Jef Raskin's Humane Interface Released

cold wolf writes "With a new site layout and information, the Raskin Center has also just released Archy (formally known at The Humane Interface). It is currently in Alpha phase and Windows only, as an executable."

Comments

Capital letters?

[AtariAmarok]

Shouldn't a proper Archy interface has no capital letters at all?

Anyway, I had enough of the whole cockroaches-in-the-computer thing when I lived in Brooklyn for a while. It gets old pretty fast.

Why?

[0x461FAB0BD7D2]

What is the point of this? After at least a decade of GUI-based computing, how many users will shift from the familiar windows-based (note the lack of capitalization) to an Archy-based one?

The web, rather than Archy, seems to be the way forward, as that is what most people are used to. Rather than focusing on the latest scientific research, they should have focused on the eccentricities of the everyday PC user.

Re:Why?

[ConceptJunkie]

The web has helped set UI development back 15 years (better graphics and faster machines have done the rest, at least in the Windows world). Something like Archy would eventually encompass the web as part of the data you can work with.

I don't think you're thinking big enough.

Re:Why?

[0x461FAB0BD7D2]

Hmm. Interesting. I always thought the point of technology was not to create the most perfect product out there, but to create a product that users would feel comfortable with.

While theoretically, UI development may have been set back, users are comfortable with the web. Therefore, that is the way technology should progress, i.e. working for the user, not getting the user to work with the product.

I believe that is the motto of Archy itself.

Re:Why?

[ConceptJunkie]

For viewing documents? Sure, the web is great.

As an application platform, it sucks big green minimize buttons. Unfortunately, a lot of apps are deployed on the Web for obvious reasons. MS tried to mitigate this lack of real app development with ActiveX, other people offered real solutions like Java and PHP. None of them however come within the same time zone as a native GUI app for ease of use and good looks. Face it, even Java is still ugly.

The nightmarish part is the a lot of GUI developers (e.g., Symantec, curse their UI people) suddenly thought this backwards hodge-podge of a paradigm was a good idea for real GUI apps. Standards have been thrown out in favor of questionable (and in some cases horrible) "innovation" along with pretty pictures that do nothing but obfuscate the user interface.

Bleh!

Re:Why?

[0x461FAB0BD7D2]

Application development on the web is at best immature. ActiveX, more than the web, pushed back UI development as well as web application development, as it bastardized an open platform.

Java applets had a future, but Sun screwed it up by bloating JREs. Flash had a future, but it is used mainly for annoying ads. Javascript is making a resurgence, but it will fail, as we've seen the amount of exploits written in JS.

The web is a far more promising platform than any other I can think of. It is something many people are comfortable with, while it has the ability to grow, something that the desktop world, entrenched in a window-based environment, can't.

At this point, with the number of users around the world, I can't see Archy taking hold. Most can't be bothered to learn new UIs. It's up to the web to deliver something.

Re:Why?

[hey!]

I always thought the point of technology was not to create the most perfect product out there, but to create a product that users would feel comfortable with.

Open your eyes my friend.

The average person views computers as upredictable, untrustworty, even malevolent. Maybe one out of five users enjoys working with them, and one out of ten has anything approaching mastery. The state of the user interface art is one of crushing mediocrity.

I will admit that much better interfaces can be consructed from the fund of ideas we have today, but it isn't reasonable to expect that a field which has been stagnant for fifteen years or more to be healthy.

Re:Why?

[0x461FAB0BD7D2]

I understand the user's frustration when it comes to computers. After all, they're used to technology just working, after using products like cars, microwaves and TVs.

However, learning how to use computers doesn't come easy to some people. The fact is that even as hard as the IT industry has tried to analogize different parts of the computer to real-life, with things such as files, folders, recycle bins, the fact is that computers are different from anything in the past. It's a general-purpose machine, and not a specialized one like a car or a toaster.

Users use computers like specialized machines, expecting them to process text and download stuff flawlessly. After all, when was the last time your type-writer tried to recover a document? The fact is that as we try to sync computing to the real-world, we're creating false hopes that the "stuff" in the computer will work like stuff in the real world.

Re:Why?

[TuringTest]

The web is read-only (for end users, at least). Archy would enable them to Write and Edit the web, with the same interface used to write and edit their desktop data.

Actually the web is very text-centric, so the extended textual capabilities of Archy would be welcome.

Re:Why?

[0x461FAB0BD7D2]

I think the fact that you actually posted your response on the web proves your point wrong. With blogs and news sites like Fark, Technocrat and Slashdot, end-users can use the web.

With products like Dreamweaver, OOo and MS Office, users can create their own webpages too. It might have been an exclusive one-way information system in the 90s, but it is definitely no longer so.

Re:Why?

[TuringTest]

I think the fact that you actually posted your response on the web proves your point wrong. With blogs and news sites like Fark, Technocrat and Slashdot, end-users can use the web.

Yes, and you have to learn a different interface for each site.

Remember when you had to use ftp, gopher, mail, BBS... to access the Internet? The web browser created a unified environment for accessing documents, but not for publishing. I see Archy as the unified interface for creating content.

Re:Why?

[TuringTest]

The fact is that as we try to sync computing to the real-world, we're creating false hopes that the "stuff" in the computer will work like stuff in the real world.
That's where Archy is useful - it doesn't try to imitate nothing in the real world, it just looks like a computer. It allows to manage streams of data in a highly efficient way. Of course, you will have to learn it before you can use it (thankfully, the interface is very easy to learn).

The construction of analogies (desktop, files, etc) was good in the 90s to educate people, but now it's time to largely deprecate it.

Re:Why?

[AtariAmarok]

"The construction of analogies (desktop, files, etc) was good in the 90s to educate people, but now it's time to largely deprecate it."

Yet, files are a lot more intuitive and easier to deal with. As someone once said, "a book is a book!". In the e-world, a book is best represented as a file (ebook), not a mushy structure. The same is true of pictures.

There are a lot of other things in Archy that are a lot worse than the "tried and true" methods which have developed over time because they work. The "you don't have to save" is a problem. It does not make things easier, it just replaces save commands with much-more-confusing "undos". "Better be much more careful when you type that letter! It is automatically saving, and you will have to use a tedious undo feature to undo any mistakes you might have made".

Other problems include enforcing someone's "morality" at the expense of the user's preference. Their principle of " giving you only one way to accomplish a task" only works if you happen to prefer that one way. Flexibility is much friendlier.

Under their "Train of Thought" part, they list 5 bullet items of problems with the non-Archy approach. The first two have already been pretty much solved on the Mac and in sophisticated office suites. The third, about someone's dialog box, is just a problem that happens sometimes in poorly designed programs. #4 (multiple ways to do things) is a strength, not a problem. #5 is more complicated: sometimes there is a very good reason for the commands to do different things, and sometimes there is not.

Re:Why?

[AtariAmarok]

"Remember when you had to use ftp, gopher, mail, BBS... to access the Internet?"

Why use a browser for non-browsing activity???? Eudora does good for mail, and there are many good FTP programs.

Re:Why?

[FidelCatsro]

I have all the knowlidge i need to construct myself a mode of transportation similar to a car .. It wont be nice ,it wont be pretty , and it wont work like its ment to ... just because people have the Means and perhaps some of the knowlidge it does not mean they have the skills and it definantly does not mean its the best way to do it.
We only dont need to reinvent the wheel because its round , however if it were octagonal or square then reinventing the wheel may have been a good step forward

Re:Why?

