Slashdot Mirror
Jef Raskin Gets $2 Million To Develop RCHI
Dr Twox writes "The Raskin Center for Humane Interfaces has received a $2 million dollar boost from a multi-national corporation to further develop Jef Raskin's RCHI project, a radical new and simple to way interact with computers. Co-creator of the Macintosh and author of The Humane Interface, Raskin hopes to have RCHI finished within 18 months. "When you actually try it," says Jef. "It actually does what we say. We've got the goods."
It's built with Python and SDL, so how long before someone ports this to *nix?"
...he got funding from a multinational corporation?
Aren't we supposed to, like, hate that, or something?
"It's built with Python and SDL, so how long before someone ports this to *nix?"
Umm... correct me if I'm wrong... but wouldn't it more or less run out of the box?
Or are you really asking how long before people take it, strip it down, and glue bits piecemeal into things like Gnome or KDE, and gut it so the old-timers don't raise heck over the changes (cf. Nautilus spatial interface instead of browsers)?
No, I don't have any love for the want-better-but-hate-change crowd.
Jeff has been promoting these extremely simple interfaces since the late 1970s. The original MacIntosh computer, before Steve Jobs co-opted it and jammed it full of Xerox GUI technology, was supposed to be like this. Then Jeff partnered with the Cannon [ copier ] company with the CAT-PC. This PC had no explicit operating system. It came up in a text edit mode. The disk was one giant piece of text you could search and edit. You could highlight sections and execute them as computation.
check out the Flash demo[8MB]:
http://www.raskincenter.org/main/img/zoomdemo.swf
yet boosts claims that it really is as good as he claims "when you try it"?
he's got a crystal ball too, then? maybe that's integrated to the product to make it guess what you want. like clippy on speed.
world was created 5 seconds before this post as it is.
Do this mean that with the new interface, his web site will actually indicate what it is he is talking about doing ?
i hope the interface they're designing is better than the one on their website...
if i'm a grammar nazi, you're an illiteracy nazi.
They created the Apple II.
--- Ban humanity.
does one of the masters of UI have such a hopeless website? Everything in some inane monospaced font, and on a single page. A specification that relies on the Find command for navigation. Gah.
As I recall from years ago, Jeff used to claim that the division between "Operating System", "Application" and "Content" was big learning barrier and slowed down computer use. So he would essentially abolish the first two items, or at least keep them largely invisible from the user.
I wonder if something like Google Desktop is along these lines. You'd use that to immediately find some information to act on, without having to muck around some cluttered file system.
Likewise MicroSoft's attempt to webify the desktop and access it through the browser is another attempt at hiding barriers. (I will make no comments on whether I think It is working adequately.)
I tried the flash demo. It's like a virtual screen that you can zoom in and out. Virtual screens have been available for years. The zooming thing is kind of interesting but not anything new. I assume the interface is based on a vector drawing back-end (display postscript/PDF?).
This has several problems. Is this thing suppose to manage all your documents and applications? Does that mean everything is being displayed and active at the same time? The CPU and memory requirements of this must be off the chart. This thing would totally choke based on the pure amount of data I have on my machine. Can this interface handle a terabyte or more of information?
Spacial interfaces suck anyway. It might seem like it is better for organizing your data because you can group things together and "zoom out" to view everything on a large scale, but in real life you're going to spend too much time zooming in and out trying to find what you are looking for. It is very much like those suck-ass 3D file managers that someone creates every once in a while.
I suppose you could query for items and they could be marked similar to MapQuest, then you could zoom in on it. That sounds like a very tedious to use interface though.
Really, the current UI system that most computers use is not a bad design, it just needs refinement. Modern UI's just need to be better about remembering which data items I've been working with recently and which items go with each other. We are already seeing the beginnings of this with things like "favorites" and "home/desktop" in most file dialogs these days. That just needs to be taken to a higher level and cleaned up.
Sorry if my post is disorganized, I just woke up...
The ratio of people to cake is too big
Let me explain how the system works. All interfaces involve a combination of directly modifying document contents, and issuing commands. Consider a word processor. In VI you have to switch between modes to issue commands and edit text. In emacs there is one mode, with shortcut keys to issue commands. However, as the number of commands grows these shortcuts become increasingly archane. In a GUI commands are issued by clicking in a menu or toolbar, but again as the number of commands increases you can end up with a very large menu tree to traverse.
