Melia - different kind of GUI

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|

Melia - different kind of GUI

adrin-3
Hi,
I just completed first stage of draft Melia GUI concept. I like to ask
you, what is wrong, what is good, how to continue and so on...
Please look here: http://adrin.ic.cz/Wiki/index.php?n=GUI.GUI

Thanks,
    Adam.

Reply | Threaded
Open this post in threaded view
|

Re: Melia - different kind of GUI

Alan Kay
Hi --

I looked at your website and I think this is a good idea (especially if you don't require window boundaries around objects -- this has always been one of the most misunderstood properties of the PARC Smalltalk GUI).

Historically, one of the first times this was done, pretty much exactly as you suggest, was the object-by-object editing as a halo of controls, and I think first implemented by Ted Kaehler at PARC in one of the very first attempts at doing a DTP system that combined modeless text editing with embedded pictures. This was before Smalltalk-76 as I recall. Here's a picture I found in the Early History of Smalltalk paper for HOPL II.

Emacs!

The larger version of this picture (which I don't have handy on my machine in Japan) shows that this illustration is embedded in a galley of text paragraphs. The idea in this interface was that when you touch any object, it should show its editing interface as a frame or halo around the object. Some remnants of this idea are to be seen in the Etoys halo of handles, and its "costume" architecture, etc.

Cheers,

Alan

At 02:17 AM 1/23/2007, Adrin wrote:
Hi,
I just completed first stage of draft Melia GUI concept. I like to ask you, what is wrong, what is good, how to continue and so on...
Please look here: http://adrin.ic.cz/Wiki/index.php?n=GUI.GUI

Thanks,
   Adam.
Reply | Threaded
Open this post in threaded view
|

Re: Melia - different kind of GUI

timrowledge

On 23-Jan-07, at 3:24 PM, Alan Kay wrote:
> . Some remnants of this idea are to be seen in the Etoys halo of  
> handles, and its "costume" architecture, etc.
And some extensions of it are seen in Sophie.

tim
--
tim Rowledge; [hidden email]; http://www.rowledge.org/tim
Strange OpCodes: START: Cancel preceding jobs in queue



Reply | Threaded
Open this post in threaded view
|

Re: Melia - different kind of GUI

johnmci
Ya, like at http://www.sophieproject.org/node/12

On Jan 23, 2007, at 3:29 PM, tim Rowledge wrote:

>
> On 23-Jan-07, at 3:24 PM, Alan Kay wrote:
>> . Some remnants of this idea are to be seen in the Etoys halo of  
>> handles, and its "costume" architecture, etc.
> And some extensions of it are seen in Sophie.
>
> tim
>

--
========================================================================
===
John M. McIntosh <[hidden email]>
Corporate Smalltalk Consulting Ltd.  http://www.smalltalkconsulting.com
========================================================================
===



Reply | Threaded
Open this post in threaded view
|

Re: Melia - different kind of GUI

adrin-3
Thank you very much for your answers.
I have read something about Etoys costumes and I like that idea. But I
don't really looked at Sophie, yet. I will :)

John M McIntosh napsal(a):

> Ya, like at http://www.sophieproject.org/node/12
>
> On Jan 23, 2007, at 3:29 PM, tim Rowledge wrote:
>
>>
>> On 23-Jan-07, at 3:24 PM, Alan Kay wrote:
>>> . Some remnants of this idea are to be seen in the Etoys halo of
>>> handles, and its "costume" architecture, etc.
>> And some extensions of it are seen in Sophie.
>>
>> tim
>>
>
> --
> ===========================================================================
>
> John M. McIntosh <[hidden email]>
> Corporate Smalltalk Consulting Ltd.  http://www.smalltalkconsulting.com
> ===========================================================================
>
>
>
>
>


Reply | Threaded
Open this post in threaded view
|

Re: Melia - different kind of GUI

adrin-3
In reply to this post by Alan Kay
Hi,

I looked closer to Etoys and Sophie and have pretty much to study.

But there is one thing, almost philosophical thing, what makes Melia
different. If I em correct, costumes and halos are extensions applied to
objects. Object will live with these extensions. In Melia I try to
really separate object from its possible visualization and possible set
of modification tools. I could be able to drag&drop object from one
window to another and object will change its appearance dynamically. For
this matter I must define standard interface (in Melia I use archetype
addressing - cool term isn't it? :) I do it, because I still try to
follow real world working with objects - take it, look at it, focus on
part of it, modify it.

