Hi,

Thanks for the detailed feedback. See my comments inline.

> On Jan 10, 2016, at 12:12 AM, Ferlicot D. Cyril <cyril.ferli...@gmail.com> 
> wrote:
> 
> Le 09/01/2016 10:32, Tudor Girba a écrit :
>> Hi,
>> 
>> Could you be specific? What exactly do you find less intuitive? To be able 
>> to react, it is useful to get concrete points.
>> 
>> Cheers,
>> Doru
>> 
>> 
>> --
>> www.tudorgirba.com
>> www.feenk.com
>> 
>> "Value is always contextual."
>> 
> 
> This was my impression but I will try to list these things.
> 
> First I find the improvements of Esteban good.
> 
> - For the icons it is true that the text help a lot. As someone mention,
> just a setting to get or not the text is possible, but a least be able
> to get the text is important IMO.
> I use Intellij for java debugging and sometime I get a little lost when
> I see the icons because for a same icon there is two different actions
> between Pharo and Intellij. So I lose always some time waiting the
> popup. And I think that a lot of students that do Java will have the
> same problem.

Interesting point, but I would not guide myself 


> - I like that the stack is fully loaded in the Spec debugger. In general
> that doesn't add this much time to show all the stack and it's one more
> line to show useful information.

It is now fully loaded in a FastTable in the GTDebugger as well.


> - On thing that I really don't like is to get 4 lines with the code of a
> block in the stack. This is too much space I think. And because of this
> we have a large horizontal scroll bar that we could avoid with a limite
> in character. I miss the fact that we had the name of the class on the
> line of a block in the Spec debugger.

This is no longer the case in the latest implementation. Could you check?


> For example I will join a screen with a block.
> 
> I would prefer something like:
> 
> In GLMMorphicTabulatorRenderer: [ :index :each | each hasId ifTrue:… ]
> And if you use FTTable you will not even get the horizontal scrollbar,
> so it's useless to get 4 long lines of the block if we cannot see half
> of it.


Indeed. Printing is a problem of the Context printing, and will be changed, but 
in the meantime we have FastTable. Is this good enough for now?


> - There is no way to close the inspector if we miss click

Indeed, this is a problem. We did not want to have an extra tab line and that 
is why we do not have a way to close inspector panes, but you can still scroll 
back. Also, in the meantime, we made the Evaluator the default second pane, so 
you can use it by default just like in the classic debugger.


> - I liked the separation between attributes and parameter/temp. I think
> that it was more readable. Before I could had a global overview of both
> easily. Now I have to scroll down to see the attributes. But for this
> one maybe I prefer the old one by habits.

I think this is indeed a habit issue.


> - The context menu of the spec debugger have more useful options
> (Suggestions, Code search…)

These can be added. Could you say more concretely which of those you actually 
used?


> These are the things that come in my mind. But maybe I missed some since
> the fact that I found the old debugger more intuitive was a feeling.
> It's hard to define why we have a feeling.

Thanks for taking the time. I know it’s hard to dig underneath an impression 
like that, but it is very valuable once you do it because it can push us 
further.

Cheers,
Doru


> -- 
> Cyril Ferlicot
> 
> http://www.synectique.eu
> 
> 165 Avenue Bretagne
> Lille 59000 France
> <screen14.png>

--
www.tudorgirba.com
www.feenk.com

"Quality cannot be an afterthought."


Reply via email to