[kepler-dev] DataSource Icon troubles

Dan Higgins higgins at nceas.ucsb.edu
Fri Feb 3 14:49:28 PST 2006

    I understand your frustration with all the debate while you are 
trying to implement some of these interface changes, but I would like to 
point out that these changes appearing do affect what other people are 
doing. I use the 'animate' feature and I would like to be able to turn 
it off. I would also like to able to see icons when working with them. 
And I don't like a white background; I think some of the existing 
workflows look better with the gray background.
    I would like to suggest that none of the incremental changes you 
refer to should appear at all in the default version of the head of CVS 
until they are completed and ready for review.


Matthew Brooke wrote:

>It seems, from the discussions we have been having, that there are some 
>general misunderstandings regarding the Kepler UI implementation 
>process. Here are a few points and explanations that I hope will help:
>1) Most of the changes are incremental - meaning they are one or two 
>steps towards the new design, but in most cases, they aren't yet all the 
>way there. Consequently, it is too early to object to changes at this 
>stage. For example, in the past week, we have spent time addressing:
>a) Dan's email about the "Run With Feedback" menu not working (which I 
>knew about - it had not yet been finished, and was not critical for 
>development purposes), which sparked off a debate about what the menu 
>item should be called.
>b) The current datasource/file folder debate. Basically, the file folder 
>had recently been changed to the same bright green as the Director 
>icons, so I decided to take 10 minutes or so to make it match the new 
>Kepler design instead (see Wiki). However, the way the code currently 
>works, I could not get the actor to use the new SVG icons easily, so I 
>worked with what the code would let me do. The new Kepler will 
>ultimately *not* use this implementation (whenever this actor finally 
>gets changed to use the proper image files). It was a step towards the 
>new design, working within the restrictions of the current code.
>Unfortunately, the new icon design didn't contrast well with the old 
>Kepler gray background; consequently, I have gone ahead and changed the 
>canvas background to white, as shown on the design wiki. If anyone 
>objects, you can change it back again for your local copy by editing the 
>"uiSettings.properties" text file.
>2) Feedback and discussion are a very important part of the design 
>process, but we should also recognize that discussions and design 
>changes need to happen at the appropriate time in the engineering 
>process.  The design has been available on the wiki for many months; 
>discussing that information *after* the code is implemented (eg menu 
>item wording etc) is not an efficient way of moving forward, especially 
>when time is of the essence. Other issues may arise, which were not 
>evident from the design documents, but most of the issues we've been 
>discussing could have been addressed before now.
>3) The visual design is and will continue to be flexible. However, this
>does not mean that it should be changed in real time; I can't implement
>a design that is constantly changing.
>What it does mean is that the look of the UI is not "set in stone"; what 
>you see now, and changes that you will see over the coming weeks, are 
>all negotiable and easily changed. However, it would be prudent to save 
>these discussions for a point *after* the end-of-February freeze date - 
>otherwise, if we continue to analyze every UI change in great detail, 
>the implementation is never going to get finished in time for the release.
>Please bear with us on this - just let me get the "first draft" of the 
>implementation finished, and *then* we can talk about potential changes 
>to the design :-)

Dan Higgins                                  higgins at nceas.ucsb.edu
http://www.nceas.ucsb.edu/    Ph: 805-893-5127
National Center for Ecological Analysis and Synthesis (NCEAS) Marine Science Building - Room 3405
Santa Barbara, CA 93195

More information about the Kepler-dev mailing list