[squeak-dev] [Ann] Cuis: A new Squeak distribution

Damien Cassou wrote:
> On Fri, Mar 27, 2009 at 9:16 PM, Juan Vuletich <[hidden email]> wrote:
>  
>> - Include code from the Squeak project (various versions, Mantis, squak-dev,
>> etc)
>>    
>
> Do you remember what you did reuse from squeak-dev?
>
>  
That I can remember right now, all the kernel fixes by Andreas, the
MessageTally multiprocess fixes by Andreas and me, numeric stuff by
Nicolas Ceiller, "blue pill" == comparison with SmallIntegers from Dan,
and lots of useful discussions and advice. I'm sure I'm missing
thousands of things...

Cheers,
Juan Vuletich

Damien Cassou wrote: If I remeber correctly, these kind of problems also happened with other
Squeak distributions and was due to a kind of bug in BitBlt, right?

Is it this problem http://bugs.squeak.org/view.php?id=7068 ?

Cheers,
Juan Vuletich

Juan Vuletich a écrit :
> :) You're right! It is also my personal image, so it has these kind of
> things I use for my projects. I use the Signal Processing stuff for
> Morphic 3 and I still intend to write an audio editor based on my
> thesis. I guess I'd remove all this from Cuis.


is your thesis available for reading ? I would be interested.

cheers,

Stef

Bert Freudenberg wrote: Agreed.

Cheers,
Juan Vuletich

Edgar J. De Cleene wrote: Thanks Edgar!

Cheers,
Juan Vuletich

Stéphane Rollandin wrote: It seems I've been too quiet recently... I would have thought you were
aware of it. It is at http://www.jvuletich.org/research.html
The ideas sketched there have evolved enough to be turned into a
commercial product. But there is always so little time...

Comments very welcome.

Cheers,
Juan Vuletich

On 28.03.2009, at 13:31, Juan Vuletich wrote:

Not a BitBlt bug, but how BitBlt is used. If all drawing operations  
would set the Display's alpha component to 1 then it would be fully  
opaque.

> Is it this problem http://bugs.squeak.org/view.php?id=7068 ?


It sounds like that's the one. Old Unix VMs do not ignore the alpha  
channel of Display.

- Bert -

>> is your thesis available for reading ? I would be interested.
>>
>> cheers,
>>
>> Stef
>>
> It seems I've been too quiet recently... I would have thought you were
> aware of it.

hmm... you may be thinking of the other Stef :)

> It is at http://www.jvuletich.org/research.html

thanks !

Stef

Stéphane Rollandin wrote: Not at all! You're the Stef who's into music stuff!

>> It is at http://www.jvuletich.org/research.html
>
> thanks !
>
> Stef

Juan Vuletich a écrit :
>> hmm... you may be thinking of the other Stef :)
>>
> Not at all! You're the Stef who's into music stuff!

ah that's right. well I'm more into composition than sound
analysis/synthesis.

cheers,

Stef

Juan,

>> I assume that, though Monticello is not in the image, it is not hard
>> to add, since it doesn't depend on much.
>
> I also think it should not be too hard to add. But I think Monticello itself
> should be an optional package. That's why I removed it.

For what it's worth, PackageInfo and Monticello more or less file in
and work with some debugger whack-a-mole.  Some of the morphic stuff
needs to be renamed from e.g. PluggableListMorph to
OldPluggableListMorph if you want the UI.

> Well, somebody would need to try. I believe Omnibrowser should be optional
> too.

I haven't tried OB yet.

Ben

Ben Matasar wrote: Cool would you be willing to get LevelPlayingField working with Cuis?

Keith

On Saturday 28 Mar 2009 5:32:39 pm Bert Freudenberg wrote:
> Having seen various attempts at that I think splitting Morphic is  
> *not* the best way forward. Instead, I think the current FullMorphic  
> (i.e. including Etoys and all the goodies) should be treated as one  
> unit and there would be complete replacements for it.
+1. Is this a proposal for Mtoys (Morphic+Etoys)?

Has anyone noticed that the new Plasma engine in KDE 4.2.1 is sounding more
like a Morphic world:
--- see userbase.kde.org/Plasma
Plasma's components are widgets called Plasmoids. Plasmoids can take on a
variety of functions, ranging from displaying your desktop and associated
wallpaper, showing your laptop's battery level, displaying your plugged in
devices, and drawing the taskbar: basically, they are small applications that
live on the desktop. Plasmoids can be grouped together in "containers" called
containments......

It doesn't sound too new... other operating systems have done that.

