Imagine a similar light-modulation encoding scheme where your car's current speed, vehicle ID number, and other specific information is transmitted from headlamps and taillights.
Imagine that this encoding and transmission was required by law and monitored by police and traffic authorities. Now our cars are tattle-tales and the need for fancy radar monitoring is greatly reduced. Kinda scary thought if you don't like speeding tickets! I suppose this would be hacked eventually.
But the same technology could also be useful for "convoy-ing" vehicles in an auto-pilot fashion. Perhaps light is not the best signal for this (radio or IR more reliable?). But wouldn't it be nice to not have to constantly tickle the accelerator and brake in stop-and-go driving? The car in front of you would tell your car that it was slowing down or speeding up, and yours could automatically adjust. Throw in a proximity detector as a failsafe and you are partway to an auto-pilot.
I've read this book when it first came out. I think it is great! Well-written and thought provoking. It makes me want to try a Cannon Cat, a relatively successful product that Jef shaped. I've thought about programming a modern user environment with these ideas, but I haven't much time to devote to it. Perhaps someone out there does!
But my biggest question after reading and agreeing with much of the book is: How the hell does it apply to business applications? I can see how Jef's ideas work well for content-creation like writing, emailing, and drawing. I don't see yet a good way to apply LEAP and non-modal interfaces to typical repetitive business applications like order entry, general ledger, and so on... especially in a networked, enterprise setting. You can't possibly have dedicated keys for every application function, and the "highlight command and press execute" gets you back to a variation on command-line interfaces or menus. Modality or navigation inefficiencies seem to creep back in!
Any thoughts on this among other readers of _Humane Interface_? Jef, are you lurking?
--
Mike
I have read Jef Raskin's book once and am currently re-reading sections of it. I like the way he shows how user-testing data and scientific analysis can determine which UI characteristics and features really help both novice and expert users. The current and popular trend to support skinning and customization of the desktop and applications in Win9x and Linux goes against usability and efficiency, and Jeff's book will show you why. I think that customization trends feed on growing boredom with computer-based work: people find it more entertaining to tinker with their desktop than do something productive.
I wish more systems software and applications would be designed with Jef's level of attention to ease of use, safety of data/content, and consistency. The book very frequently returns to the Cannon Cat for examples of a Humane Interface. From descriptions, this early-to-mid 1980's "information appliance" took Jef's early ideas to the max, and offered very integrated ways to do word processing, calculations, email, and other functions in a consistent and forgiving environment. I wish I could have tried one out.
I'd love to see a new incarnation of the Cannon Cat using modern hardware and the new insights Jef and others have gained over the last 20 years. -- Mike Losh
It's like trying to build a skyscraper with a hammer and a saw.
Why do so many people and organizations want to build skyscrapers? Metaphorically speaking, the needs of many organizations and people are better met with a tent, or a hut, or a simple house. People waste so much time trying to save time, but when you are building a skyscraper, its complexity and sheer size make it difficult to do efficiently.
Steve M. had written some excellent books. I have not read ATGR, his earlier SPSG tried to strike a good balance between rigor and efficiency. I think that a better and more realistic balance is described in Kent Beck's Extreme Programming Explained. In this book, they tell you "You ain't gonna need it" when it comes to adding extra generallity (you might change your mind later, or have the project cancelled). Think about it before you jump into building a new skyscraper! -- Mike
I concur. In May I ordered a new PC. I received the new Microsoft Elite natual keyboard despite my order for another type. I tried it for a few hours. Awful! The arrow keys are in a one-two-one diamond pattern (not the familiar inverted-T), and the Ins-Del-Home-End-PgUp-PgDown keys are laid out in a two-by-three pattern, not the familar three-by-two. I could not adjust, and quickly sent it back. I'm on a traditional-layout genuine IBM-brand 'board now.
Slashdot Mirror
This is at New Deal, Inc. -- Mike Losh
Imagine a similar light-modulation encoding scheme where your car's current speed, vehicle ID number, and other specific information is transmitted from headlamps and taillights.