[Anonymous Coward]

> The web is read-only ... he says, on a web board

Re:Why?

[AtariAmarok]

" The web is read-only ... he says, on a web board"

No real human could possibly say what he did. I think "TuringTest" as failed the turing test.

Re:Why?

[eraserewind]

The web has helped set UI development back 15 years

If the web is so bad, why can everyone and their dog can surf it, but the family expert gets called in to do anything related to a using a UI application.

Re:Why?

[AtariAmarok]

"The web has helped set UI development back 15 years"

You might be onto something if you replaced the word "back" with "ahead".

Re:Why?

[ConceptJunkie]

Because the Web offers a far more restictive set of functionality, primary focused around reading documents.

Just like a car is easier to use than an airplane. Almost anyone can drive a car. Flying a plane takes a significant investment in training and study.

The Web is easy to use because, for the most part, it only has butt-simple functionality. For what it does, what functionality is largely sufficient, but for full-blown app development it is hopelessly limited.

Re:Why?

[Tim Browse]

And you could search with Archie...hey, wait a minute...

Re:Why?

[eno2001]

Who cares about what the users want? They can barely keep their asses clean, and you want to build a UI around what they want??? This is a UI that give people what they NEED. There is an important distinction made here. The machines are tools. They are supposed to make life easier for people. The best way to make life easier for some is to give them what they need, not what they want. An alcoholic might want booze, but what he needs is coffee and a good kick in the ass and constant reminder of why he is a failure as long as he drinks. This is what a good UI/machine will do for the user. The user may not like it, but that's their tough luck. ;P

Archy: An Introduction

[smittyoneeach]

Imagine a system where you can send email, write a book, make calculations, manipulate pictures, navigate the Web -- do whatever you want to do anytime at all, without having to switch in and out of applications. Imagine a system that never loses your work or buries it in a maze of folders, a system that doesn't wrestle with you on your way to getting something done, a system that effortlessly boosts your speed and productivity by 20 percent or more.
That's Archy. It's the answer to a host of problems that have made you mistrust, and at times hate your computer. Up until now, you've blamed yourself when your computer went off the rails. Guess what? You were right and your computer was wrong.
For two decades now, the graphical user interface -- or "GUI" (pronounced "GOO-ey") -- has been the de facto standard for human-computer interaction. But researchers have known for a long time that GUIs are inherently flawed. Nevertheless, this gooey environment has reigned supreme for so long that we've come to accept it as normal and necessary. Up until now we've had no choice.
Now we do.
In his book, The Humane Interface, Jef Raskin -- creator of the Macintosh project at Apple -- said, "Creating an interface is much like building a house: If you don't get the foundations right, no amount of decorating can fix the resulting structure."
When Jef began designing Archy, he didn't try to tweak or tinker with the GUI interface. He didn't try to decorate it. He cleared the blackboard and built a system from the ground up, giving prime consideration to the latest scientific research on human cognition.
The result is a new user interface that looks and feels completely different. Where your current computer still demands that you conform to its way of doing things, Archy adapts to your way of doing things, the humane way.
The principles behind Archy's design are applicable to all kinds of information appliances and the machines that depend on them. Today that includes aircraft, automobiles, scientific instruments, and industrial machinery. In this sense, Archy Alpha Release 1 is the beginning of a movement. Our long-range goal is a world where enlightened user interface design -- taking account of our limitations and taking advantage of our natural abilities -- becomes the standard. Our first product demonstrates that computers can add ease, convenience, power, and efficiency to our lives without adding to the list of our frustrations.

Does Tufts use it?

Re:Archy: An Introduction

[Anonymous Coward]

s/tufts/tufte/

Re:Archy: An Introduction

[Hamled]

Imagine a system that never loses your work

Interestingly, this current version (perhaps they will fix it later on), will happily lose many of the paragraphs I wrote before I told Windows to kill the process... Even after letting the text sit there for several minutes, none of it was ever saved. I'd expect a system designed to "never lose your work", to do some level of automatic saving. In addition, the application uses staggering amounts of memory, simply browsing up and down a text file, it ended up using over 100MB!

Re:Archy: An Introduction

[ConceptJunkie]

Can you say "alpha"?

Expect problems. They are just trying to let you see what it will be able to do.

Re:Archy: An Introduction

[Anonymous Coward]

If there are problems now, who's to say other problems aren't going to show up in the final release?

AppleDoc rehashed?

[obrienb]

Is this really anything new or just the latest version of the old AppleDoc concept?

Re:AppleDoc rehashed?

[pavon]

I think I am correct in assuming that you are referring to the OpenDoc system, which both Apple and IBM worked on.

This project shares several technical similarities with OpenDoc but it has a significantly different goal. The purpose of the OpenDoc system (and OLE, COM, ActiveX, .NET, KDE KParts, and GNOME Bonobo after it) was to reduce duplicate effort on the behalf of programmers by implementing high level document objects which could be easily reused in applications. This would mean that similar tasks in different applications would be done the same way, because they were using the same code, but you would still have applications.

Now if this method of programming became widespread, applications would basically just be a bunch of documents object glued together with extra features added - commands and tools that operate on the document object. This project would do away with applications all together, by creating a system framework and graphical shell that provides the "glue" that applications used to. To compare the difference consider a library written in C vs. a set of UNIX command line tools. Both implement the same functionality, both can be save the programmer time in implementing a task, and both can be flexibly combined to do new things. However to you the C library you have to be a programmer, and have to create a library write a full program with its own interface, and finite feature set. With the command line tools, you can right a full program with them (a shell script), but you can also issue the commands in real-time from the shell. With the C library, you have to do extra work to make the application extendable (your own plug-in system), with the command line tools, the ability to add new commands and recombine them in new ways is built into the shell.

The UNIX shell has limitations though. Its interface and framework (stdin, stdout, pipes, redirect) is primarily useful for filtering documents, not for interactively editing them. Which is why all the unix editors are basically monolithic applications (with their own plug-in and scripting systems).

The goal of this project to make an operating environment that is more pleasant and powerful to use. It intends to do so by making a graphical shell and system framework that will allow for the developer to make software that consists purely of documents and tools that operate on documents, with no applications to wall things off.

I hope this explanation is useful - it is a bit abstract, but the details could fill an entire book. In fact they did :)

Re:AppleDoc rehashed?

[n1ywb]

Don't forget the Xerox PARC Alto before ALL of those. Alto was highly document-centric.

Integrated applications -- This industry buzzword has been used to describe many things; here it means that text, graphics, tables, and mathematical formulae are all edited inside documents. In many other systems, different types of content are edited in separate application windows and then cut/pasted together. For example: a MacDraw drawing put into a Microsoft Word or Aldus PageMaker document can no longer be edited; rather, the original must be re-edited with MacDraw and then substituted for the old drawing in the document.

Not even Star is fully integrated in the sense used here. For example, though the original structured graphics editor, the new one (see History of Star Development, below), and the table and formula editors all operate inside text files, spreadsheets and freehand drawings are currently edited in separate application windows and transferred into documents, where they are no longer fully editable.

Re:AppleDoc rehashed?

