
Why is this useful? Think of all the times you’ve chased phantom
paragraph commands with the backspace key in an effort to realign
sentences and eliminate shortened lines. There’s also the havoc
wrought by a forgotten tab. In THE, you see the tab by mousing over
the space without having to flip back and forth between viewing
WordPerfect Reveal Codes
This article alwasys seems to pop up every few months… nothing new here
If you want an userfriendly environment, take bash or any other shell of your choice. The CLI really IS user-friendly.
My gosh, Raskin is a genius! NOT!!
Using the latest Mozilla? Hold down SHIFT and type a few letters. Gee, you say it jumps forward to the first matching pattern on the page? How about that.
Seems there are a lot of different reasons for serifs being used in body text. I won’t argue any point other than there is solid scientific data that text written in serif fonts are retained better than sans serif (dyslexics notwithstanding).
As for removing choice, hell, I WISH I only had to learn thirteen interfaces. Many of the tasks within those interfaces are duplicates to retention would be even easier.
I’d like to see some screenshots.
Also, has anyone actually used this program?
The whole argument about improved readability of serif fonts is a load of crap. The only reason that serif is preferred is cultural and historical.
There are about 400 years between Gutenberg’s press and the popularity of sans-serif fonts when the only typeface available to the masses was a serifed roman alphabet. When the Bahaus movement of the early 1930’s began articulating the concept of “utilitarian beauty”, newly created fonts such as Futura were shouted down as being ugly and grotesque. This had more to do with the innate conservatism of old school typographers than actual aesthetic considerations.
Studies which claim to show improved readability of serif fonts rarely take these aesthetic considerations into account. Nor do they take into account the fact that there are many foreign language publications that use the roman alphabet (and even a few English language ones) that get along quite happily with sans serif as body text, with apparently no harm coming to their readership, or readability. Nor, for that matter, do they take into account that no non-roman letterform requires serifs for readability.
In my experience, thrashing the “improved readability” argument is the last desparate rationalisation of a lazy designer in the face of difficult decisions.
Err, Sofa, there are numerous SCIENTIFIC studies that support that serifs are better for reader comprehension; notwithstanding your interesting history lesson, discussions regarding aesthetics are a poor substitute for empirical evidence. I feel that designers that fall back on the aesthetics argument are elitist boobs who can’t stand to not be “experts” in regards to design. Next you’ll tell me that 1 + 1 doesn’t equal 2.
Can you point us to one or two of these scientific studies?
“As a graphic designer, the concept of design itself is thrown out the window when you let people “roll their
own”, so to speak. The point of design is to create an overall effect of unity and cohesion… specification of
typeface is important in that respect… ”
I know, I know. I’ve worked in design education for many years.
A web page is not the same as a printed page. There is actually little
or no need for a graphic designer on a web site. The whole thing is
dynamic – there is no finished artwork.
The designer cannot claim control over what the user sees. Apart from
anything else, the user can resize his browser window to any shape and
resolution that he wants. A good web page will flow to fit. Bad ones
are “designed” with fixed widths in pixels.
A web site is not a magazine or a brochure.
In a good operating system, the user can customise the whole
appearance of the browser – buttons, window borders, scroll bars, menu
fonts, colours. People have a lot of control over their screens. You
can’t then try to control what font they use to read the text on a web
Have you seen how a lot of these studies are done? You put your test subjects into an artificial environment and you present them with pages of randomly selected texts in either serif or sans serif fonts. Then, you time how long it takes to read each page, and ask them at the end what they understood about each passage of text.
Okay, that’s a highly simplified explanation, but you get the point I hope. And this is supposed to yield the result that serifed fonts are better? Please.
And discussions about aesthetics have EVERYTHING to do with making good design decisions, elitist boobery though that may be.