Cheers,
    Adam

Alan Kay napsal(a):

> Hi --
>
> I looked at your website and I think this is a good idea (especially
> if you don't require window boundaries around objects -- this has
> always been one of the most misunderstood properties of the PARC
> Smalltalk GUI).
>
> Historically, one of the first times this was done, pretty much
> exactly as you suggest, was the object-by-object editing as a halo of
> controls, and I think first implemented by Ted Kaehler at PARC in one
> of the very first attempts at doing a DTP system that combined
> modeless text editing with embedded pictures. This was before
> Smalltalk-76 as I recall. Here's a picture I found in the Early
> History of Smalltalk paper for HOPL II.
>
> Emacs!
>
> The larger version of this picture (which I don't have handy on my
> machine in Japan) shows that this illustration is embedded in a galley
> of text paragraphs. The idea in this interface was that when you touch
> any object, it should show its editing interface as a frame or halo
> around the object. Some remnants of this idea are to be seen in the
> Etoys halo of handles, and its "costume" architecture, etc.
>
> Cheers,
>
> Alan
>
> At 02:17 AM 1/23/2007, Adrin wrote:
>> Hi,
>> I just completed first stage of draft Melia GUI concept. I like to
>> ask you, what is wrong, what is good, how to continue and so on...
>> Please look here: http://adrin.ic.cz/Wiki/index.php?n=GUI.GUI
>>
>> Thanks,
>>    Adam.
> ------------------------------------------------------------------------
>
>
>  


Reply | Threaded
Open this post in threaded view
|

Re: Melia - different kind of GUI

Alan Kay
Hi --

The costumes are supposed to be separate. An example in Croquet is
that when you go from a world where you might look like the Rabbit
into the waterworld, you change into a fish. Also, Croquet has done a
lot of experimentation with filtered viewing, which is worth looking
at. Logical separation is one thing, but there is also the question
of identity, namely, when is a view of something to be taken as a
manifestation of the thing, and when should it be considered a new thing?

Cheers,

Alan

At 12:37 AM 1/25/2007, Adrin wrote:

>Hi,
>
>I looked closer to Etoys and Sophie and have pretty much to study.
>
>But there is one thing, almost philosophical thing, what makes Melia
>different. If I em correct, costumes and halos are extensions
>applied to objects. Object will live with these extensions. In Melia
>I try to really separate object from its possible visualization and
>possible set of modification tools. I could be able to drag&drop
>object from one window to another and object will change its
>appearance dynamically. For this matter I must define standard
>interface (in Melia I use archetype addressing - cool term isn't it?
>:) I do it, because I still try to follow real world working with
>objects - take it, look at it, focus on part of it, modify it.
>
>Cheers,
>    Adam
>
>Alan Kay napsal(a):
>>Hi --
>>
>>I looked at your website and I think this is a good idea
>>(especially if you don't require window boundaries around objects
>>-- this has always been one of the most misunderstood properties of
>>the PARC Smalltalk GUI).
>>
>>Historically, one of the first times this was done, pretty much
>>exactly as you suggest, was the object-by-object editing as a halo
>>of controls, and I think first implemented by Ted Kaehler at PARC
>>in one of the very first attempts at doing a DTP system that
>>combined modeless text editing with embedded pictures. This was
>>before Smalltalk-76 as I recall. Here's a picture I found in the
>>Early History of Smalltalk paper for HOPL II.
>>
>>Emacs!
>>
>>The larger version of this picture (which I don't have handy on my
>>machine in Japan) shows that this illustration is embedded in a
>>galley of text paragraphs. The idea in this interface was that when
>>you touch any object, it should show its editing interface as a
>>frame or halo around the object. Some remnants of this idea are to
>>be seen in the Etoys halo of handles, and its "costume" architecture, etc.
>>
>>Cheers,
>>
>>Alan
>>
>>At 02:17 AM 1/23/2007, Adrin wrote:
>>>Hi,
>>>I just completed first stage of draft Melia GUI concept. I like to
>>>ask you, what is wrong, what is good, how to continue and so on...
>>>Please look here: http://adrin.ic.cz/Wiki/index.php?n=GUI.GUI
>>>
>>>Thanks,
>>>    Adam.
>>------------------------------------------------------------------------
>>
>>
>>
>


Reply | Threaded
Open this post in threaded view
|

Re: Melia - different kind of GUI

Howard Stearns
The answer is: a costume is considered it's own thing precisely when  
the user says so.



That sounds like a platitude, but I intend every word of it.