[Anonymous Coward]

I think he's talking about OLE2. The whole point of it was to allow in-place editing. When you double-click on the Excel chart in a Word document, the menus change to Excel's commands and the Excel chart is fully editable.

dom

looking good...

[Anonymous Coward]

...but how will it accommodate existing apps? I think a more practical approach would be to write a window manager (on X based systems at least). Window managers like Ion, TreeWM and WindowLab are pretty interesting...

(")Revolutionary(") Ideas

[pavon]

Sweet, I'm going to have to boot into windows when I get windows when I get home to try this out again. I have been following this project from the sidelines for a while now ever since I read his book, and have to admit to being a little giddy about seeing it actually getting somewhere.

I know that they popular trend on slashdot is to love or hate ideas and people, and that is what most of the posts will be about, but my opinion of Raskin has never been one of idol worship or supreme cynicism of anything visionary. I (false)started grad school a couple of years ago, with all sorts of ideas about how to make computing environments better, more pleasant more powerful, only to find that all my "revolutionary ideas" had already been thought of before, sometimes decades ago, and have sat on the shelf ever since. There was really no fundamental research for me to do - all the ideas had already been thought of, and were waiting for someone to do the grunt work of turning them into a practical working system. I became very disillusioned with what I was doing at school - the whole program seemed like a big sham - everyone pretended as if they were doing meaningful research but not one thesis seemed to be anything more than BS. Because of that, and other personal reasons, I dropped out after one semester.

Raskin was one of many of the researchers who ideas I latched onto. I don't know if I agree with all of his ideas, but really want to seem them attempted in something more than a simple proof-of-concept. Universities are not interested in practical grunt work - even if it is pushing the boarders. The huge amount of risk involved in creating an operating environment to compete with MS, not to mention the fact that the ideas are still just ideas, means that no one would dare take this on as a business venture. It seems that the open source community is really the entity most capable of doing projects like this.

Right now the project has mainly focused on the text-editing portion of Raskins ideas, which while interesting, are for the most part a known quantity - they are an incremental improvements on the ideas used in the Canon Cat. What I am really interested in is how they can be expanded to a system environment. For those that haven't read about him, he talks about a computing environment where there are no applications, just documents and tools that act on documents. This would create an incredible amount of flexibility, as is effectively bringing the Unix philosophy to the GUI world. Or alternately it takes the plug-in, undo and scripting functionality that the most powerful applications have and bringing it to the system level, so that everything has those features "for free", and they all interoperate for free, since you don't have a bunch of applications each with their own different, incompatible and likely proprietary methods. You now just have the core document objects, and a bunch of small tools that interface the document object. Apple's CoreData also has me really interested as it seems to implement many of the technical requirements that I have concluded such a system will need.

My other half keeps reminding me that all the attempts at wonderful unified systems have failed, and that it is ugly systems that are good at gluing together disparate, but existing technologies that succeed. But I don't care. I still would like to see it tried even if it does fail.

Re:(")Revolutionary(") Ideas

[vmcto]

Or alternately it takes the plug-in, undo and scripting functionality that the most powerful applications have and bringing it to the system level, so that everything has those features "for free", and they all interoperate for free, since you don't have a bunch of applications each with their own different, incompatible and likely proprietary methods.

Not trying to be a troll... From this description it sound like it's the inverse of Linux from a desktop user experience persepctive?

Re:(")Revolutionary(") Ideas

[TuringTest]

It's the same as the Unix command line. A single interface, where all available commands interact at user's whish through a unified syntax. But in Archy the syntax is more powerful than in Bash.

Re:(")Revolutionary(") Ideas

[pavon]

Yeah, Linux on the desktop follows the same basic WIMP model just like Windows and Mac OS. The problems are not exclusive to Linux. There are tons of "redundant" applications for windows each of which with its own selection of features, and incompatable ways of doing things. It is just more visable in Linux Systems, because alternate applications have more possibility of being used, unlike in Windows and Mac where a one appication became the dominant defacto-standard for each task, for economic and interoperability reasons related to the proprietary nature of the applications.

Which does bring up something that I think I've commented on before. Even the open source community is not perfecly suited for this kind of project because it is very much a cathederal type project (that will hopefully create a bazaar environment when done). It will requires a level of cooperation and integration that is difficult to achieve in a situation where the standard operation procedure is to just implement what you think with improve the project, and if the project maintainers disagree then call them Nazis :)

Re:(")Revolutionary(") Ideas

[dr.badass]

You, sir, have just made my friends list.

My other half keeps reminding me that all the attempts at wonderful unified systems have failed, and that it is ugly systems that are good at gluing together disparate, but existing technologies that succeed. But I don't care. I still would like to see it tried even if it does fail.

Revolutionary new interfaces suffer more from being new than being revolutionary. New interfaces are never perfect -- they must evolve over time. You have to give yourself a wide berth to be wrong, and the worst way to do that is to start from scratch with a rigid plan.

There's also the fact no matter how unified your desktop interface is, it still has to interact with a world that doesn't use the same. I think that most "thesis UI's" fail because bridging them with the rest of the world's tools winds up feeling even more hackish than the interfaces they're intended to replace.

I tend to think of Gelernter's LifeStreams. It's a very elegant and interesting concept, with lots of good ideas. The trouble is that it doesn't work unless everybody else is using it, or if you limit it's functionality to things that work with it's metaphor. That is, when you try to communicate with the world that doesn't use it, it starts to suck. (The commercial implementation is a pale reflection of the grand idea.)

I basically agree with you, but I think the approach to such an interface has to be mostly evolutionary -- instead of dictating a unified environment, create a framework that encourages (and indeed, attracts) developers and users to adopt something closer to that ideal form.

Re:(")Revolutionary(") Ideas

[mrsbrisby]

Sweet, I'm going to have to boot into windows when I get windows when I get home to try this out again. I have been following this project from the sidelines for a while now ever since I read his book, and have to admit to being a little giddy about seeing it actually getting somewhere.

You don't have to! It runs fine on Linux right now. Go download the source and run it. Just have pygame and a couple other python extensions installed and it works fine.

Karma Whoring and Lazy Linking

[fm6]

It'd be easier to just send us to the Archy Introduction instead of copying the entire page onto Slashdot.

And that link should have been in the story itself, instead of a link to the download page. You want to read something about a program before you go to all the hassle of downloading and installing it.

Re:Karma Whoring and Lazy Linking

[Anonymous Coward]

Agree with your second point, but your first is not true for all values of 'easy'.

Re:Karma Whoring and Lazy Linking

[BigJimSlade]

You want to read something about a program before you go to all the hassle of downloading and installing it.

You must be new here...

Stupid

[Anonymous Coward]

This interface is awful. I remember checking it out a long time ago. Nothing has improved.

Imagine a text console running a program that is a cross between EMACS and VI (at the same time). It's wide, flat, hard to use, cryptic, etc...

Ugh, it feels like something that came out of 1970's mainframe computer science.

Re:Stupid

[TuringTest]

Yes, just like if it were a command line...

Re:Stupid

[FidelCatsro]

I currently can't try archie , though i intend to the minute a version is released for Mac ,linux or BSD.
Simple reason as to why its hard to use , well a two fold awnser
1: its alpha
2: its a completly new way of working and still well into development.