The key difference here is that plasmoids can interact together. You want a
better view of your laptop battery in order to find out when you are running
low? You just drag it away from the taskbar and put it on the desktop. Also,
applets can be resized and rotated at will, thanks to the use of Scalable
Vector Graphics (SVGs). As you can see, the desktop not only interacts with
you, as the user, but also with itself in new and interesting ways. You are
now able to control how your workspace behaves and what it displays, in a
visually pleasing and user-friendly manner. Since Plasma is the sum of its
plasmoids, every element, even the desktop itself, is a widget. This allows
you to move your desktop anywhere with respect to the windows (back and
forward). It is no longer rooted behind everything and becomes instead
another element of real interaction.
---

On Sat, 28 Mar 2009 13:17:19 +0100, Juan Vuletich wrote:
...
>>
>> I'm under the impression that system windows flash when focused; is  
>> this intended or can the flash be turned off somehow?
>>
> No, it was not intended, and I don't see it... How do you make them  
> flash? Just clicking on each Workspace doesn't make them flash here...

Right, clicking into the text pane of your two workspaces doesn't make  
them flash but, clicking instead their title bar does make their window  
flash.

Okay; how would I do that?

...
>>> - Backwards compatibility not important
>>
>> :) can I have my (#scrollBarsOnRight false) preference back?
>>
> Sure! You can implement it. It could be another community supported  
> optional package. Much simpler than FT2, I guess.

I'll give it a try next week.

...
--
"If at first, the idea is not absurd, then there is no hope for it".  
Albert Einstein

On Sat, Mar 28, 2009 at 2:24 PM, Juan Vuletich <[hidden email]> wrote:
> That I can remember right now, all the kernel fixes by Andreas, the
> MessageTally multiprocess fixes by Andreas and me, numeric stuff by Nicolas
> Ceiller, "blue pill" == comparison with SmallIntegers from Dan, and lots of
> useful discussions and advice. I'm sure I'm missing thousands of things...

These are not part of squeak-dev images. Squeak-dev images do not
contain any fix.

--
Damien Cassou
http://damiencassou.seasidehosting.st

On Sat, Mar 28, 2009 at 2:31 PM, Juan Vuletich <[hidden email]> wrote:
> If I remeber correctly, these kind of problems also happened with other
> Squeak distributions and was due to a kind of bug in BitBlt, right?
>
> Is it this problem http://bugs.squeak.org/view.php?id=7068 ?

No, this is not the same bug: on your image, most of the text can't be
seen but the window itself is not translucent. Moreover, I'm using the
same VM to run pharo and it works quite well.

--
Damien Cassou
http://damiencassou.seasidehosting.st

Hi Ben,

Ben Matasar wrote: Thanks! I'll try to remove the Old prefix this week. I'm not crazy about
compatibility, but this creates a lot of unneeded incompatibility and
I'll fix it.
>> Well, somebody would need to try. I believe Omnibrowser should be optional
>> too.
>>    
>
> I haven't tried OB yet.
>
> Ben
>  
Cheers,
Juan Vuletich

Hi Klaus,

Klaus D. Witzel wrote:
>>>
>>>
>> No, it was not intended, and I don't see it... How do you make them
>> flash? Just clicking on each Workspace doesn't make them flash here...
>
> Right, clicking into the text pane of your two workspaces doesn't make
> them flash but, clicking instead their title bar does make their
> window flash.
>
Oh, yes. A feature I still haven't documented is keyboard focus change.
Alt/cmd + arrow keys lets you navigate morphs who handle keyboard focus.
The flash occurs when setting focus activates a different SystemWindow.
The pane that will get keyboard focus will flash. Try it on a browser,
you can do quite a lot without going to the mouse. Now I think it might
flash even if no new SystemWindow is activated... Will experiment a bit
with this.

>>>
>> No. But you can load those fonts in Cuis anytime. You just need to
>> build a couple of files for importing, using some other Squeak
>> environment that does support FT2.
>
> Okay; how would I do that?
>
Please allow 1 or 2 days for me to start documenting.

>>>
>> Sure! You can implement it. It could be another community supported
>> optional package. Much simpler than FT2, I guess.
>
> I'll give it a try next week.
>

That would be great. Cheers,
Juan Vuletich

Damien Cassou wrote: LOL! We have a name clash. When I say "squeak-dev" I'm referring to this
mail list. Google "squeak-dev" and you'll see what I mean.

Cheers,
Juan Vuletich

Damien Cassou wrote: Reading Bert's message, I realize that setting Display alpha channel
might be considered a bug. I'll try to fix it this week.

BTW, this is supposed to be a developer's community. I mean, people with
some level of technical ability. I believe you that after reading #7068
on Mantis, you should realize that it would be most helpful if you
checked the alpha values in Display, and set them to opaque (1.0, 255),
and see if the problem persists. Then, you could tell if it is the same
bug or not, and the bug report would be much useful. And obviously, the
problem not happening in Pharo doesn't mean anything about being the
same or a different problem.

Moreover, if the problem is in 3.10 and other Squeak releases, and it
was fixed in Pharo, you could ask there and tell about the fix, right? I
don't see the Pharo fix at 7068 nor at 7001.

Cheers,
Juan Vuletich