pharo and ui again

Hi Stephan, Hi All,

Joachim and I come from the same, commercial world, and we are not used
to initial problems regarding the UI we are used to huge limitations
once your models become more and more complex and custom vizualisations
are desired.

On first sight the videos and examples are tempting and give a beginner
the feeling that everythign is possible in Pharo/Squeak.
Well it definitely is and I myself are also more the "do it your self" guy.
But there are simple tasks that should not need a huge involvement to
accomplish a visualization and manipulation of a model.

I think there is most stuff out there in Pharo that is needed to built a
nice UI with Model interaction.
It took me jsut a few hours to built a nice Glamour based data model
browser.
So why do I struggle to built a login view?!
I think this is because Glamour is its own example.... It is implemented
in one way and there are few distractions.
GTSpotter is already too difficult to understand again....

What Pharo needs and I do not think that this is a too huge task is.
A clear destiction of M, V and C accomplished by ClassNamingConvertions
and package sub-division of these.
A browser that just lists all the available basic morphs and their examples
A sushi store demo according to the current Pharo release.
A clear documentation what, Spec, Polymorph, Morphic,... are
representing and on which "layer" they are positioned. OS-Layer,
Common-Layer, Layout-Layer, Event-Layer.

Based on first steps experiance with Seaside I can still remember that
the only reason why I was able to get started fast and even during the
early days of Seaside were the two examples, Scriptacoulous demo site
and the Sushi Store.
With the help of those I even learnt some CSS and JS ...
The jQuery example was already to much limited to the usage of jQuery
and it was much more difficult to use the exampel site as an example to
go on with an own project.

How much time would it cost to implement a small sushi store with in
image persitsancy?
Also,.... a documentation on how to deplay this store and make a real
Pharo app.

I think such a demo would also show the community how huge a impact a
new Pharo release does to a commercial product.
A steady update from release to release would be useful as a pharo
version migration guide.
It could also start a discussion on how to do things and show each of us
where things are unclear and under specified.

We should take what we have and put an example together instead of
thinking/discussing what kind of framework might be best or better.
There are so many books an Smalltalk patterns, MVC,.... but there are
actually no real official examples available

Stephan! Thank you so much for starting this!
I will have a look asap.
Right now I am more focused on the local Software group.
I would like to have some application "WITH UI" and even 3D where I can
show off live programming!

I admire the energy the community and the board puts into Pharo and I
can only imagine how difficult it must be to find and lead into a
certain direction.
  Thank you!

So and now her eis the question that I think needs to be discussed since
it is the most controversial answered to me.
"Spec or not to Spec?"

I am kidding!
But my current counter is 5 for Spec and 5 against Spec.
I think some rason could be that Spec is misunderstood, or still under
development/inmature.
I can't tell, but  a beginner wants to know:
"WHAT AM I SUPPOSED TO DO!?"

Thanks for reading!
Sebastian






Am 12.02.2015 um 07:22 schrieb Stephan Eggermont:

They are not yet public but soon will be.

BTW, Johan, I cannot find your slides in http://www.slideshare.net/pharoproject/presentations ?

> On 12 Feb 2015, at 17:53, Sven Van Caekenberghe <[hidden email]> wrote:
>
> BTW, Johan, I cannot find your slides in http://www.slideshare.net/pharoproject/presentations ?
>
I have the .key files.. I will double check and upload the missings…

Kilon, your second paragraph can be understood as a description of Spec. There is a low level code and an extensions API, and ALL widgets are implemented as such extensions and you can assemble a GUI framework. What you propose as your ideal solution is reimplementing this wheel all over again.

In any case, my slides are pretty useless without me giving the complete song and dance routine as well ;-) Best to see them together with the video ...

> On Feb 12, 2015, at 17:59, Marcus Denker <[hidden email]> wrote:
>
>
>> On 12 Feb 2015, at 17:53, Sven Van Caekenberghe <[hidden email]> wrote:
>>
>> BTW, Johan, I cannot find your slides in http://www.slideshare.net/pharoproject/presentations ?
>>
> I have the .key files.. I will double check and upload the missings…
>



---> Save our in-boxes! http://emailcharter.org <---

Johan Fabry   -   http://pleiad.cl/~jfabry
PLEIAD lab  -  Computer Science Department (DCC)  -  University of Chile

Some restricted comments inline

> On Feb 12, 2015, at 17:15, Sebastian Heidbrink <[hidden email]> wrote:
>
> I think there is most stuff out there in Pharo that is needed to built a nice UI with Model interaction.
> It took me jsut a few hours to built a nice Glamour based data model browser.
> So why do I struggle to built a login view?!
> I think this is because Glamour is its own example.... It is implemented in one way and there are few distractions.