Its an experiment into the best user enviroment, It is not like the standerd Windowing GUI you are used to , nor is it that much like shell.
EMACS is a good example and the closest thing to it i have used (remembering i have only read up on the interface not had a chance to use it yet).
I as a computer user really look forward to learning a new interface or way of interacting with my computer and i for one am not entierly sold as of yet on Raskin's interface , But i will give it a fair try but please dont judge it from such an early version and please explain what you find awfull (in more detail)as the developers are probably reading this and the human interface devs are as i recall very open to suggestions about problems.

it is only cryptic i would say as you have only used it for a short while.

Re:Stupid

[Anonymous Coward]

A truely functional interface would be easy to use. If it truely was humane, I could sit down and start using it right away.

Hell, I sat my 4-year-old down and they were using Windows, OS X, or GNOME with a mouse in less than 60 seconds.

This isn't that way. It's like sitting the same 4-year-old down in front of Vi and saying "go to town!" Pfffft, that will never fly.

Note that I actually am a Vi user and I love it to death, but it took a long painful learning curve to realize the full power (which I can use because I'm a programmer/power user). A casual user has no need to expend the effort on such things. This interface requires too much effort to use, it's not humane. Dispite all their humane-this, humane-that, it still feels a lot like a computer nerd coming up with a technical solution to a human problem.

Re:Stupid

[AtariAmarok]

"Dispite all their humane-this, humane-that"

I don't think any minks were slaughtered to create it, were there? Perhaps the "humane" claim is the only one of its wild boasts that really is true.

Re:Stupid

[mrsbrisby]

This interface is awful.

Subjective. I happen find it quite comforting.

Imagine a text console running a program that is a cross between EMACS and VI (at the same time). It's wide, flat, hard to use, cryptic, etc...

I find VI very easy to use, and I know folks who say EMACS is very easy to use. I think you mean it's not approachable, instead of "not easy to use."

Notepad is very approachable, but it's certainly not easy to use. A better way for you to look at this is think: how productive can people be?

Archy promises to combine the things that I think make me most productive in VI, and some of the things from EMACS that I think could make me more productive (some of which are showing up in VIM all the time), and at the same time, help me be more productive still, by blurring the concept of data and documents.

Ugh, it feels like something that came out of 1970's mainframe computer science.

I think you're painfully confused. What you're using now on your computer is something that came out of 1970's mainframe computer science.

Archy is what comes out of 30 years of reflecting on how bad of an idea that was.

Trying it out now

[snorklewacker]

First impressions: really awful.

Forward and back arrows do what you expect. Up and down scroll the screen. Page up and down do nothing.

The mouse, of course, does nothing at all.

Keys you expect to repeat don't. That triple-tap thing holds firm for everything. Even backspace. Even the arrow keys.

Tildes and backticks are impossible to type, they've become control characters.

The cursor blinks frantically and distractingly in not one, but two colors.

To access help, you have to hold down capslock while you type.

I stopped there. Guess it needs a little more time in the oven, but so far it's flying in the face of usability.

Re:Trying it out now

[Anonymous Coward]

No, this is what happens when people develop with blinders on. They purposefully "wiped the slate clean and started over" and by doing that they ignored all the lessons learned over the last 50 years.

This thing is archaic and unusable.

Idiots!

Re:Trying it out now

[TuringTest]

First impressions: really awful.
I find it different, but really interesting. Much better that any other GUI used only with keyboard. *Those* are awful.


# Forward and back arrows do what you expect. Up and down scroll the screen. Page up and down do nothing.
I wrote to Brad Lauster just two days ago on this very point, and he was quite open to commentary. Main movement is supposed to be done through text search, but maybe page up/down will be implemented. Remember that this is in very alpha stage.


# The mouse, of course, does nothing at all.
The mouse will be used for graphic manipulation and navigation in the zooming interface. For text processing, it uses the keyboard. This is the main point that makes habit forming possible.


# Keys you expect to repeat don't. That triple-tap thing holds firm for everything. Even backspace. Even the arrow keys.
All keys do repeat, thank you very much. You have to tap it three times, not press and wait until it repeats. This avoids the error of having autorepeat when you don't want it, and is faster when you do.


# Tildes and backticks are impossible to type, they've become control characters.
This program is a prototype. The final product is suposed to run under dedicated hardware (a special keyboard). Afaik- there is a USB "LeapBar" extension for current keyboards.


# The cursor blinks frantically and distractingly in not one, but two colors.
Maybe you could file a usability bug report? This project is being user-tested. If it is annoying, it will be removed.


# To access help, you have to hold down capslock while you type.
The LeapBar will have a dedicated Command key.
With the zooming interface you could simply "zoom out" until you see the Help section.
And yes, the textual interface is designed for good typists - that's not strange, since it's supposed for advanced text manipulation.


I stopped there. Guess it needs a little more time in the oven, but so far it's flying in the face of usability.
No, its only strange because you have to learn it from scratch. Is not more difficult than a current WIMP "point-and-click" interface is for novice users. Actually it's easier - more consistent, more simple, you don't have to think what you want to do in advance.

Re:Trying it out now

[TuringTest]

Actually they've learned the lessons from the last 50 years, and they're trying to solve the problems. Mainly the big problems from the last 20 years' GUIs.

Re:Trying it out now

[nate nice]

"Tildes and backticks are impossible to type, they've become control characters"

Hmm, it's going to be hard (even harder I should say) to write C++ programs that don't have memory leaks now.

Re:Trying it out now

[snorklewacker]

Is not more difficult than a current WIMP "point-and-click" interface is for novice users. Actually it's easier - more consistent, more simple, you don't have to think what you want to do in advance.

So let me get this straight: it requires special hardware to navigate properly and fails to follow conventions that have been around since electric typewriters themselves, and you're telling me that I am the one that doesn't get it?

Your arrogance makes me sad, because it will condemn this whole project to irrelevance, sinking whatever good it did have to offer along with it.

Re:Trying it out now

[brunes69]

No, its only strange because you have to learn it from scratch. Is not more difficult than a current WIMP "point-and-click" interface is for novice users. Actually it's easier - more consistent, more simple, you don't have to think what you want to do in advance.

I am really wick of this argument - namely, because there *are* no users anymore who don't know the WIMP interface. It is what everyone in the world has used for the past 30 years to control their PC. The ship has sailed guys - seriously. The mouse has drawbacks, but not enough to justify the cost required to adopt a non-WIMP interface - and there is always a cost. Even if you are somehow a novice who has never used a WIMP interface before (who is this again? Two year olds are using WIMP these days...) you would still be at a disadvantage, because even if you were marginally more productive with this interface, the fact that you are now incompatible with the rest of the world makes it not worth it. It like someone coming along today proclaiming they have a great new interface for driving a car, that uses something other than a wheel. It doesn't matter *how* good the idea is, or how simple the UI is - the cost to adopt is too great.

Short of whenever we get direct brain manipulation, it is highly doubtful that there is ever going to be any major use of a non-WIMP interface in computing, ever.

Re:Trying it out now

[SA Stevens]

Oh, you won't be allowed to PROGRAM in this brave new world.

Or if you are, it'll be very, very friendly. I.e. forget all about C++.

Re:Trying it out now

[fade-in]

Not C++; it's all about Python

Actually, the choice of an interpreted language for this project is really cool because you can just plug code into your document and run it on a whim, without spending time compiling it.