Raskin's environment involves breaking a program like a wordprocessor apart so that there is no monolithic application - just a document handler and a bunch of small command that operate on the document - ie he is bringing the UNIX philosophy to the GUI world. Since all the commands are issued at the system level (like in unix) as opposed to the application level (like in Word) there will be a large number of commands in a single namespace (like in unix). Therefore shortcut keys are not acceptable, and both emacs-like hyroglyphics and vi-like modal interfaces have their own problems.
The solution he proposes can be thought of in two ways. One is to look at it a shortcut keys of infinate length. IE addition to the 30-odd single letter shortcuts possible on a keyboard (^c ^x etc) you can also type ^ls or ^repaginate. The Caps Lock key is not really a Caps Lock Key - it is a command key that just happens to be where the caps lock Key is on PC keyboards (BTW as any unix guru will tell you, that is where God intended the control key to be).
Another way of thinking about it is that holding down the command key brings up a floating command prompt, where you can type your command. Once you release the command key the command is executed.
This seems a little weird but it isn't that hard to type with your pinky held down (I DO IT ALL THE TIME WHEN TYPING ALL CAPS), and if someone does have difficulty, then special hardware like foot pedal, keyboard with command key under the thumb etc, or in worst case making the command key sticky (toggled) will easily solve the problem.
>>Umm... correct me if I'm wrong... but wouldn't it more or less run out of the box?
Nope. Not until you glue on a LEAP(tm) key and install a SwyftCard.
I am a big fan of The Humane Interface; I reviewed the book in its early phases.
Raskin is a big fan of buttons, as long as each button does exactly one thing. He says that the best way to use a computer is to develop habits, so that you can do things without thinking about them.
That works best when things are incredibly consistent. Modes are the enemy of habits; you have to remember that in this context the right button does X, but in that content the right button does Y.
He goes for something he calls "quasimodes", where you press and hold a button to temporarily and actively shift into a different mode. You only have one mouse to do a lot of gestures, but you want to press and hold a "zoom in" button rather than clicking into a "zoom in" mode and then clicking out.
The theory is good, but I was never completely comfortable with the idea. It seems to create rather a proliferation of buttons, and new applications can't add new buttons to your keyboard. His ideas are heavily centered around everything being a word processor or spreadsheet, and I have a hard time adapting his ideas to applications that are basically forms instead. Those cases are heavily modal: typing in one field means something very different from typing in another field.
I've been reading "Revolution in the Valley" from OReilly (a really fun read, I highly recommend it) and really, Jef had very lettle to do with the Macintosh interface. Steve Jobs forced him out very early on into the Macintosh development process - well before they had a working prototype. In fact, Jef's vision of the Macintosh was much more text-based than what we know the Mac to be today. Jef's vision of the Macintosh looked more like the Osborne 1 than the Banana Junior 2000 form factor we're all familiar with. The interface borrowed heavily from the Apple Lisa for which Bill Atkinson developed.
Jef was a music major by training, so while I still respect him and what he's done, it's not like he was formally educated in the field.
You can read about this at folklore.org as well but the book is great.
Raskin is "the man" for UI?
In a word; no.
And, if you don't believe me, check out the Canon Cat. Really.
Post Mac, and has NOTHING that is on your standard UI list. Big (BIG) flop.
Check out Raskins ramblings -- boils down to "The UI should be vi; and people will love it". Especially, vi with dedicated function keys.
In a sense, he *is* right. It would be a better UI. But, he *is* very wrong; people will *not* love it. So its a non-starter.
Ratboy.
Just another "Cubible(sic) Joe" 2 17 3061
I'm sure Mr Raskin is an excellent man. He appears to have a solid reputation, and I certainly wish him no ill.
However, much to my bemusement I received an e-mail nastigram from him three or four years back requesting an open source project I was (and still am) in charge of change it's name. My project had a resemblance in name to a *patent* he'd registered some years ago. My projects in an entirely different field of computing: he had issue with my software name clashing with the name of an operation within a patent. It was - in my view - utterly groundless.
So I ignored it. I never heard any more from him. It helped too that I was neither in the same country to him, privy to the same laws as him, and that ignoring such things usually does help them go away when individuals are involved. None the less, Mr Raskin was implying lawyers. I always worry when lawyers are mentioned.
Still, in the grand scheme of things, having veiled legal threats from a co-creator of the Mac (of which I'm a big fan) is an interesting experience to talk about over a beer.
I wish him all the best, but I do hope he isn't still firing nastigrams off to open source developers.
In Jef Raskin's world, a world that will undoubtedly become everyone's world someday
I like good ideas. I like good thinking. I like good implementations.