For example, in Brie (but you could imagine similar stuff in Melia or  
any of it's references):

* Croquet users observe objects through interactors/filters (3D  
viewing windows), which can be shared or private.  Objects have a  
view for each interactor that is only visible when seen through that  
interactor.  Of course, interactors stack, and one interactor's view  
is another's object.

*The default interactor behavior is to map left mouse click to #use.  
The default #use behavior of a view is to #select the object that it  
is a view of.  However, buttons typically do something else on #use.  
The selection state is held by the interactor.

* The default interactor behavior is to map right mouse click to  
#author. An #author gesture is typically handled by a #authorSelect  
of the view object, which suppresses any temptation for the view to  
act out.

Thus "left click for normal use, right click to get inside of stuff  
for authoring.  Fences -- but very low fences.

http://opencroquet.org/Site%20PDFs/2006%20BrieUserExperience.pdf
(From last year's C5. Section 6 is most apropos of para-selection,  
but of course I hope for folks to read all in order.)

-Howard

On Jan 25, 2007, at 11:48 AM, Alan Kay wrote:

> Hi --
>
> The costumes are supposed to be separate. An example in Croquet is  
> that when you go from a world where you might look like the Rabbit  
> into the waterworld, you change into a fish. Also, Croquet has done  
> a lot of experimentation with filtered viewing, which is worth  
> looking at. Logical separation is one thing, but there is also the  
> question of identity, namely, when is a view of something to be  
> taken as a manifestation of the thing, and when should it be  
> considered a new thing?
>
> Cheers,
>
> Alan
>
> At 12:37 AM 1/25/2007, Adrin wrote:
>> Hi,
>>
>> I looked closer to Etoys and Sophie and have pretty much to study.
>>
>> But there is one thing, almost philosophical thing, what makes  
>> Melia different. If I em correct, costumes and halos are  
>> extensions applied to objects. Object will live with these  
>> extensions. In Melia I try to really separate object from its  
>> possible visualization and possible set of modification tools. I  
>> could be able to drag&drop object from one window to another and  
>> object will change its appearance dynamically. For this matter I  
>> must define standard interface (in Melia I use archetype  
>> addressing - cool term isn't it? :) I do it, because I still try  
>> to follow real world working with objects - take it, look at it,  
>> focus on part of it, modify it.
>>
>> Cheers,
>>    Adam
>>
>> Alan Kay napsal(a):
>>> Hi --
>>>
>>> I looked at your website and I think this is a good idea  
>>> (especially if you don't require window boundaries around objects  
>>> -- this has always been one of the most misunderstood properties  
>>> of the PARC Smalltalk GUI).
>>>
>>> Historically, one of the first times this was done, pretty much  
>>> exactly as you suggest, was the object-by-object editing as a  
>>> halo of controls, and I think first implemented by Ted Kaehler at  
>>> PARC in one of the very first attempts at doing a DTP system that  
>>> combined modeless text editing with embedded pictures. This was  
>>> before Smalltalk-76 as I recall. Here's a picture I found in the  
>>> Early History of Smalltalk paper for HOPL II.
>>>
>>> Emacs!
>>>
>>> The larger version of this picture (which I don't have handy on  
>>> my machine in Japan) shows that this illustration is embedded in  
>>> a galley of text paragraphs. The idea in this interface was that  
>>> when you touch any object, it should show its editing interface  
>>> as a frame or halo around the object. Some remnants of this idea  
>>> are to be seen in the Etoys halo of handles, and its "costume"  
>>> architecture, etc.
>>>
>>> Cheers,
>>>
>>> Alan
>>>
>>> At 02:17 AM 1/23/2007, Adrin wrote:
>>>> Hi,
>>>> I just completed first stage of draft Melia GUI concept. I like  
>>>> to ask you, what is wrong, what is good, how to continue and so  
>>>> on...
>>>> Please look here: http://adrin.ic.cz/Wiki/index.php?n=GUI.GUI
>>>>
>>>> Thanks,
>>>>    Adam.
>>> --------------------------------------------------------------------
>>> ----
>>>
>>>
>>>
>>
>
>


Reply | Threaded
Open this post in threaded view
|

Re: Melia - different kind of GUI

adrin-3

I see I have more and more materials to study. You have mentioned many
thing that is very interesting, so I will look how this is done and why.
In fact, for the first time I try to use decorator pattern for manage
visualization changes (but I finally change my mind - don't know why at
this moment).