Though I'm no expert in Python, I guess it is safe to assume that the backtick and the tilde aren't used?

Re:Trying it out now

[nate nice]

They are not used directly in Python, but Python uses C++ very directly. It's essentially built with and upon C++, thus making it useless in this environment.

Re:Trying it out now

[nate nice]

I Soviet Russia, C++ forgets about you!

OK, sorry, I had to do it. I never had a chance and now I don't have to post that again!

Re:Trying it out now

[TuringTest]

I don't think that Archy will totally replace GUIs (the same way that GUIs haven't totally replaced CLIs), but it will provide a model to follow for them. System-wide commands and non-blocking alerts are two benefits that are already implemented in Mac OS X.

Some hard core users, good typists but bad programmers (secretaries, doctors, lawyers, all those who just "don't get" the filesystem and "open and close" operations) will use an Archy-like system because it makes more sense to them (all my data is always accesible and it doesn't dissapear if I turn off the machine? Good!). The rest of GUIs will tend to be more Archy-fied, and mouse will be more used for navigation and less for command invocation.

Re:Trying it out now

[TuringTest]

There's a "Tilde" command that introduces a tilde. It won't be needed with the special Archy keyboard, of course.

Re:Trying it out now

[fred fleenblat]

>> Keys you expect to repeat don't. That triple-tap thing holds firm for everything. Even backspace. Even the arrow keys.

Here's my little wish. I'd like a keyboard with pressure sensitive keys. The idea is:

1. movement keys (arrows, page up/down, backspace) repeat like normal, but repeat faster as you press harder. there is tactile feedback at the points where repeat start accelerating and at the limit point where it won't go any faster.

2. normal keys that correspond to printable characters do not repeat, unless you press somewhat harder than usual. they repeat at a constant speed.

Re:Trying it out now

[Anonymous Coward]

> Here's my little wish. I'd like a keyboard with pressure sensitive keys. The idea is:

Some old electrics had this feature -- some repeating keys would engage repeat if you pressed them harder. There was a noticable tactile feedback to this, not unlike the push-downshift on the accellerators on automatic transmission volvos (and no doubt other cars).

You would likely not want this feature in a keyboard unless it also gave tactile feedback.

Re:Trying it out now

[snorklewacker]

Python has a backtick operator, which stringifies any expression you enter into it.

>>> 1+2
3
>>> `1+2`
'3'

It's hardly ever used ... in fact I believe it's deprecated (though it doesn't warn you)

Tilde is the unary two's-complement negation. You might use it with modules like struct, which gives you C datatypes.

Re:Trying it out now

[mrsbrisby]

So let me get this straight: it requires special hardware to navigate properly and fails to follow conventions that have been around since electric typewriters themselves, and you're telling me that I am the one that doesn't get it?

Close! It encourages special hardware (the way Windows encourages a keyboard with a Windows key, or Emacs encourages a keyboard with separate Super, Meta, and Hyper keys), and reexamines the foolishness of various conventions.

I'm certain many of the people working on the STAR project heard this before- about not following conventions that have been around since electric typewriters themselves, but right now, you're glad they examined, experimented, and improved upon the working environment DESPITE the fact that it is remarkably different than the "accepted" interfaces of the day because your modern WIMP interfaces are the direct result of that research.

Which pretty much leaves the "you don't get it" part, and I'd say that's still right-on.

Mixed first impressions

[n1ywb]

First of all, don't complain about how it's not what you're used to because THATS THE FARKING POINT! It's supposed to be revolutionary! Forget what you think you know and approach Archy with an open mind, for crying out loud!

I do and have done a LOT of typing and I would welcome a powerful text editing system consistant accross platforms that combines the best of all worlds and I think this has a LOT of potential.

The biggest disappointment for me is the non-free license. It's only free for non-commercial use. That's pretty much means it will never go anywhere, IMO. If it was GPL'ed or BSD'ed, then I think it would have a chance.

Re:Mixed first impressions

[AtariAmarok]

"It's supposed to be revolutionary! Forget what you think you know and approach Archy with an open mind, for crying out loud!"

It would be a lot to consider using if not for the colossal blunder of getting rid of the tilde key, which makes it useless for file access. That is badly crippled text editing, if they can't even get the small core typewriter characters correct (incuding the backtick/accent).

Re:Mixed first impressions

[snorklewacker]

That's twice you've mentioned the tilde key for file access.

I don't even have short filenames turned on. I haven't had any problems. Maybe you need to learn Windows post win95 or so.

BTW, Sun invented that naming pattern for PCNFS.

Re:Mixed first impressions

[zero_offset]

I do and have done a LOT of typing and I would welcome a powerful text editing system consistant accross platforms that combines the best of all worlds and I think this has a LOT of potential.

We call those "word processors".

No tilde? Kiss file access goodbye.

[AtariAmarok]

"Tildes and backticks are impossible to type, they've become control characters"

No tilde????? That makes it pretty much useless for Windows file access, with all the c:\progra~1, etc. That's a huge mistake for something that is starting out on the windows platform.

Re:No tilde? Kiss file access goodbye.

[TuringTest]

Not a problem at all, since Archy doesn't have "files".

Don't try it if you have a dual monitor system

[gothzilla]

I'm using dual monitors and Archy sized itself so that you can't fit the window on one monitor. You can't resize it either. I also noticed that the X in the corner doesn't close the program. I had to type "quit" while holding capslock to close it. Definately Alpha but it does look promising.

The tilde means a lot less typing.

[AtariAmarok]

"I don't even have short filenames turned on. I haven't had any problems"

I don't think I have them turned on, either. At least I have never gone to that setting. The tilde makes a lot of files and their locations quicker and easier to type, not to mention shorter.

c:\progra~1

is a lot shorter than

"c:\program files\"

and it becomes even more advantageous for much longer file and folder paths (c:\docume~1 is a lot shorter than "c:\documents and settings")

So you have no problem with it. I do have a problem with the idea of making many common filenames and paths twice as long (or much longer). I was considering trying this, but won't after someone described this glitch.

Re:The tilde means a lot less typing.

[n1ywb]

Archy doesn't even use files in the conventional sense so your argument is sophomoric. Go away.

Re:The tilde means a lot less typing.

[mooingyak]

I don't know what version of windows you're on, but I typically use cd doc* or prog* for 'documents and settings' or 'program files' I always hated typing the ~

Re:The tilde means a lot less typing.

[RevAaron]

Ugh, I'm with you. I hate typing ~n. The more annoying part is when you don't know which number to use in n- it's not like like everything is ~1 and magically works.

He must be using Windows 95/98, which in a DOS shell will list both Program Files and Progra~1 when you do a 'dir.' you can't type "cd \program files" in 9x; you have to use either "cd \progra~1" or 'cd "\program files"' which is also annoying. On NT4/2k/XP it shows up as c:\program files, not c:\progra~1, though "cd \progra~1" works for me on XP.

But yeah, I really doubt anyone else is so worked up about this. Very few folks give even a third of a damn about the decrepid OSes of 95 and 98, especially when 2000 runs better than either of them even on pretty crappy and old hardware (233 MHz p2 w/ 96 MB of RAM or more). Especially in an app that doesn't even use filenames.

wouldn't even run

[100lbHand]

well, be glad that you could get it to start, downloaded and installed it smoothly, then when i try to start the program, it launches up firefox and asks for a bug report.
archy has been humanely deleted.

