When I hear Alan Kay talk dismissively about current applications and interfaces, and the lack of attention given to what was developed at PARC 40 or 50 years ago, I often wish he was more explicit about WHAT was developed. (I have watched the mother of all demos, which is truly awesome, but partial information). This video is another significant chunk, and it puts modern interfaces to shame for their lack of power and imagination. The depth of power here is analogous to the power of Lispy languages, where, until you really understand the concepts, you are ignorant as to how (for example) C++ is in no way "Object Oriented" in the way Alan Kay meant it, how impoverished it is, and how critical late binding is.
You might be interested in Brad Myers' new book "Pick, Click, Flick! The Story of Interaction Techniques". He's a prominent researcher in HCI at Carnegie Mellon University (and one of my colleagues). It gives a great overview of the history of how we interact with computers.
Here's a summary of the book:
This book provides a comprehensive study of the many ways to interact with computers and computerized devices. An “interaction technique” starts when the user performs an action that causes an electronic device to respond, and includes the direct feedback from the device to the user. Examples include physical buttons and switches, on-screen menus and scrollbars operated by a mouse, touchscreen widgets and gestures such as flick-to-scroll, text entry on computers and touchscreens, consumer electronic controls such as remote controls, game controllers, input for virtual reality systems like waving a Nintendo Wii wand or your hands in front of a Microsoft Kinect, interactions with conversational agents such as Apple Siri, Google Assistant, Amazon Alexa or Microsoft Cortana, and adaptations of all of these for people with disabilities. The book starts with a history of the invention and development of these techniques, discusses the various options used today, and continues on to the future with the latest research on interaction techniques such as presented at academic conferences. It features summaries of interviews with the original inventors of some interaction techniques such as Larry Tesler (copy-and-paste), David Canfield Smith (the desktop and icons), Dan Bricklin (spreadsheets), Loren Brichter (Pull-to-Refresh), Bill Atkinson (Menu Bar and HyperCard), Ted Selker (IBM TrackPoint pointing stick), and many others. Sections also cover how to use, model, implement, and evaluate new interaction techniques. The goal of the book is to be useful for anyone interested in why we interact with electronic devices the way we do, to designers creating the interaction techniques of tomorrow who need to know the options and constraints and what has been tried, and even for implementers and consumers who want to get the most out of their interaction techniques.
I’m not sure how this puts things now to shame. Yes it’s very playful and neat, but it’s wildly impractical. It doesn’t really “do” anything by itself. Instead you’ve to build up all interactions, most of which are physical, and know how to do so. And what’s the result? Some spring interactions and markers on a bitmap?
Compare this to how the web or apps are used today. They’re task/purpose driven, and all of the UX has already been thought out. It’s a far more simple and straightforward approach to build the functionality you want which you refine over time than to give a blank canvas that does everything and nothing and tell people to go figure it out themselves.
It seems to me this is really just wanting to take the underpinnings of small talk and turn it into a physical UI representation. That’s fun… but now what? I’m zero percent surprised this hasn’t lasted.
I couldn't disagree more strongly. Part of the vision in PARC (and Alan Kay's life work) was to create learning environments for children, and this demo is set in that realm, so having kids figure it out themselves is enormously empowering. Obviously not what you want if you're trying to sell widgets, but the widget-sellers themselves could benefit from this kind of environment to develop a usable widget-buying experience.
> The UX has already been thought out ... [and] ... refine[d] over time.
In such an impoverished way, in my experience. One of the things Alan Kay has said (elsewhere) is that the environment at PARC allowed them to experiment with radically different user interfaces in the afternoon, after thinking about it at lunch. This implies an in-depth knowledge of the tools, yes, but developers need that in any environment. I'm doing volunteer work now and having to learn to use several software systems (EventBrite, MailChimp, Salesforce). The difficulty of using these systems is in range for me as a retired software engineer, (although endlessly annoying), but other volunteers need extensive training. There's is nothing 'refined' about them, in terms of ease-of-use to the end users. The current tools for developing the interfaces (HTML, CSS, SQL, Javascript, etc.) are also primitive compared to the environment shown here. Again, it's a lengthy task to become a power use of the system demo-ed here, but no more so than to become a power use of the 10 or more technologies you need to learn to make a modern web site.
As a web site developer, with these kinds of tools, you'd develop your own tools and abstractions (an intermediate layer) to build with, none of which would appear anything like the physical world manipulation tools that have been built here.
>I often wish he was more explicit about WHAT was developed. (I have watched the mother of all demos, which is truly awesome, but partial information).
Here's one of several demos Kay has done of the Squeak/Croquet/Etoys/Frank systems, all built to extend on the PARC original. The others are good, too:
There is some nice concepts in here that you can see in some things (like Unreal Blueprints) but the problem it has was you had to use a button for everything, and it just seems a bit clunky.
I would love to see something like this for the web, there's never really been a tool that captures it as close as Yahoo Pipes did.