Glamour is excellent! But only as long as you want to do something that fits the overall philosophy of it. That’s the advantages and the disadvantages of a domain-specific solution. Spec is a general-purpose solution.

Your login UI should be able to be built in Spec in less than 20 minutes. (Supposing it is more or less like a traditional login UI).

> What Pharo needs and I do not think that this is a too huge task is.
> A clear destiction of M, V and C accomplished by ClassNamingConvertions and package sub-division of these.
> A browser that just lists all the available basic morphs and their examples

Browse the hierarchy of ComposableModel and you have all the Spec widgets and UIs that are in the image.

> A sushi store demo according to the current Pharo release.

I don’t know this example, can you give a link to an example app?

> A clear documentation what, Spec, Polymorph, Morphic,... are representing and on which "layer" they are positioned. OS-Layer, Common-Layer, Layout-Layer, Event-Layer.

Have a look at the Spec book chapter (in progress). It also relates the position of Spec with respect to Morphic https://ci.inria.fr/pharo-contribution/view/Books/job/PharoBookWorkInProgress/lastSuccessfulBuild/artifact/Spec/

[…]

> So and now her eis the question that I think needs to be discussed since it is the most controversial answered to me.
> "Spec or not to Spec?"
>
> I am kidding!
> But my current counter is 5 for Spec and 5 against Spec.
> I think some rason could be that Spec is misunderstood, or still under development/inmature.
> I can't tell, but  a beginner wants to know:
> "WHAT AM I SUPPOSED TO DO!?"

To answer your BIG QUESTION ;-) Read the book chapter that I linked to above, or at least the first few sections, and try to build the login UI with Spec. Then you will have better information that can serve to make a choice for or against your use of Spec.

---> Save our in-boxes! http://emailcharter.org <---

Johan Fabry   -   http://pleiad.cl/~jfabry
PLEIAD lab  -  Computer Science Department (DCC)  -  University of Chile

Kilon, I repeat and paraphrase from my previous mail:

Some restricted comments inline

On Feb 12, 2015, at 17:15, Sebastian Heidbrink [hidden email] wrote:

I think there is most stuff out there in Pharo that is needed to built a nice UI with Model interaction.
It took me jsut a few hours to built a nice Glamour based data model browser.
So why do I struggle to built a login view?!
I think this is because Glamour is its own example.... It is implemented in one way and there are few distractions.

Glamour is excellent! But only as long as you want to do something that fits the overall philosophy of it. That’s the advantages and the disadvantages of a domain-specific solution. Spec is a general-purpose solution.

Your login UI should be able to be built in Spec in less than 20 minutes. (Supposing it is more or less like a traditional login UI).

What Pharo needs and I do not think that this is a too huge task is.
A clear destiction of M, V and C accomplished by ClassNamingConvertions and package sub-division of these.
A browser that just lists all the available basic morphs and their examples

Browse the hierarchy of ComposableModel and you have all the Spec widgets and UIs that are in the image.

A sushi store demo according to the current Pharo release.

I don’t know this example, can you give a link to an example app?

A clear documentation what, Spec, Polymorph, Morphic,... are representing and on which "layer" they are positioned. OS-Layer, Common-Layer, Layout-Layer, Event-Layer.

Have a look at the Spec book chapter (in progress). It also relates the position of Spec with respect to Morphic https://ci.inria.fr/pharo-contribution/view/Books/job/PharoBookWorkInProgress/lastSuccessfulBuild/artifact/Spec/

[…]

So and now her eis the question that I think needs to be discussed since it is the most controversial answered to me.
"Spec or not to Spec?"

I am kidding!
But my current counter is 5 for Spec and 5 against Spec.
I think some rason could be that Spec is misunderstood, or still under development/inmature.
I can't tell, but  a beginner wants to know:
"WHAT AM I SUPPOSED TO DO!?"

To answer your BIG QUESTION ;-) Read the book chapter that I linked to above, or at least the first few sections, and try to build the login UI with Spec. Then you will have better information that can serve to make a choice for or against your use of Spec.

---> Save our in-boxes! http://emailcharter.org <---

Johan Fabry   -   http://pleiad.cl/~jfabry
PLEIAD lab  -  Computer Science Department (DCC)  -  University of Chile