Is it really as useless as you say?

[AtariAmarok]

"Archy doesn't even use files in the conventional sense so your argument is sophomoric"

Is it really as utterly useless as you say? No files means you can't use it with email attachments (which are files) or digital cameras (which create picture files)....to name two common uses of files. Any so-called "revolutionary" OS idea that is incompatible with such ubiquitous and useful things as email and digital cameras will "go away" for sure.

Re:Is it really as useless as you say?

[snorklewacker]

It's a document-centric system. Attachments are merely part of a document. You might have structured collections (it's how you'd browse your email and digital camera in the first place), but you access things via direct manipulation, not their filenames, much as you generally would in a mac.

Just try it for godsakes, I promise you'll be disappointed, but for different reasons. Your filesystem arguments are really looking sillier by the post.

Seems familiar

[chrismear]

Let's see, what do we have here... a kind of glorified text editor full of cryptic key commands that you have to memorise before you can get anything done, and with a few hooks into a web browser and other apps to give it the vague semblance of an operating system.

Yes, yes, I think we've seen this before.

From The FAQ

[Goo.cc]

Q. Is Archy an application or an OS?

A. Archy doesn't quite fit into the traditional mold of either an application or an operating system. It is an application in that it runs on top of your current operating system, but it is more than that. It's also like an operating system in that it provides a framework for issuing commands.

So, uh, you recreated Emacs?

Re:From The FAQ

[TuringTest]

So, uh, you recreated Emacs?
Yes, but modeless.

Re:From The FAQ

[sydb]

Balls, there's no such thing as modeless. Triple-tapping to repeat doesn't count as invocation of a mode?

Re:From The FAQ

[Anonymous Coward]

Perhaps you're thinking of vi. Emacs does its damndest to be modeless.

Re:From The FAQ

[SteeldrivingJon]

Seems like they should implement it as an OS X input manager.

Re:From The FAQ

[Malfourmed]

So, uh, you recreated Emacs?

Not really...

Q. Archy sounds a lot like a program I use called 'GNU Emacs'. How is Archy different?

A. Indeed, Archy does share a number of similarities with GNU Emacs. Like Emacs, Archy has been designed to allow the user to accomplish a disparate variety of tasks quickly, without leaving the program. Like Emacs, Archy uses commands to manipulate content. Like Emacs, Archy provides a mechanism to navigate a document that is far faster than standard GUI methods.

However, Archy is also very different from Emacs in a number of salient ways.

For one thing, Emacs is rife with modes. In general, a system is in a mode when a given gesture is interpreted in a different way than it is when the system is in a different state. For instance, when you press C-x in Emacs, you put it into a mode whereby the next character you enter is the next key in a hotkey sequence, rather than a letter inserted into your document. Even seasoned Emacs users encounter problems with modes on a daily basis; Archy was designed with a philosophy that eschews modality, because we believe, based on principles of cognitive science, that it is a significant source of errors and confusion in user interfaces. As such, Archy has no modes, resulting in an extremely positive user experience.

Part of the reason Emacs is only really used by computer "experts" is because it's so difficult to learn and habituate to: often a user must learn a huge number of hard-to-remember hotkey combinations to do everyday tasks. Archy has no such hotkey combinations, yet it remains easy to learn and efficient to use because of its reliance on a small set of commands with simple, easy-to-remember names to perform operations.

Similarly, Archy achieves remarkable simplicity by providing a small set of fundamental operations that allow you to accomplish a wide range of tasks with ease. For example, Emacs provides a number of hard-to-remember hotkey combinations that allow the user to move the cursor one character, word, sentence, or paragraph at a time. Archy conflates all of these actions into a single, unified mechanism called Leap, which is similar to Emacs' incremental search, only easier to use and more powerful. To move to the next paragraph, for instance, simply Leap to a carriage return character; there's no need to remember a separate hotkey combination.

There are many other ways in which Archy is fundamentally different from Emacs, but we hope this is enough to whet your appetite. While Emacs and Archy are similar in some ways, they are built on entirely different philosophies, which results in very different user interfaces. We believe that Archy offers at least as much productivity as Emacs, while being far easier and less frustrating to use; we encourage you to test our theory and provide feedback if you are so inclined.

Re:From The FAQ

[TuringTest]

Triple-tapping to repeat doesn't count as invocation of a mode?

No, that's called a Quasimode.

Re:From The FAQ

[sydb]

The inventor of a system with modes invents a new word for those modes so that he can call it modeless.

Re:From The FAQ

[Goo.cc]

You know, I noticed that after I posted. I was wondering if someone was going to bust me!

Re:From The FAQ

[TuringTest]

You could also say that commands are modes, but it's simply not true.

A quasimode is not a mode. The mode is persistent, the quasimode requires a user action to exist and dissapears when the user is not maintaining it. That's a BIG difference with respect to users attention.

Bingo!

[TuringTest]

It's the same concept of Emacs. Must be good, since Emacs is considered by programmers as one of the best development environments.

The main benefit of Archy over Emacs is that it has been engineered with ease of use in mind, not just ease to extend.

Re:Bingo!

[AtariAmarok]

"The main benefit of Archy over Emacs is that it has been engineered with ease of use in mind, not just ease to extend."

Ease of use? Only partially. Some of their "principles" defy ease of use (forced automatic saving which takes word processing back to the BETTER MAKE SURE YOU WANT TO HIT THAT KEY days of the manual typewriter, only one way to do things which can go against user preferences.)

Re:Bingo!

[Anonymous Coward]

some programers,some others prefer to use VI. some others prefer nano and some prefer a massive IDE with more bells and whistles than a referees convention in a cathedrall

Re:Bingo!

[TuringTest]

forced automatic saving which takes word processing back to the BETTER MAKE SURE YOU WANT TO HIT THAT KEY days of the manual typewriter
Ah, but did manual typewriters have an automatic undo? Archy does, for everything (except printers).

Universal Undo is regarded as one of the main advances that a system can provide for ease of use.

Re:Bingo!

[sahala]

Ease of use? Only partially. Some of their "principles" defy ease of use (forced automatic saving which takes word processing back to the BETTER MAKE SURE YOU WANT TO HIT THAT KEY days of the manual typewriter, only one way to do things which can go against user preferences.)

This is only because file-saving is something you've been trained with. It's actually quite nice not having to worry about saving. The auto-save feature is present in IDEA Intellij. Make changes to source, run a build, step back in the undo or local history to revert incorrect changes. I wish more applications did this.

Hello Mr. 1995

[AtariAmarok]

"The web is read-only (for end users, at least)."

For end users? Except for every web user who plays online games, uses Slashdot, participates in blogs, uses online non-AOL chatrooms, purchases anything online, uses eBay, or does banking online.

Other than that, the web is read-only for everyone!

Re:Hello Mr. 1995

[SA Stevens]

What about those of us who use Mozilla's Composer to throw up project pages from time to time?

Re:Hello Mr. 1995

[AtariAmarok]

"What about those of us who use Mozilla's Composer to throw up project pages from time to time?"

That explains something. I always thought that the Mozilla mascot was breathing fire in that splash screen. Now I know better: he's ralphing a really huge ravioli dinner.