Reminds me of this one [1]
[1] https://www.youtube.com/watch?v=j_lxBwvf3Vk
I remember when I saw it I wanted to play with it - just seemed like a good way to organize things and 'play' with your desktop
When I hear Alan Kay talk dismissively about current applications and interfaces, and the lack of attention given to what was developed at PARC 40 or 50 years ago, I often wish he was more explicit about WHAT was developed. (I have watched the mother of all demos, which is truly awesome, but partial information). This video is another significant chunk, and it puts modern interfaces to shame for their lack of power and imagination. The depth of power here is analogous to the power of Lispy languages, where, until you really understand the concepts, you are ignorant as to how (for example) C++ is in no way "Object Oriented" in the way Alan Kay meant it, how impoverished it is, and how critical late binding is.
You might be interested in Brad Myers' new book "Pick, Click, Flick! The Story of Interaction Techniques". He's a prominent researcher in HCI at Carnegie Mellon University (and one of my colleagues). It gives a great overview of the history of how we interact with computers.
https://www.cs.cmu.edu/~bam/ixtbook/
Here's a summary of the book: This book provides a comprehensive study of the many ways to interact with computers and computerized devices. An “interaction technique” starts when the user performs an action that causes an electronic device to respond, and includes the direct feedback from the device to the user. Examples include physical buttons and switches, on-screen menus and scrollbars operated by a mouse, touchscreen widgets and gestures such as flick-to-scroll, text entry on computers and touchscreens, consumer electronic controls such as remote controls, game controllers, input for virtual reality systems like waving a Nintendo Wii wand or your hands in front of a Microsoft Kinect, interactions with conversational agents such as Apple Siri, Google Assistant, Amazon Alexa or Microsoft Cortana, and adaptations of all of these for people with disabilities. The book starts with a history of the invention and development of these techniques, discusses the various options used today, and continues on to the future with the latest research on interaction techniques such as presented at academic conferences. It features summaries of interviews with the original inventors of some interaction techniques such as Larry Tesler (copy-and-paste), David Canfield Smith (the desktop and icons), Dan Bricklin (spreadsheets), Loren Brichter (Pull-to-Refresh), Bill Atkinson (Menu Bar and HyperCard), Ted Selker (IBM TrackPoint pointing stick), and many others. Sections also cover how to use, model, implement, and evaluate new interaction techniques. The goal of the book is to be useful for anyone interested in why we interact with electronic devices the way we do, to designers creating the interaction techniques of tomorrow who need to know the options and constraints and what has been tried, and even for implementers and consumers who want to get the most out of their interaction techniques.
I’m not sure how this puts things now to shame. Yes it’s very playful and neat, but it’s wildly impractical. It doesn’t really “do” anything by itself. Instead you’ve to build up all interactions, most of which are physical, and know how to do so. And what’s the result? Some spring interactions and markers on a bitmap?
Compare this to how the web or apps are used today. They’re task/purpose driven, and all of the UX has already been thought out. It’s a far more simple and straightforward approach to build the functionality you want which you refine over time than to give a blank canvas that does everything and nothing and tell people to go figure it out themselves.
It seems to me this is really just wanting to take the underpinnings of small talk and turn it into a physical UI representation. That’s fun… but now what? I’m zero percent surprised this hasn’t lasted.
> Wildly impractical
I couldn't disagree more strongly. Part of the vision in PARC (and Alan Kay's life work) was to create learning environments for children, and this demo is set in that realm, so having kids figure it out themselves is enormously empowering. Obviously not what you want if you're trying to sell widgets, but the widget-sellers themselves could benefit from this kind of environment to develop a usable widget-buying experience.
> The UX has already been thought out ... [and] ... refine[d] over time.
In such an impoverished way, in my experience. One of the things Alan Kay has said (elsewhere) is that the environment at PARC allowed them to experiment with radically different user interfaces in the afternoon, after thinking about it at lunch. This implies an in-depth knowledge of the tools, yes, but developers need that in any environment. I'm doing volunteer work now and having to learn to use several software systems (EventBrite, MailChimp, Salesforce). The difficulty of using these systems is in range for me as a retired software engineer, (although endlessly annoying), but other volunteers need extensive training. There's is nothing 'refined' about them, in terms of ease-of-use to the end users. The current tools for developing the interfaces (HTML, CSS, SQL, Javascript, etc.) are also primitive compared to the environment shown here. Again, it's a lengthy task to become a power use of the system demo-ed here, but no more so than to become a power use of the 10 or more technologies you need to learn to make a modern web site.
As a web site developer, with these kinds of tools, you'd develop your own tools and abstractions (an intermediate layer) to build with, none of which would appear anything like the physical world manipulation tools that have been built here.
>I often wish he was more explicit about WHAT was developed. (I have watched the mother of all demos, which is truly awesome, but partial information).
Here's one of several demos Kay has done of the Squeak/Croquet/Etoys/Frank systems, all built to extend on the PARC original. The others are good, too:
https://www.youtube.com/watch?v=prIwpKL57dM
There is some nice concepts in here that you can see in some things (like Unreal Blueprints) but the problem it has was you had to use a button for everything, and it just seems a bit clunky.
I would love to see something like this for the web, there's never really been a tool that captures it as close as Yahoo Pipes did.
Interesting; this somehow feels like a hybrid between NI LabView and Baba Is You.
[dead]
Thought it was an alternative fact kit LOL …