Imagine that this encoding and transmission was required by law and monitored by police and traffic authorities. Now our cars are tattle-tales and the need for fancy radar monitoring is greatly reduced. Kinda scary thought if you don't like speeding tickets! I suppose this would be hacked eventually.
But the same technology could also be useful for "convoy-ing" vehicles in an auto-pilot fashion. Perhaps light is not the best signal for this (radio or IR more reliable?). But wouldn't it be nice to not have to constantly tickle the accelerator and brake in stop-and-go driving? The car in front of you would tell your car that it was slowing down or speeding up, and yours could automatically adjust. Throw in a proximity detector as a failsafe and you are partway to an auto-pilot.
Enough random thoughts for today...
Mike
I've read this book when it first came out. I think it is great! Well-written and thought provoking. It makes me want to try a Cannon Cat, a relatively successful product that Jef shaped. I've thought about programming a modern user environment with these ideas, but I haven't much time to devote to it. Perhaps someone out there does!
But my biggest question after reading and agreeing with much of the book is: How the hell does it apply to business applications? I can see how Jef's ideas work well for content-creation like writing, emailing, and drawing. I don't see yet a good way to apply LEAP and non-modal interfaces to typical repetitive business applications like order entry, general ledger, and so on... especially in a networked, enterprise setting. You can't possibly have dedicated keys for every application function, and the "highlight command and press execute" gets you back to a variation on command-line interfaces or menus. Modality or navigation inefficiencies seem to creep back in!
Any thoughts on this among other readers of _Humane Interface_? Jef, are you lurking?
--
Mike
I have read Jef Raskin's book once and am currently re-reading sections of it. I like the way he shows how user-testing data and scientific analysis can determine which UI characteristics and features really help both novice and expert users. The current and popular trend to support skinning and customization of the desktop and applications in Win9x and Linux goes against usability and efficiency, and Jeff's book will show you why. I think that customization trends feed on growing boredom with computer-based work: people find it more entertaining to tinker with their desktop than do something productive.
I wish more systems software and applications would be designed with Jef's level of attention to ease of use, safety of data/content, and consistency. The book very frequently returns to the Cannon Cat for examples of a Humane Interface. From descriptions, this early-to-mid 1980's "information appliance" took Jef's early ideas to the max, and offered very integrated ways to do word processing, calculations, email, and other functions in a consistent and forgiving environment. I wish I could have tried one out.
I'd love to see a new incarnation of the Cannon Cat using modern hardware and the new insights Jef and others have gained over the last 20 years.
--
Mike Losh
It's like trying to build a skyscraper with a hammer and a saw.
Why do so many people and organizations want to build skyscrapers? Metaphorically speaking, the needs of many organizations and people are better met with a tent, or a hut, or a simple house. People waste so much time trying to save time, but when you are building a skyscraper, its complexity and sheer size make it difficult to do efficiently.
Steve M. had written some excellent books. I have not read ATGR, his earlier SPSG tried to strike a good balance between rigor and efficiency. I think that a better and more realistic balance is described in Kent Beck's Extreme Programming Explained. In this book, they tell you "You ain't gonna need it" when it comes to adding extra generallity (you might change your mind later, or have the project cancelled). Think about it before you jump into building a new skyscraper!
--
Mike
kiddybytes for ~1,000.
Manlybytes for ~1000,000!
Giantbytes for ~1,000,000,000!!
OK, enough silliness.
--
MLOSH
I concur. In May I ordered a new PC. I received the new Microsoft Elite natual keyboard despite my order for another type. I tried it for a few hours. Awful! The arrow keys are in a one-two-one diamond pattern (not the familiar inverted-T), and the Ins-Del-Home-End-PgUp-PgDown keys are laid out in a two-by-three pattern, not the familar three-by-two. I could not adjust, and quickly sent it back. I'm on a traditional-layout genuine IBM-brand 'board now.