Celestia UI redesign

[Vincent]

I agree that the marker stuff already uses a lot of space. However, putting the marker size on the same line as the marker color and symbol boxes is a very logical arrangement: all the marker style options on a single row of the UI. Making room for a size selector is one reason why I ditched that 'Marker Color' push button and integrated it into the color swatch.

The following layout for the markers group saves some room:

[ElChristou]

I agree that the marker stuff already uses a lot of space...

If you could implement a UI designer file as soon as possible it would be really great to do some serious search on the topic...

[chris]

I just checked in a prototype solar system browser for Qt4. It's separate from the star browser, but it could perhaps be integrated. Some features:

- Possible to switch between two views: the normal view and a grouped view
- Grouped view adds categories for asteroids, spacecraft, and minor moons to help keep the browser organized and uncluttered.
- Works with multiple star systems--all stars, barycenters, and planets in a system appear in the browser. This is immediately apparent when inspecting our own solar system: the solar system barycenter is the root object.
- Context menu works just like in the deep sky and star browser

There's currently no support for markers in the solar system browser. It could easily be added, but it's probably worth making the markers panel a separate widget first in order to avoid cutting and pasting code.

--Chris

[chris]

And here's a screenshot of the solar system browser:



--Chris

[cartrite]

I like the way you can undock it and move it around.





I think maybe a minimize button so you could store it on the desktop tool bar or a tab at the bottom of the screen.

When Sol is collapsed, it seems kind of bare. I see what was meant by integrating the solar system browser into star browser.
cartrite

[BobHegwood]

And here's a screenshot of the solar system browser:

Four what it's worth, Chris, I like this approach very much.
The MinorMoons box will open a similar panel capable of displaying
100+ other moons?

Thanks, Bob

[chris]

I like the way you can undock it and move it around.



I think maybe a minimize button so you could store it on the desktop tool bar or a tab at the bottom of the screen.

Yes that would be nice. Right now, you can close the browser and then restore it using the view menu, but there's no other indication that the browser is still around. On the other hand, we want to make sure that it's possible to banish all clutter from the screen and see just the 3D view.

When Sol is collapsed, it seems kind of bare. I see what was meant by integrating the solar system browser into star browser.
cartrite

Yes . . . The solar system browser should automatically expand the top level star items.

--Chris

[chris]

And here's a screenshot of the solar system browser:

Four what it's worth, Chris, I like this approach very much.
The MinorMoons box will open a similar panel capable of displaying
100+ other moons?

That's exactly how it works. There's no minormoon class right now however, so the browser just fills the "Minor moons" group with every moon that has less than 1/1000th the radius of the planet it orbits.

--Chris

[chris]

A tree-based solar system browser would indeed be useful. But should it be integrated with the star browser, as Christophe has done in the KDE version? There's a logic to doing it that way. On the other hand, the spaceflight professionals using Celestia would I think prefer a dedicated solar system browser.

If the Sun is in first place in the star browser, then there's no need of a dedicated browser.

I add few ideas for the GUI :

* Changing time scale should be easier to find. Vincent's slide bar is a good solution.

Agreed. I've implemented a basic time control toolbar for the Qt4 front-end, though I'm sure there are ways it can be improved.

* Another point from Edu Tools : a system to visualize scene orientation quickly (like a simple 3-axis display on a corner).

OK. What should the axes represent? The axes of the ecliptical coordinate system that Celestia uses? Or does it not matter so much as long as there's some sense of how the orientation is changing?

* I'd like the bookmark facility to be extended to save views : if I set 3 views and cameras to look to a spacecraft mission, I'd like to save the full set of settings to reload it or to share it with others.

Yes, that would be useful. I thought that there had already been some work on adding multiview capability for bookmarks, but I'll have to check again.

--Chris

[cartrite]

Chris,
I noticed that you included a new time toolbar. I guess it's not activated yet?

I kinda like the calender and date time edit widgets. It's really simple to create the ui with the qtdesigner and create the header file with uic. I think Celestia used to use a calender in a previous version. Currently, it's beyond my programming skill to tie the header to Celestia though. So I can't really see how it would really look in Celestia but..........


There does seem to be a problem with the minimum date being 1752 in the dime date edit widget. The calender only goes back to 2000. Actually, the calender can be set back to 1752 but.........Oh well.
I don't know if this could be worked around.

cartrite

[MKruer]

Chris, I don't know if you missed this, but I just wanted to know what you thought. Part of it was in jest, but its also a real question. Right now you were thinking of making the distinction between major and minor based upon the minimum radius of the moon based upon the parent planet size.

Chris,

One thing almost immediately comes to mind for me. That involves
the presentation of moons for selected planets. Jupiter, for example,
with ALL of its moons, presents a very unwieldy interface when one
is trying to select a particular moon from the interface. Now, granted,
not everyone is going to have all of Jupiter's moons installed, but I
am running into this problem with both Jupiter and Saturn.