Thank you very much again, I em grateful for your comments :)

Cheers,
    Adam


Howard Stearns napsal(a):

> The answer is: a costume is considered it's own thing precisely when
> the user says so.
>
>
>
> That sounds like a platitude, but I intend every word of it.
>
> For example, in Brie (but you could imagine similar stuff in Melia or
> any of it's references):
>
> * Croquet users observe objects through interactors/filters (3D
> viewing windows), which can be shared or private.  Objects have a view
> for each interactor that is only visible when seen through that
> interactor.  Of course, interactors stack, and one interactor's view
> is another's object.
>
> *The default interactor behavior is to map left mouse click to #use.
> The default #use behavior of a view is to #select the object that it
> is a view of.  However, buttons typically do something else on #use.
> The selection state is held by the interactor.
>
> * The default interactor behavior is to map right mouse click to
> #author. An #author gesture is typically handled by a #authorSelect of
> the view object, which suppresses any temptation for the view to act out.
>
> Thus "left click for normal use, right click to get inside of stuff
> for authoring.  Fences -- but very low fences.
>
> http://opencroquet.org/Site%20PDFs/2006%20BrieUserExperience.pdf
> (From last year's C5. Section 6 is most apropos of para-selection, but
> of course I hope for folks to read all in order.)
>
> -Howard
>
> On Jan 25, 2007, at 11:48 AM, Alan Kay wrote:
>
>> Hi --
>>
>> The costumes are supposed to be separate. An example in Croquet is
>> that when you go from a world where you might look like the Rabbit
>> into the waterworld, you change into a fish. Also, Croquet has done a
>> lot of experimentation with filtered viewing, which is worth looking
>> at. Logical separation is one thing, but there is also the question
>> of identity, namely, when is a view of something to be taken as a
>> manifestation of the thing, and when should it be considered a new
>> thing?
>>
>> Cheers,
>>
>> Alan
>>
>> At 12:37 AM 1/25/2007, Adrin wrote:
>>> Hi,
>>>
>>> I looked closer to Etoys and Sophie and have pretty much to study.
>>>
>>> But there is one thing, almost philosophical thing, what makes Melia
>>> different. If I em correct, costumes and halos are extensions
>>> applied to objects. Object will live with these extensions. In Melia
>>> I try to really separate object from its possible visualization and
>>> possible set of modification tools. I could be able to drag&drop
>>> object from one window to another and object will change its
>>> appearance dynamically. For this matter I must define standard
>>> interface (in Melia I use archetype addressing - cool term isn't it?
>>> :) I do it, because I still try to follow real world working with
>>> objects - take it, look at it, focus on part of it, modify it.
>>>
>>> Cheers,
>>>    Adam
>>>
>>> Alan Kay napsal(a):
>>>> Hi --
>>>>
>>>> I looked at your website and I think this is a good idea
>>>> (especially if you don't require window boundaries around objects
>>>> -- this has always been one of the most misunderstood properties of
>>>> the PARC Smalltalk GUI).
>>>>
>>>> Historically, one of the first times this was done, pretty much
>>>> exactly as you suggest, was the object-by-object editing as a halo
>>>> of controls, and I think first implemented by Ted Kaehler at PARC
>>>> in one of the very first attempts at doing a DTP system that
>>>> combined modeless text editing with embedded pictures. This was
>>>> before Smalltalk-76 as I recall. Here's a picture I found in the
>>>> Early History of Smalltalk paper for HOPL II.
>>>>
>>>> Emacs!
>>>>
>>>> The larger version of this picture (which I don't have handy on my
>>>> machine in Japan) shows that this illustration is embedded in a
>>>> galley of text paragraphs. The idea in this interface was that when
>>>> you touch any object, it should show its editing interface as a
>>>> frame or halo around the object. Some remnants of this idea are to
>>>> be seen in the Etoys halo of handles, and its "costume"
>>>> architecture, etc.
>>>>
>>>> Cheers,
>>>>
>>>> Alan
>>>>
>>>> At 02:17 AM 1/23/2007, Adrin wrote:
>>>>> Hi,
>>>>> I just completed first stage of draft Melia GUI concept. I like to
>>>>> ask you, what is wrong, what is good, how to continue and so on...
>>>>> Please look here: http://adrin.ic.cz/Wiki/index.php?n=GUI.GUI
>>>>>
>>>>> Thanks,
>>>>>    Adam.
>>>> ------------------------------------------------------------------------
>>>>
>>>>
>>>>
>>>>
>>>
>>
>>
>
>
>