Archy = pogo stick. Usual GUI = bicycle.

[AtariAmarok]

"So let me get this straight: it requires special hardware to navigate properly and fails to follow conventions that have been around since electric typewriters themselves, and you're telling me that I am the one that doesn't get it?"

Give it up. It is like arguing with the idiot who commutes to work on a pogo stick who keeps insisting "but it IS better than your bicycle!!!". Part of the reason this analogy is so apt is that you will hardly find anyone using Archy ever, and hardly find anyone using a pogo stick.

Re:Archy = pogo stick. Usual GUI = bicycle.

[QuantumG]

pah-lease, it's more like people complaining that they have to learn how to use an accelerator, a steering wheel and blinkers all at the same time when they're used to just peddling.

Re:Archy = pogo stick. Usual GUI = bicycle.

[AtariAmarok]

"pah-lease, it's more like people complaining that they have to learn how to use an accelerator, a steering wheel and blinkers all at the same time when they're used to just peddling."

The analogy would work better if the car in question was a lot slower than the bicycle.

Re:Archy = pogo stick. Usual GUI = bicycle.

[QuantumG]

That's the whole point of Archy, using your keyboard is a lot faster than fussing around with a mouse. Therefore my analogy is sound.

Re:Archy = pogo stick. Usual GUI = bicycle.

[AtariAmarok]

"That's the whole point of Archy, using your keyboard is a lot faster than fussing around with a mouse. Therefore my analogy is sound."

Sometimes it is, sometimes it isn't. Mature UI's that have had to muster the "real world" of being tested against users take this into account. They offer a mix of mouse and keyboard. They also take into account user preferences: some users want to do things differently than other users. Archy even has a principle against this: they want to straightjacket the users into doing it just one way.

"Step away from the mouse" might be good advice for some users, but not for all. A UI must take into account these different preferences.

Re:Archy = pogo stick. Usual GUI = bicycle.

[QuantumG]

Sorry, I'm an objective realist when it comes to user interfaces. One method must be better than another. The user should have the option of using the best method or not using the system at all. If their preference is to use some inefficient way of doing things they can go use someone else's product (dare I say, Microsoft's). Which, I think, is what this all boils down to. If you want to use the most efficient system, use Archy. If you want to exercise your "preference" to do things some inferior way, don't.

Re:Archy = pogo stick. Usual GUI = bicycle.

[AtariAmarok]

"Sorry, I'm an objective realist when it comes to user interfaces. One method must be better than another"

Sorry, I am not completely up on my UI objectivism. Time to attempt to read those old Ayn Rand masterpieces "The Fountainmouse" and "Linus Shrugged".

Re:Archy = pogo stick. Usual GUI = bicycle.

[fade-in]

You can tell that this project wasn't written in Perl: TOOWTDI doesn't fly in Perl.

Neither does the lack of tildes and backticks, for that matter.

Re:Archy = pogo stick. Usual GUI = bicycle.

[Arkaein]

One method must be better than another.

Care to prove this? I'm serious. If you're really advocating dismissing the majority of existing interfaces for something that has never moved beyond the laboratory/prototype phase you'd better have some way of backing these statements up.

Are there theorems of human-computer interaction efficiency that prove that Archy is more efficient than WIMP interfaces? Have extensive user studies been conducted with test subjects having wide ranges of computer experience?

I'm not trying to rain on your parade here, but lots of people make grandiose claims about revolutionizing just about any field you can think of. Very rarely do these claims amount to much, in most cases because no matter how good an idea seems in theory there usually exists one or more flaws that bring it down in practice. It's up to Archy's developers and supporters to demonstrate otherwise with hard evidence, not just compelling ideas.

Re:Archy = pogo stick. Usual GUI = bicycle.

[QuantumG]

Jesus, yes, that's the whole freakin' point of Archy. It's made by a foundation that does professional usability testing and was created by a man who was a god in his field.

Re:Archy = pogo stick. Usual GUI = bicycle.

[TuringTest]

I don't think that Archy will totally replace GUIs (the same way that GUIs haven't totally replaced CLIs), but it will provide a model to follow for them. System-wide commands and non-blocking alerts are two benefits that are already implemented in Mac OS X.

Some hard core typists but bad programmers (secretaries, doctors, lawyers, all those who just "don't get" the filesystem and "open and close" operations) will use an Archy-like system because it makes more sense to them (all my data is always accesible and it doesn't dissapear if I turn off the machine? Good!). The rest of GUIs will tend to be more Archy-fied.

Re:Archy = pogo stick. Usual GUI = bicycle.

[Arkaein]

I realize that's the point, but has it actually produced any results? Any theorems, validated user tests? I don't know many details about it other than that Jef Raskin supposedly worked on this thing or ideas like it for something like the past 20 years, but seems to have very little to actually show for it. When someone rails against the establishment for that long, they better actually produce something.

As far as Raskin being a "god", I haven't seen it. He had more than enought time to prove the viability of his ideas. Why is it that we hardly see anything that actually works until after he dies? My guess is that any prototypes he's made were flawed in some way, and he's never been able to correct them such that they could pass muster in the public. Something was likely missing in the transition from good sounding idea to practical application.

In any case, if they ever make a version of Archy that runs on Linux I'll be glad to test it. It should be trivial to make it cross platform since it doesn't sound like ti would rely on any standard widgets for drawing, although I suppose the ridiculous keymapping and USB keyboard extension requirements would obviously make it more difficult.

Re:Archy = pogo stick. Usual GUI = bicycle.

[mrsbrisby]

Mature UI's that have had to muster the "real world" of being tested against users take this into account.

I don't think you understand that THE/Archy has been tested with real people. It's the result of quantitative decisions about real work, not about baseless arguments.

some users want to do things differently than other users.

This is ridiculous. Users do things "differently" (in this sense) because current interfaces aren't universally good. If they were, then they wouldn't. Arguing this is futile.

What is helful to argue is whether or not THE/Archy is a universally good interface. I'd say, "not yet", but then even the authors of Archy believe that.

A UI must take into account these different preferences.

Let me guess, you also think productivity went up when Microsoft started offering a candy coated shell for their users...

What the heck is it?

[glebfrank]

OK, can somebody explain what Archy is, exactly? The intro document seems remarkably content free, and basically just talks about how wonderful it is, and how it's going to make you more productive, and enlarge your penis. But what is it all about, really? I don't have a Windows box handy, so I can't try it out...

Wheee!!!!

[Anonymous Coward]

So this is, as it stands now, the crappiest parts of Emacs and the crappiest parts of a Wiki stirred together. In the future they hope to bolt on the crappiest parts of MapQuest....

Profit

[AtariAmarok]

"So this is, as it stands now, the crappiest parts of Emacs and the crappiest parts of a Wiki stirred together. In the future they hope to bolt on the crappiest parts of MapQuest"

1. Mangle Emacs and Wiki together
2. Name it after a cartoon cockroach
3. Talk as if the Mac never existed: "You need to calculate the product of two numbers while you're typing a document, but you don't want to have to launch a calculator program in order to do so"
4. Add MapQuest
5. To complete the business model, revive "Flooz" to make money from it.
6. Profit!!!

Uh...

[Toba82]

yuck. The site doesn't even say what it really is. Unless you can tell me on your own homepage wtf your product is/does, I'm not buying. Thanks anyways...

