[ka-Map-users] yet another ka-map interface
lorenzo at ominiverdi.com
Tue Aug 8 10:28:38 EDT 2006
In my recent experience I've seen that many persons were not used to
a custom layout interface as the one of version 0.2
The winManager is still futuristic, so next link shows my candidate
to be the user interface for ka-Map version 1.0
First of all I have to say that I have not yet the chance to test it
with Internet Explorer. I'm in a remote part of Dominican Republic
and have rare internet connections.
I've stressed this code with Safari and Firefox.
Many changes have been introduced so I try to explain them as they
arise in my mind:
As it's simple to notice, I've tried to stick to a much more
"windows" friendly layout. Who's used to ESRI products can naturally
see a very close similarity to ArcReader/ArcExplorer (quite
explicit). Until I've maintained the Nomad icon set, I've adapted it
at a windows button style. I had to invent some more buttons, hope
that Aleks Herzog can once again come and help us.
To be honest I have to say that I've never had a good feeling with
Windows layouts in general but I've understood how is easier for a
newbie to understand ka-Map features if the user interface in much
Stats on my web server show that 80% of the people who's looking at
our ka-Map implementations have Windows.
More than newbies, I've seen that old GIS men are so Windows/ESRI
focused that have big problems to positive evaluate different
Hope this layout will help to spread ka-Map to a wider audience.
in the toolbar from left to right:
1) map selector -> no changes
2) Map Info -> the default selector. it shows the keymap and the
mouse geo coords. I've divided it form the legend to make it cleaner.
3) Legend -> always the same legend but I'm intended to change it a
bit in the next days to avoid HTML tables and to allow a more
4) Print -> as easy as possible. Choose an output and the document
should be downloaded automatically. PDF layout is always the same
(sorry David, hope to have time to merge your code soon).
5) Zoom functions -> always the same. I've reintroduced the scale
select to stick to the interface.
6) Identify/query -> some improvements. First of all no more pop-ups
(I hate them). I've introduced raster identification but it should be
refined. The system now looks in LAYER METADATA, as Andrea's query
system, looking for this fields:
the syntax is used to determine which fields to show and which alias
to use in the printed table. As usual, LAYER needs a TEMPLATE
argument to be queryable for mapserver.
Both tollerance and tolleranceunits are useful to refine the system. es:
I'm intended to add a button to "hilite" vectorial objects using
Andrea's system. This will take a while more maybe... still to decide.
7) Search system -> same layout that Identify. looking for same
fields in the metadata. Naturally it skips raster layers. That should
be refined to fulltext search and multiple field search but I've fear
that's not a mapserver skill (can somebody help with this?). I'm sure
I'll have to better study mapserver query and search systems.
8)Link to this view -> I've decided to add a stupid function to send
links by e-mail. To be refined.
I've took away all panels except the scalebar that I've refined with
a different CSS to make it smaller (I've fear It's not working fine
THINGS I STILL DON'T LIKE
1) Identify -> I've not yet decided the politic to limit output or
allow a full output control. At the moment it's limited to 20 record
per layer. It would be nice to use a dynamic page selector to show
all data. No clear idea about it for the moment.
2) Search system -> too simple.
1) Tooltips -> I've not yet integrated the tooltip system. I'll have
to refine it.
2) Measure -> it should be implemented. no technology is convincing
me... it's a stupid feature that many want.
3) Navigation History -> it should be easily implemented. I have to
decide it's priority.
4) cache clean up for hilite and print system.
5) Search system - > missing all advanced features.
6) Identify system -> missing all advanced features.
I've fear there's a little bug in extent calculation. It seems to me
(still to confirm) that the viewport extent is calculated with the
browser margin as left margin instead that the viewport real margin.
to a more intimate view, all panels have been reported as HTML code
in the index_kaExplorer.html (service div) to easily allow some text
So after all this comments, I'm still convinced we are close to a
I will really appreciate all kind of comments and suggestions. They
will influence my priorities.
thanks to all
More information about the ka-Map-users