Re: Implementing stuff in morphic 3.

[Jerome Peace]

--- Juan Vuletich <juan@...> wrote:

> > Running your image under the classic os vm crashes
> the
> > image and often the system before I can do too
> much of
> > any type of exploring.
> >
> >  
> That's too bad. I wonder what would be the reason.
> My image is smaller
> than a standard 3.9 or 3.10 image, and uses less
> memory. I would like
> you to be able to use my image without major
> problems.

The size of the image is secondary. The size that
matters is the peak size the application requires as
it runs. With graphic intensive things you can hit the
ceiling by requiring too much memory for just the
intermediate forms.

The other showstoppers for me are trying to use the
halos in ways I "know" work and running into the fact
they haven't been implemented for your morphs.

And drag and drop doesn't work. So I can't figure out
how to embed one morph in another.  

You are familiar with what you wrote in a way that I
am not. I am more or less poking at things from the
outside. And I lose interest after about finding 5-9
(seven plus or minus two)  things that don't work.

Then its a matter of waiting until you make progress.

I know you are serious about this. So I would
encourage you to get an official enough status for
this that we can have a catagory for it on mantis**.
The I can turn may annoyances with the way things work
into bug reports. Which will help ratchet progress and
may even help document the ideas and issues we come
across as you develop your baby.


> > I also think it would be the right way to
> > incrementally develop morphic 3.0.
> >
> >  
> Stef also asked me that. Unfortunately to start
> doing that I would have
> to halt the development of Morphic 3 itself.

Yeah, at this point it would be a big disruption. But
maybe keep it in mind if the project itself calls for
a big disruption.




> I have
> modified lots of
> things in "old" Morphic, and it will not be that
> easy.

Like what?


> I have very
> little time available for this project, and I'm
> trying to spend it
> carefully.

I know that situation.


> >And we must follow them if they
> >are to let us lead.
> >
> >  
> I think I don't agree with this.

All my experience has shown this to be true. I need to
understand where other are and wish to go inorder to
provide a path they wish to follow.

There has to be something they want in what we do
inorder for them to participate.

This does not have to be important to us as we
experiment. What I want is the learning that comes
from the task. I would like to see my better ideas
widely distributed because that will be more useful
and generate more and better feedback as well as
dividends.



> I do what I do because it is what I
> need. I invite everybody to participate. But I won't
> change what I'm  doing to seduce others.


> They would need to convince
> me that I was wrong,
> and to show me a better way, for me to follow them.
> After all this is
> exactly what Dan Ingalls always did.

Dan had the benifit of a nacent system and a paycheck.

> And at the end he was always right.

He didn't really get to the end. He went back to take
care of the family farm and the team he had assembled
broke up. See the references in Berts comment on
fabrik.
 
http://lists.squeakfoundation.org/pipermail/squeak-dev/2007-June/118049.html

-----

In conclusion,

I will support you as much as I can from the outside
by playing with things as you improve them. I am not
too interested in diving into your separate morphic
3.0 code.

If you get a catagory for Morphic 3.0 set up on Mantis
I will write bug reports as I come across them. I very
much recommend this approach.

I am entertaining my curiosity by coding the
place/morph layers concept in the current development
stream of squeak. I am going to do what is practical
to do but I am keeping in mind your concept of each
morph having its own coord system. So if you look over
my work and I yours we should benifit from each others
insights.


Yours in curiosity and service, --Jerome Peace


--------

**This should probably be a catgory under the squeak
project rather than its own separate project. That is
due to limitations of mantis in dealling amoung
projects. Ken Causey may be able to provide some
advice.  _Jer


      ____________________________________________________________________________________
Fussy? Opinionated? Impossible to please? Perfect.  Join Yahoo!'s user panel and lay it on us. http://surveylink.yahoo.com/gmrs/yahoo_panel_invite.asp?a=7 

_______________________________________________
Morphic mailing list
Morphic@...
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/morphic

[Stéphane Ducasse-3]

>
>>> I also think it would be the right way to
>>> incrementally develop morphic 3.0.
>>>
>>>
>> Stef also asked me that. Unfortunately to start
>> doing that I would have
>> to halt the development of Morphic 3 itself.
>
> Yeah, at this point it would be a big disruption. But
> maybe keep it in mind if the project itself calls for
> a big disruption.