DOs rehashed?

[Anonymous Coward]

"I hope this explanation is useful - it is a bit abstract, but the details could fill an entire book. In fact they did :)"

ISBN:0471-12993-3

It'll be a pain to get this to work on a Mac.

[NeoBeans]

One screenshot shows an interesting caption.

"To invoke a command, hold down "Caps Lock"...

Isn't the fact that the Mac requires a bit of trickery involving rewriting the keyboard driver to get the Caps Lock key to report it's state like the Control key going to make this paradigm a bit difficult to port to the Mac?

Anarchy

[pudge]

I have no idea really what Archy is -- despite reading a lot of Raskin's writings about THE in the past, I still have no clue, and have ceased caring -- but I am hereby announcing that I will be writing something that contrasts itself to Archy, called Anarchy.

I just wanted to get that out of the way.

Nice try, misses the point

[Julian Morrison]

This has the same problems as any other introspective UI (eg: Squeak), no matter how revolutionary - that it misses the primary point of computer software nowadays. Modern computers run seperately developed software tools, not one big do-everything app. For each of these tools, form follows function. If you shoehorn an MP3 player and a spreadsheet into the same UI paradigm, at least one of them is going to look like an ugly hack. Equally importantly, you'd kill off UI innovation. It's an irony that Archie would have precluded its own development.

Re:Nice try, misses the point

[RevAaron]

In Squeak's defense: you can use it to create your single-purpose apps. And thanks to support for wxWidgets, you can even create native-looking, single-purpose apps in Squeak. Or, you can still use Morphic and do it the cooler way, but either way it's pretty easy to package an app for delivery, having two files (the VM and the .image) as a part of installer/zip/tgz. I think there's a way to even embed the .image file in the VM's exe as a resource, but no one has found that much of a concern for quite a while...

Gelernter's LifeStreams

[Eustace Tilley]

LifeStreams requires Outlook and Office. The overlap between "users of Outlook and Office" and "early adopters of novel ways to use computers" is slim. Targeting that sliver made the whole project seem dimwitted to me.

Easy Interface?

[waynegoode]

Features of the easy interface that I learned in the first tutorial:

• to start the tutorial, press and hold the CAPS LOCK key and type EX1 then release the CAP LOCKS key
• to start a new document, type the accent grave key (`)
• to start a new page, type the tilde key (~)
• to repeat a key, press 3 times and hold the key
• to back up to a previous jump point, press and hold the left ALT key and type the grave accent (`) repeatedly until you are there
• the cursor has two parts, a head and a tail

Thiis is supposed to be easier?

Re:Easy Interface?

[Anonymous Coward]

Exactly my thinking.

Holding down the ALT key while typing is an exercise in torture, and is hardly humane! I watched my fingers contort cruely as I attempted to Leap(TM) around.

And Page Up and Page Down don't work? Huh?

Then try leaping to something that you aren't sure where it is on screen (or off screen). After the leap, you have to look all over the place to find the cursor, especially on a larger screen.

Re:Easy Interface?

[RevAaron]

Precisely. The first time I tried this- before it was renamed Archy, over a year ago- I just thought (rather, really hoped) it was a very well crafted joke on Raskin's part. What's easier than Windows, Icons, Menu and Pointers? Having to memorize textual commands and do some backassword things to invoke them! Maybe Archy will have tab completion one day (I use it all the time in emacs after I hit M-x), but that is still light years from optimal...

Well, it was a cute dream. Maybe next time?

It misses the point of separate applications

[Anonymous Coward]

The whole reason you have separate applications is because different kinds of data work in different ways. There's no way that you would want the same interface to an MP3 player as you would to a photo editor or tax preparation software.

Also, the demo screenshots are all of small documents. There's nothing that shows how you manipulate huge documents. How do you get to page 500 if there are no scroll bars or even page up/down keys? Do I zoom out until the document is just a vertical line and zoom back in on where I think page 500 is?

What's with all these commands I have to know? The typing is cumbersome and there's no way to figure out what commands are available. The great thing about a good GUI is that all options are discoverable. Right now it reminds me of DOS versions of AutoCAD, only with Python as its programming langauge instead of LISP.

Now don't get me wrong here -- combining document types has been a sort of holy grail for ages -- but if it was that easy, somebody would have done it before. How do you handle expanding a spreadsheet in the middle of your text document?

I think that this is just another nice Modernist theory, too obsessed with purity of design to be able to be used to its fullest extent. Just as the way Mac keyboards originally had no arrow keys (I'm supposed to use the mouse to move the cursor one character?), or Java originally had no enumerated types (I'm supposed to type "public static final int" before every item?), I predict that Archy will eventually have to include things like shortcut keys which are currently antithetical to its initial design goals.

And as for this LEAP thing, it definitely looks like a viable idea -- and there are similar implementations like in Plan 9 -- but I think that eye-tracking interfaces will become common-place before LEAP does.

dom

Re:It misses the point of separate applications

[mrsbrisby]

The whole reason you have separate applications is because different kinds of data work in different ways. There's no way that you would want the same interface to an MP3 player as you would to a photo editor or tax preparation software.

You have no idea what you are talking about. The reason you have separate applications is because different people work on them. This has absolutely nothing at all to do with presentation.

The reason tax preparation software and photo editors have different interfaces is because they represent different concepts. Archy is an interface for text processing, and there's nothing about it that says that the Photoshop-killer for Archy will be written in LOGO.

Also, the demo screenshots are all of small documents. There's nothing that shows how you manipulate huge documents. How do you get to page 500 if there are no scroll bars or even page up/down keys? Do I zoom out until the document is just a vertical line and zoom back in on where I think page 500 is?

How do you presently do it? Are you telling me you know you need to get to page 500 without knowing anything that's on page 500, what it might be about, and without any reference near what you're looking at to indicate it?

The LEAP-system means if you know what's there, you can get to it. There's no reason THE/Archy has to preclude hyperlink-style references, and even without them, a table of contents or even a reasonable citation can be enough to use LEAP.

What's with all these commands I have to know? The typing is cumbersome and there's no way to figure out what commands are available. The great thing about a good GUI is that all options are discoverable. Right now it reminds me of DOS versions of AutoCAD, only with Python as its programming langauge instead of LISP.

This isn't true at all. Even presently, in Archy, all the commands are in the commands document. They're all discoverable.

Now don't get me wrong here -- combining document types has been a sort of holy grail for ages -- but if it was that easy, somebody would have done it before. How do you handle expanding a spreadsheet in the middle of your text document?

By not aggregating data-types into documents. Yes, it really is that easy. If you need a table-layout for statistical information, go ahead and insert one. There's no reason at all you have to think in terms of "document embedding".

And as for this LEAP thing, it definitely looks like a viable idea -- and there are similar implementations like in Plan 9 -- but I think that eye-tracking interfaces will become common-place before LEAP does.

I don't think you have any idea how LEAP works. Plan9 most certainly does NOT have anything like it. With Plan9 everything is an interactive hyperlink, so you press the right-mouse button to bounce between documents frequently.

LEAP is much more like the new Mozilla type-ahead-find in that, you hold a key, and start typing and the cursor jumps to the next position having your text. VIM can do this right now if you press the forward-slash in command mode.