RE: What about preparing scarab for dojo support ? your opinions ...

Hussayn Dabbous wrote:

>I am currently working on a project, that involves the usage of dojo-1.2
>Since i am about to start with some gui-related enhancements for Scarab 
>soon,
>i wonder, if  adding Dojo on the client side would be appreciated, or 
>severely disliked.
>
> From what i learned so far, we could reduce the complexity of the velocity
>scripts significantly, OTOH we would add a bit of dojo-coding, which in fact
>will be quite a bit of javascript...
>

I'm all for reducing the complexity of scarabs source code.
But I think the templates are complex because they are containing too much business logic, which should be pushed one tier down, not because they are velocity templates.
Converting velocity code into java script code wouldn't help much with the complexity-issue.

I'm not against introducing dojo (or any other javascript-framework).
In fact I see much potential, to improve the usability of Scarab in this way (e.g. an autocompletion-textfield for assigning users, instead of the clunky list-dialog which we have now).
But I'm also with Johannes here, that we should be careful when adding it. 
It must not become a hack on top of all the other hacks.
We must avoid adding even more complexity to Scarab.

If we decide to add dojo, we should branch before that. That means no dojo in the upcoming release.

Ronny