What I was saying is that if you want that other people
give a try and at the end helps you need to offer them
a way to work on their image and to be able to run side by side
Morphic3 and Morphic.

>> I do what I do because it is what I
>> need. I invite everybody to participate. But I won't
>> change what I'm  doing to seduce others.

The main point is to make sure that others feel comfortable enough to  
help.

Stef

_______________________________________________
Morphic mailing list
Morphic@...
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/morphic

[Juan Vuletich-4]

Hi Jerome,

I apologize for the delay. My second daugher, Diana was born on 6/28,
and I'm slowly coming back.

Jerome Peace escribió:
> The size of the image is secondary. The size that
> matters is the peak size the application requires as
> it runs. With graphic intensive things you can hit the
> ceiling by requiring too much memory for just the
> intermediate forms.
>  
Setting Display depth to 16 (instead of 32) reduces memory usage
significantly. I should study Form allocation, to see if there's
something to optimize, though.
> The other showstoppers for me are trying to use the
> halos in ways I "know" work and running into the fact
> they haven't been implemented for your morphs.
>
> And drag and drop doesn't work. So I can't figure out
> how to embed one morph in another.  
>
>  
Yes, those still missing. I do it all with code. Quite easy, anyway.
> You are familiar with what you wrote in a way that I
> am not. I am more or less poking at things from the
> outside. And I lose interest after about finding 5-9
> (seven plus or minus two)  things that don't work.
>
> Then its a matter of waiting until you make progress.
>
>  
That's ok.

> I know you are serious about this. So I would
> encourage you to get an official enough status for
> this that we can have a catagory for it on mantis**.
> The I can turn may annoyances with the way things work
> into bug reports. Which will help ratchet progress and
> may even help document the ideas and issues we come
> across as you develop your baby.
>
>
>  

Ok. I agree.

>> I have
>> modified lots of
>> things in "old" Morphic, and it will not be that
>> easy.
>>    
>
> Like what?
>
>
>  

I didn't log all I did. But if you browse OldMorph, OldMorphExtension,
and OldPasteUpMorph, you'll see I removed many instance variables. I
also removed a lot of methods, and modified their senders. As new Morphs
live in an old World, to avoid needing methods in Morph that I didn't
want, I removed their senders. Basically, I feel free to delete whatever
I don't like, anywhere in the system.

>  
>> They would need to convince
>> me that I was wrong,
>> and to show me a better way, for me to follow them.
>> After all this is
>> exactly what Dan Ingalls always did.
>>    
>
> Dan had the benifit of a nacent system and a paycheck.
>
>  

:)

>> And at the end he was always right.
>>    
>
> He didn't really get to the end. He went back to take
> care of the family farm and the team he had assembled
> broke up. See the references in Berts comment on
> fabrik.
>  
> http://lists.squeakfoundation.org/pipermail/squeak-dev/2007-June/118049.html
>  

Ok, but what I meant is that he did Smalltalk the way he felt, without
worrying if others would understand him or not. And I really believe
nothing better than Smalltalk was made. I.e. I believe Smalltalk is
better than C++, C#, Java, Python, Ruby, Self, etc.

> -----
>
> In conclusion,
>
> I will support you as much as I can from the outside
> by playing with things as you improve them. I am not
> too interested in diving into your separate morphic
> 3.0 code.
>
> If you get a catagory for Morphic 3.0 set up on Mantis
> I will write bug reports as I come across them. I very
> much recommend this approach.
>
> I am entertaining my curiosity by coding the
> place/morph layers concept in the current development
> stream of squeak. I am going to do what is practical
> to do but I am keeping in mind your concept of each
> morph having its own coord system. So if you look over
> my work and I yours we should benifit from each others
> insights.
>
>
> Yours in curiosity and service, --Jerome Peace
>
>
> --------
>
> **This should probably be a catgory under the squeak
> project rather than its own separate project. That is
> due to limitations of mantis in dealling amoung
> projects. Ken Causey may be able to provide some
> advice.  _Jer
>
>  

Great!
Thanks,
Juan Vuletich

_______________________________________________
Morphic mailing list
Morphic@...
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/morphic