I don't like when somebody tells me about something being in its (not so early) infancy that this will be your way of doing things. Let me decide that one. Thanks.
I am putting myself to the fullest possible use, which is all I can think that any conscious entity can ever hope to do.
To answer the specific questions, another feature of this system is that all documents are saved automatically. Think a palm pilot, where you just edit your document and turn it off, then when you turn it back on everything is just like you left it in the real world. There is also global undo (works for any command - if not you are prompted before command is executed) which is saved between power cycles. There is also talk of having an automatic revision history so you can come back to a document at any time later, and revert to a previous version (this isn't infeasable - the VAX had a crude mechanism like this way back when).
:)
The quit command is to exit the entire environment (ie to lougout). There is not quit command for a document - when you are done editing there is nothing special that you need to do just go on to your next task.
As to the global namespace, it may or may not be problematic (and actually calling it a global namespace might have been inacurate on my part). I don't know how they are actually implementing it (I am not involved in the project just read his book and have been watching from the sidelines). From what I have seen though, the commands will be able to do different things to different objects. For example, if I select some text and ^copy it will copy the text, if I select 5 whole documents and ^copy it will copy the documents. I can't imagine that they would attempt to write a single command that handles all the possible object that could act on.
The way I see implementing it is that each document type has a document handler class that provides the direct manipulation interface, as well as a programatic interface to manipulate the data. Then commands are written using this programatic interface. When the user issues a command it is the same as sending a message to a smalltalk object - if the command is recognised for that document it gets executed, otherwise you get an error. With this approach different document types could have commands with the same name, and it would not conflict.
However, from a user-interface point of view, if two documents types support a simular opperation, it is highly desirable for them to share a command name (so you don't have to remember to use ^find on one document type, ^search in another). From a technical point of view, this can all be done very easily with a late binding language like python, and sharing commands between document types can even make development easier - if the people work together. So the "single" command space creates a social problem not a technological one.
Which makes me sceptical of Raskins claims that this system will work well with comercial companies. To begin with, they don't like the idea of being demoted to writing commands, as opposed to developing full applications. This is one of the main reasons that OpenDoc died. Secondly, what happens when Alias writes a set of 3D modeling commands whose names conflict with Discrete's set of modeling commands? What if Maya doesn't like the 3D document handler and writes it's own incompatible one, with incompatible tools? You are back to where you started with walled off applications that don't integrate into the rest of the system, and potentially even conflict with one another, defeating most of the purpose of this new architecture.
To me is seems that this project will only work if it is managed as a coherent whole, like BSD or Squeak, and that means being open source with a strong leader. And now that I've gotten completely off-topic of your question I'll end my post
Would I be trolling if I say that I think Jef Raskin is totally overrated? He likes to promote himself as the "creator of the Macintosh" and an expert in optimal user interfaces, but let's remember that he opposed the use of GUIs, and believe that the "optimal" user interface involves chording combinations of arcane keystrokes. Just read the description of Raskin's [url=http://www.jagshouse.com/swyft.html]Canon Cat,[/url] then compare it to your favorite user interfaces, and realize how way off-base Raskin is.
To be fair, Jef does have some nice ideas, such that a computer should turn on instantly, and that commands across different applications should be consistent. But hey, we've already got [url=http://www.apple.com/ibook/]computers that do that.[/url]
The worship of Jef Raskin as some sort of unparalleled visionary has no basis in reality.
--R.J.
Electric-Escape.net
That is not the real problem, though. The real problem is triangle trading schemes that let corporations sell products to themselves at a "loss" so they can claim they made no money. Almost all multinational corporations do this; it's no secret.
In case you aren't familiar with the scheme, the multinational company has subsidiary X in the US, its main headquarters. In some third world country, they have subsidiary Y, which produces, say, tennis shoes. Then they have subsidiary Z, a tiny, unofficial office in the Virgin Islands. Subsidiary Y sells the shoes to subsidiary Z for $3 a pair. Then subsidiary Z sells those same shoes to subsidiary X for $50 a pair. But since subsidiary Y is not officially part of the multinational, so it appears the company is LOSING $47 on each pair of shoes. They sell them in the US for $97 a pair, and the net balance is zero. No taxes to pay. Or in some cases, when there is a negative net balance, they ask for bailout money from the government (and that money sure didn't come from taxes the corporation paid).
You can easily imagine a company using subsidiary Y in the opposite way, to artificially inflate corporation income if necessary to meet Wall Street's expectations.
Si la vida me da palo, yo la voy a soportar Si la vida me da palo, yo la voy a espabilar