Better morphic architecture? (was: Re: Changing window size)

>
> tR> complained for *years* about how utterly essential  host windows are
> tR> and yet in the nearly three years since we wrote Ffenestri there has
> tR> been..... <crickets>....
>
> It seems some of those "You guys" will have to change the whole morphic
> drawing, BitBlt whatever to paint on a HostWindow. Which looks like an
> awful lot of work. So I guess people having a need will do a bit in
> this area and maybe it sums up one day.
>

>>
>> tR> complained for *years* about how utterly essential  host  
>> windows are
>> tR> and yet in the nearly three years since we wrote Ffenestri  
>> there has
>> tR> been..... <crickets>....
>>
>> It seems some of those "You guys" will have to change the whole  
>> morphic
>> drawing, BitBlt whatever to paint on a HostWindow. Which looks  
>> like an
>> awful lot of work. So I guess people having a need will do a bit in
>> this area and maybe it sums up one day.
>>
> A good point! There is need to make substantial changes with
> Morphic/Display infrastructure to support multiple OS windows/hardware
> Displays/Desktops.
> Current design implies that you having only single display surface and
> i find this is very archaic approach.
>
>> I'm happy to see that more and more people are on the way to use
>> Squeak commercially and not only Seaside.
>>
>> Cheers
>>
>> Herbert                            mailto:[hidden email]
>>
>>
>>
>

> On 7/13/07, John M McIntosh <[hidden email]> wrote:
>
>> Laslty I'll note Morphic was rewritten over in
>>
>> http://scratch.mit.edu/
>>
>> I believe under the grant rules the source should be available,
>> surprised  no-one has gone digging to
>> see if that has a better Morphic to build on?
>
> In http://news.squeak.org/2007/01/23/scratching-the-surface/ John said
> that "We plan to make source code available by mid-2007 under the MIT
> License so that others can experiment with extensions and variations."
>
> In early versions of Scratch it was possible to get a debugger in some
> situations and from there get to the full development environment,
> which looked like a pretty standard (but snappy) Morphic circa Squeak
> 2.2 or so.  I haven't tried doing that with a recent release...

> Ah, well Tim and note there were oh say in an arbitrary 3.8 image
>
> 424 uses of Display and
> 265 uses of Sensor
>
> which is 689 places to visit and understand. Also each of those has
> the decision
> which sensor and which display to use?
>
> In that era of Morphic is Dead, long live Tweak we decided not to do
> anything.
> In Tweak you ask for the world's Display or Sensor so it always has
> the right context if you
> tie a TweakWorld to a particular HostWindow.
>

> Laslty I'll note Morphic was rewritten over in
>
> http://scratch.mit.edu/
>
> I believe under the grant rules the source should be available,
> surprised  no-one has gone digging to
> see if that has a better Morphic to build on?
>
> On Jul 13, 2007, at 1:45 PM, sig wrote:
>
> >>
> >> tR> complained for *years* about how utterly essential  host
> >> windows are
> >> tR> and yet in the nearly three years since we wrote Ffenestri
> >> there has
> >> tR> been..... <crickets>....
> >>
> >> It seems some of those "You guys" will have to change the whole
> >> morphic
> >> drawing, BitBlt whatever to paint on a HostWindow. Which looks
> >> like an
> >> awful lot of work. So I guess people having a need will do a bit in
> >> this area and maybe it sums up one day.
> >>
> > A good point! There is need to make substantial changes with
> > Morphic/Display infrastructure to support multiple OS windows/hardware
> > Displays/Desktops.
> > Current design implies that you having only single display surface and
> > i find this is very archaic approach.
> >
> >> I'm happy to see that more and more people are on the way to use
> >> Squeak commercially and not only Seaside.
> >>
> >> Cheers
> >>
> >> Herbert                            mailto:[hidden email]
> >>
> >>
> >>
> >
>
> --
> ========================================================================
> ===
> John M. McIntosh <[hidden email]>
> Corporate Smalltalk Consulting Ltd.  http://www.smalltalkconsulting.com
> ========================================================================
> ===
>
>
>
>