I agree completely--all the minor moons of Jupiter and Saturn clutter both the 3D view (when orbits are enabled) and the GUI, and it's only on rare occasions that you actually want to see them. My current plan is to add a new "minormoon" category and organize the UI so that they appear in a separate category from the more significant moons. The distinction between what is a moon and what's a minor moon is somewhat arbitrary, which is why it will be settable in an ssc file rather than be automatically determined by Celestia.

Chris, you might want to use the term "Dwarf Moon" considering the IAU obsession with dwarfs. Then just use the same general criteria that would be used for the planets and apply them to the moons.

[chris]

Chris, I don't know if you missed this, but I just wanted to know what you thought. Part of it was in jest, but its also a real question. Right now you were thinking of making the distinction between major and minor based upon the minimum radius of the moon based upon the parent planet size.

Chris,

One thing almost immediately comes to mind for me. That involves
the presentation of moons for selected planets. Jupiter, for example,
with ALL of its moons, presents a very unwieldy interface when one
is trying to select a particular moon from the interface. Now, granted,
not everyone is going to have all of Jupiter's moons installed, but I
am running into this problem with both Jupiter and Saturn.

I agree completely--all the minor moons of Jupiter and Saturn clutter both the 3D view (when orbits are enabled) and the GUI, and it's only on rare occasions that you actually want to see them. My current plan is to add a new "minormoon" category and organize the UI so that they appear in a separate category from the more significant moons. The distinction between what is a moon and what's a minor moon is somewhat arbitrary, which is why it will be settable in an ssc file rather than be automatically determined by Celestia.

Chris, you might want to use the term "Dwarf Moon" considering the IAU obsession with dwarfs. Then just use the same general criteria that would be used for the planets and apply them to the moons.

I still prefer the term minor moon to dwarf moon.

I think applying the same general criteria to determine what is a minor moon is a reasonable idea. A lot of the small outer planet moons are in groups with similar orbits, indicating that they originated from the same fragmented parent body. But this isn't a terribly serious issue either--regardless of what we call them, the moons are still there in Celestia if you want to see them. And I hope no one gets their knickers in a twist over whether, say, Himalia is a minor or major moon.

--Chris

[chris]

Chris,
I noticed that you included a new time toolbar. I guess it's not activated yet?

I checked in the modifications to enable it in the main window. You can move it around to position it wherever you like, or disable it completely. You can reenable it via the View menu.

I kinda like the calender and date time edit widgets. It's really simple to create the ui with the qtdesigner and create the header file with uic. I think Celestia used to use a calender in a previous version. Currently, it's beyond my programming skill to tie the header to Celestia though. So I can't really see how it would really look in Celestia but..........

There does seem to be a problem with the minimum date being 1752 in the dime date edit widget. The calender only goes back to 2000. Actually, the calender can be set back to 1752 but.........Oh well.
I don't know if this could be worked around.

I was worried about this limitation I had to make custom date/time widgets for Windows just to get around this.

--Chris

[chris]

Regarding the time toolbar . . . I organized it like Vincent's time controller from the EDU tools. There are handy tool tips that popup to tell you what each of the buttons mean. The buttons themselves are rather ugly text based ones, but it's very simple to add icons. You can see the code in qttimetoolbar.cpp that is currently disabled; you can enable it, then copy icons from the Tango icon set into (celestia directory)/icons/qt if you want something a little prettier. We'll have to develop a nice set of custom icons for the new Celestia UI.

--Chris

[Vincent]

Regarding the time toolbar . . . I organized it like Vincent's time controller from the EDU tools.

I just tested it, and I have to say that I like it !
I agree that it could look prettier with these nice icons that Christophe (ElChristou) knows how to make...

Chris, what about rearranging the marker group as I suggested above ?

[cartrite]

Chris,
Now I see what it is doing. I was wondering why I wasn't seeing any reference to dates, hours, minutes, etc in the code.

Looks like there is a lot of room for other tools too.



cartrite

[Vincent]

We'll have to develop a nice set of custom icons for the new Celestia UI.

Looks better indeed, even with quite rough icons:

[cartrite]

I was worried about this limitation I had to make custom date/time widgets for Windows just to get around this.

I was looking at qtassistant QDate Class reference and found this line.

QDate::QDate ( int y, int m, int d )
Constructs a date with year y, month m and day d.
If the specified date is invalid, the date is not set and isValid() returns false. Any date before 2 January 4713 B.C. is considered invalid.
Warning: Years 0 to 99 are interpreted as is. If you want to specify the year 2010, you must write 2010, not 10.

It seems that only the widgets are limited to Sept 1752. But you wanted dates to back further?
cartrite

[chris]

We'll have to develop a nice set of custom icons for the new Celestia UI.

Looks better indeed, even with quite rough icons:

Very nice . . . I think that you should check those in, though first we should discuss on the dev list where the proper location for icons should be.

--Chris

[MKruer]

I was thinking (yeah I know that can be dangerous) Looking at the buttons, reminded me of a VCR/DVD player whatever, Anyway I though, would it be a good idea to add a record option for point as you browse a system and save it to a cel script. It might be helpful in the creation of the cel scripts if you can do the basics gotos inside Celestia, but I am not sure if this is outside the scope of the next release.